Meta Advertiser Field Notes
Weekly observations from inside Meta ads

This week’s updates cover a follow-up on Push Delivery to This Ad, several ChatGPT ads changes, Meta’s plans to use website activity for Feed and AI recommendations, Reels post-view ads, and the growing perception that LinkedIn is overrun with AI-generated content.

1. Can you push delivery to more than one ad?
2. ChatGPT ads updates
3. Website activity to inform feed content, Meta AI recommendations
4. Reels post-view ads
5. What percentage of LinkedIn content is AI-generated?

Let’s get to it…

1. Can You Push Delivery to More Than One Ad?

Earlier this week, I wrote about how I finally have the option to push delivery to an ad. This allows you to force Meta to dedicate a percentage of your budget to an ad, which could solve issues related to neglected ads or time-sensitive promotions.

At the time of writing the post, I was unclear about the implications of this message that comes up when trying to push delivery to more than one ad in a campaign or ad set at a time…

This certainly sounds like Meta doesn’t want you to push delivery to more than one ad at once. It suggests that if you do, it could make it difficult to reach minimum spend for each ad.

At the time, I was able to verify that this was at least a soft warning. There was nothing preventing me from pushing delivery to more than one ad. But the question I couldn’t answer then was whether this could create a delivery problem.

This isn’t something I want to test at high budgets or for any significant length of time since I don’t want to force delivery unnaturally without a reason. But what I can say is that I tested this with six ads, where I pushed 10% of the budget to five of the ads (10% each, totaling 50% of the budget).

Using the first $75.02 of the budget spent since making this change and ignoring partial days, Meta is pushing budget accurately.

While I still can’t say confidently that this will always be the case, regardless of your budget, number of ads, or percentages you’re dealing with, this is at least encouraging that we shouldn’t be limited to pushing budget to a single ad — assuming it’s necessary.

One question I still have is whether the percentage requested is used as a minimum that will be exceeded in some cases. As you can see in the screenshot, the amount dedicated to each push is about the same. If one ad were recognized as a top performer, would Meta spend more?

My assumption is that Meta can spend beyond that requested percentage. In the example above, I only pushed delivery to ads that weren’t otherwise getting consistent spend. I can’t say for sure that this would happen, but it’s not unusual that these ads aren’t getting more spend than they are.

2. ChatGPT Ads Updates

I received an email from OpenAI last week that announced several features, some new to me and some not.

Edit campaign budget type: While I’ve always had the daily budget option, apparently not everyone did. All advertisers can now switch existing campaigns from lifetime budgets to daily budgets.

Clone to convert CPM campaigns to CPC: You can now convert a campaign to CPC bidding.

Custom CPM max bids: Once again, this update isn’t new to me. But apparently other advertisers had a fixed max bid before. Now, all advertisers can set a custom max bid.

Bulk edits in Ads Manager UI: You can now make bulk edits to campaigns, ad groups, and ads directly in the Ads Manager interface or using CSV uploads.

Flexible budgets in Ads Manager: Starting May 5th, OpenAI will allow some flexibility on daily budget pacing, similar to Meta ads. Daily budgets will be enforced as average daily budgets with a weekly allocation.

Expanding campaign targeting to additional countries: This one is brand new to me. Campaigns can currently target the United States, Canada, Australia, and New Zealand. In the coming weeks, targeting will also become available in the United Kingdom, Japan, South Korea, Brazil, and Mexico.

OpenAI also announced early testing of multi-advertiser ads.

Up until now, there have only been single-advertiser ad units. This announcement says that multi-advertiser “experiences build on our existing ad unit by presenting multiple relevant ads together within a single placement.”

Meta advertisers are well aware of multi-advertiser ads. You won’t find many big fans of the unit.

I haven’t seen an example of multi-advertiser ads in ChatGPT yet, and it will be interesting to see how OpenAI gets creative with ad placements. It feels as though the potential for inventory is limited at this point.

Some other updates I noticed were related to product ads. Within the Tools section of the OpenAI Ads Manager, I noticed a new menu item for Feeds.

It wasn’t at all obvious at first what this was for. At this stage of the ChatGPT ads beta experience, there are few tooltips or links to documentation that help clarify setup or how to use certain features. If you click to create a feed, you can provide a merchant name and supported currencies.

Once you click “Create,” you can provide SFTP connection details to bring in your product feed.

That’s relevant since there is now a Campaign Type option when creating a new campaign. “Standard” is the default, but you can also select “Product Feed.”

3. Website Activity to Inform Feed Content, Meta AI Recommendations

Meta made an interesting announcement about a change to how website and event data will be used.

We’re updating how we use information that other businesses already share with Meta. We already use this data — like games you play or purchases you make on other websites — to make the ads you see more relevant. In the future, we’ll use this information to personalize other parts of your experience, including the content you see in your Feed and AI responses.

Event data from advertiser websites, apps, and Conversions API connections is already used to inform the ads people see. But it hasn’t been used to help determine the organic content people see in their feeds.

I’ll admit I didn’t realize this. I’ve long assumed that Meta used advertiser data to help personalize the algorithmic feed — not just for surfacing ads, but organic content, too. That apparently hasn’t been the case. And if this is indeed changing, the potential is there to make the algorithmic feed more relevant.

But maybe more important is the news that this data will also be used to inform AI responses. If you’re having a chat with Meta AI, it will already know the websites you visit and apps you use, assuming that data was sent to Meta. And that could significantly impact the background the AI has for making recommendations.

And that could be both good and bad. You shouldn’t need to explain as much to the AI, which should lead to fewer frustrating conversations and more relevant responses. But that could also be incredibly creepy, depending on how that “knowledge” is surfaced.

If Meta can leverage this in a meaningful way, of course, it could be a significant advantage over Claude, ChatGPT, and other LLMs. It’s data that they simply do not have.

4. Reels Post-View Ads

Social Media Today reported on a press release indicating that Meta will make post-view ads available to more advertisers.

Here’s how they work…

These ads can be displayed after 60-second Reels, which is intended to capture highly engaged users. They include a 5-second countdown on the organic content, warning that an ad is coming. Once the ad starts, users can choose to manually skip, returning them to the Reel they were viewing.

I am not currently seeing this placement, but it’s something to watch.

5. What Percentage of LinkedIn Content Is AI-Generated?

I’ve been increasingly annoyed by the concentration of AI-generated content across social media. But it can’t possibly be worse than what we’re seeing on LinkedIn right now.

I decided to create a poll to see what others felt about the percentage of LinkedIn posts and comments that are currently AI-generated.

While unscientific in every way, it’s interesting how popular the belief is that such a high percentage of content on LinkedIn isn’t human-created. Could it really be more than 50%? My estimate is consistent with that vote. Though it’s certainly possible that my bubble of tech and marketing folks would reflect a greater level of AI experimentation than most.

The irony also isn’t lost on me that if a substantial percentage of LinkedIn posts and comments are AI-generated, that may also apply to participation in my poll. But perception is important here.

Whether the percentage is 15%, 35%, or 65%, it’s impacting the way we consume and engage. I assume that most of the comments on my posts are AI-generated now. I only react or respond to comments if I believe they actually represent the feelings of a human. I also won’t engage with posts that were clearly written by AI.

I’m certainly not alone in this change of behavior. What percentage of users are conscious of this and intentionally engage less as a result? At what point does it become a problem for the platform?

This isn’t just an issue for LinkedIn, of course. AI fatigue is everywhere. While it may still fall in that shiny object window for now, will people get fed up and abandon it all? Will it eventually just be a place for AI to engage with itself?

I’m curious what impact that change in behavior will have on these platforms and how the true human connection void will be filled.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Push Delivery Tests, ChatGPT Ads Updates, and More appeared first on Jon Loomer Digital.

More than two months ago, I briefly covered a new test feature called Push Delivery to This Ad. I didn’t have it at the time, but I had enough information to at least confirm its existence.

The concept is simple. When running multiple ads in one ad set, Meta’s delivery algorithm tends to favor one ad or a small handful of ads, often at the expense of others. You can “push delivery to this ad” to guarantee that ad gets shown for a period of time.

I finally have it, so I may be part of a broader rollout. Since I’m still not seeing official documentation, I assume it remains a test feature for now.

But now that I have it, I’ve pushed some boundaries to better understand the functionality. In this post, I’ll help you understand how it works, the limitations, and how you might use it.

Requirements

Since there isn’t any official documentation on this feature yet, I’ve had to do my best to figure this out based on trial and error.

Campaign Objective and Performance Goal

I do not believe the campaign objective or performance goal matters. I’ve tried each campaign objective, and the feature appears. I can’t confirm whether it actually works when published with each objective and performance goal.

Ad Format

I do not believe the ad format matters. The feature appears for manual upload or Advantage+ Catalog Ads. It also appears for single image or video, flexible format, and carousel.

New and Existing Ads

The feature also appears for both new and existing ads, whether they are currently running or not.

Number of Ads

The only requirement I’ve found that impacts whether or not the feature appears is the number of ads in an ad set. If there is only one ad, whether active or in draft, the option to “push delivery” will not appear. That changes once there are at least two ads.

How It Works

Once you have at least two ads in an ad set, either active or in draft, go to the Ad Setup section of ad creation. Under Multi-Advertiser Ads, click the link to “show more settings.”

Under the ad scheduling option, you’ll see a setting to “push delivery to this ad.” It reads:

Help ensure delivery by dedicating a portion of your budget to your ad.

The tooltip reads:

Select a portion of your budget to push delivery to this ad for up to 7 days or until your ad set ends. After that, the ad will use your existing budget.

We’ll come back to that seven days maximum in a minute, but the details here are important. You can force Meta to spend a designated percentage of your budget during the “push.” After that, delivery will return to normal.

The setting is off by default. When turned on, it looks like this…

You will need to manually enter the percentage of your campaign or ad set budget that should be dedicated to this ad. Then select how long that push will last.

Limitations

There are a few limitations to be aware of.

1. Minimum and maximum length of the push.

First, recall that the tooltip suggests that you can push delivery for up to seven days. That actually appears to be a recommendation, rather than a hard limit. There is nothing preventing you from choosing to run the push for eight, 20, or even 30 days.

The actual maximum appears to be 30 days, with the minimum being one day.

2. Number of ads pushed at once.

This is more of a recommendation, but Meta suggests that there may be an impact to delivery if you attempt to push to more than one ad at a time. If you’ve turned on “push delivery” for one ad and attempt to turn it on for a second, you’ll get this message:

Another ad is using this setting
If you push delivery to multiple ads in the same campaign or ad set, minimum spend per ad may not be met.

It’s not clear what impact pushing delivery to multiple ads would have. As an experiment, I’ve chosen to push 10% of my budget to five different ads within an ad set. So far, that doesn’t seem to be a problem.

Obviously, there will be limits to this. You can’t have 10 ads in an ad set and tell Meta to dedicate 15% of your budget to each one. So it’s understandable why this warning exists.

But it’s unclear how much this matters if you’re pushing delivery to a second ad while keeping percentages low. I’ll know more from my experiment with time. So far, it’s running fine and dividing my budget as expected.

3. No creative tests.

The creative testing tool will be unavailable while you’re pushing delivery to an ad. The option to “set up test” is grayed out.

When you hover over the disabled button, it reads:

You’re pushing delivery to one or more ads in this campaign. You can only set up a creative test after these push delivery periods have ended.

Presumably, this limitation goes both ways. I’m not running a creative test right now, but it’s safe to assume that you cannot push delivery to an ad when running a creative test in the same campaign or ad set (wherever the budget is defined).

Creative Testing Tool Comparisons

If the functionality of the “push delivery” tool looks familiar, it’s because it clearly steals from the Creative Testing Tool. Here’s how the setup looks in that case…

A slight variation here is that you’re selecting a specific dollar amount to spend, rather than a percentage of your budget. But it still cuts into that overall campaign or ad set budget. The length of the test also needs to be between 1 and 30 days.

But there are some very important differences…

1. Number of ads involved.

The creative testing tool requires that you include at least two ads at a time. The maximum depends on the version you have of the creative testing tool, but mine is up to five.

When pushing delivery, it’s to a single ad. While you can individually push delivery to multiple ads, each decision to do so is separate. And Meta clearly prefers that we limit the number of ads we push delivery to at one time.

2. Active or draft ads.

One of the most annoying limitations of the creative testing tool is that you can only test new ads. By the time an ad is running and you decide you want to test it, it’s too late. You’d need to create a duplicate for the purpose of testing.

There is no such limitation to push delivery. You can push delivery to an active ad or choose to push delivery to a new ad as it’s published.

3. Test or independent.

And then the obvious difference: The creative testing tool lets you set aside budget to test multiple ads against one another. It’s an A/B test, and a winner will be determined based on your chosen metric.

No such test is happening when you push delivery. You’re just asking Meta to dedicate a portion of your budget to an ad.

When to Push Delivery

First, this feature impacts how I approach the creative testing tool. In the past, I saw it as a solution for understanding what would happen if Meta dedicated a percentage of your budget to each ad. But this could only be done with new ads, so I initially made it part of my process to test new ads when I published them.

But the introduction of pushing delivery completely changes things. I see no reason to start new ads with a creative test anymore. In most cases, I recommend letting the algorithm run without manual manipulation. If you ever want to push delivery later, you can.

There are two primary situations when I believe this could be useful…

1. Minimal delivery to an ad from an underperforming ad set.

I want to be very careful how I word this because I don’t want this to be misunderstood. You should not be pushing delivery to every ad when it’s not delivering. You should prioritize minimizing your involvement where possible.

But all bets are off when you have an underperforming ad set. When you’re not getting the results you want in aggregate, feel free to mess with things and experiment. While my general recommendation would be to create new ads, you could always push delivery to a low-impression ad for a few days and see what happens.

If it’s a single ad, you probably won’t even need to push delivery for more than a few days, depending on the budget going to it. It does appear you can turn push delivery off before it is scheduled to end.

2. Time-sensitive promotion.

Let’s say you have an ad set with multiple ads to promote multiple products at similar price points. But you have a promotion that’s going to run for seven days.

You create new ads for that promotion. But simply adding them to the existing ad set would be dangerous since Meta may not show the ads. In the past, the solution may have been to create a separate campaign or ad set for that promotion.

But now this gives you another option. You could publish ads for the time-sensitive promotion to the existing ad set and then push a designated percentage of your budget to those ads until the promotion ends.

Final Thoughts

This is another rare feature that gives advertisers back some level of control. We’ve generally been at the mercy of the ad delivery gods up until now, unless you wanted to force control in other ways.

And that forced control does concern me. Advertisers are far too tempted by control they don’t necessarily need. And when they find creative ways to take that control, they often do so at a cost.

That doesn’t mean that there aren’t problems to be solved. If there is a provable problem with ad delivery in this case, the push delivery feature could be a solution.

But making a habit of using it feels like a recipe for driving up costs. And when you’re doing that with intentionality, that’s just careless.

Your Turn

Have you started using this feature to push delivery yet? What do you think?

Let me know in the comments below!

The post Hands-On With Push Delivery to This Ad appeared first on Jon Loomer Digital.

Almost every advertiser screws this up…

One of the most common mistakes I see from advertisers is related to audience segments. They either fail to define them at all or do so incorrectly.

What Are Audience Segments?

Audience segments provide incredibly helpful access to additional reporting data. When running sales campaigns, you can use the breakdown by audience segments.

This helps uncover how your spend and results were distributed by audience. When defined correctly, you can see how results are segmented by existing customers, engaged audience, and new audience.

When you have this information, you’ll understand how much remarketing happens naturally with algorithmic targeting. Those who don’t use this feature are often those who feel the need to lean heavily into remarketing strategies.

How to Define Audience Segments

Audience segments are defined in Advertising Settings. Your Engaged Audience should be the broadest group of people connected to you, using custom audiences. I use custom audiences for my entire email list and all website visitors from the past 180 days.

Existing Customers should be defined with whatever custom audiences you have for those who have bought from you. I use a website custom audience for all purchase events from the past 730 days and an email custom audience reflecting all of my paying customers.

In May, Meta updated the longest retention window for purchase events to 730 days (2 years) from 180.

Note that audience segments are only used for reporting purposes. They will not influence delivery. So there’s no reason to limit these definitions.

The post Common Audience Segments Failures appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

This week’s updates cover a couple of custom audience exclusion changes, a new audience control for app advertisers, and what may be the end of page like ads.

1. Hands-on with Customer Lifecycle Strategy
2. “Legal and operational” custom audience exclusions
3. App Install Status audience control
4. No more page like ads?

Let’s get to it…

1. Hands-On With Customer Lifecycle Strategy

A couple of months ago, I wrote about how many advertisers on LinkedIn were sharing a new feature called Customer Lifecycle Strategy. It seemed like everyone was sharing a screenshot that looked a lot like this…

The screenshot was rather straightforward. Choose whether or not you want to limit your audience to new customers.

But how that would be accomplished was less clear. If you listened to the advertisers sharing the screenshot, you’d assume this was some “game-changer” feature that improved on manual exclusions of existing customers. The claims all suggested that this was some AI-powered magic that gave you access to special data that can better exclude those people.

But it was all talk. No one would share a screenshot or link to official documentation. My assumption was that, at best, this might use our audience segment definitions.

And now I finally have the feature in a couple of ad accounts to confirm. As far as I can tell, I was right to be skeptical.

If you select the option to acquire new customers, it looks like this…

See that? “Exclude these existing audiences: None selected.” It appears you need to manually select custom audiences to exclude.

If you hover over that, you can edit. It looks like this…

You have the option to manually select custom audiences to exclude that reflect your existing customers and engaged audience. But note that this is an entirely manual process.

It doesn’t pull from your predefined audience segments. This was a test account, so I defined those to confirm it. Nothing changed.

It doesn’t pull from labels used with custom audiences. I’ve covered these related to a beta test of Audiences for value rules. When I labeled a custom audience as a customer, it had no effect on this either.

Look, it’s always possible that some people have a different version of this than I do. But everything I’m seeing confirms precisely what I assumed: There is no magic here.

This is just Meta surfacing a second location to exclude your custom audiences. In either case, they’re manually selected.

Now, would it make sense if Meta automatically pulled these custom audiences from audience segment definitions? Absolutely. But even then, that’s just a time saver. It’s like a saved audience for custom audience exclusions.

2. “Legal and Operational” Custom Audience Exclusions

The next change is directly related.

I wondered whether the custom audiences selected to exclude in the Customer Lifecycle Strategy section would also be reflected in the Audience Controls section of the ad set to exclude custom audiences. Strangely, they’re not. But what I found was interesting.

I’m weirdly fascinated by small changes Meta makes to the language used in labels and descriptions. It interests me because these changes are never accidental. They’re purposeful.

In the example above, it reads “Legal and operational custom audience exclusions.” If you don’t have Customer Lifecycle Strategy, this section is labeled “Exclude these custom audiences.”

Here’s the tooltip for the new version…

Only exclude custom audiences if you need to for legal or other operational reasons. To exclude your existing customers, use the settings in Customer Lifecycle Strategy.

So, this suggests that you should use each section for a specific reason.

Customer Lifecycle Strategy: Exclude your existing customers or engaged audience.

Legal and Operational Custom Audience Exclusions: Exclude if you need to for legal or other operational reasons.

Now, there’s a lot that’s weird about this. There’s nothing keeping you from excluding existing customers in the “Legal and Operational” section, just as there’s nothing preventing you from excluding “legal and operational” custom audiences in the Customer Lifecycle Strategy section. It’s just two places to exclude custom audiences.

Second, and maybe I’m missing something, but why would you have custom audiences to exclude for legal reasons?

It’s important to remember that this is still an early version of… something. It doesn’t appear to be at all helpful at this point. But there’s likely a vision for something far more dynamic than what I’m seeing.

3. App Install Status Audience Control

While I was investigating custom audience exclusions, I noticed something I hadn’t seen before in Audience Controls: App Install Status.

Now, it’s entirely possible that this has been here for a while since it’s a feature that I wouldn’t necessarily care much about. I don’t promote my own app, after all. But I’ve spent enough time recently in Audience Controls that I’m reasonably certain that this is new(er).

Here’s what appears under the tooltip:

If you’ve implemented the Facebook SDK or a mobile measurement partner, this can help improve the performance of the app install status to target ads to people who have your app.

There are two options in the Install Status dropdown menu.

• Any: Reach people who either have your app or don’t.
• Not Installed: Reach people who don’t have your app.

Maybe I’m crazy, but I’m pretty sure that this used to be handled with App Activity Custom Audience exclusions. Now you simply select “Not Installed” and the mobile app that’s connected to your account.

At the very least, I’m finally documenting this somewhere.

4. No More Page Like Ads?

I’m seeing this message (“Page likes performance goal is no longer available”) in one ad account when using the Engagement objective and Instagram or Facebook conversion location…

The option to “maximize number of page likes” is grayed out.

Look, I’m not going to suggest that running ads to increase page likes is a particularly smart use of your money. But this still surprises me.

Assuming this reflects a global decision and not an isolated test, the removal of page like ads represents an end of an era. This was one of the first strategies advertisers used in the early days of Facebook advertising. Before conversion tracking, you could build your page likes.

And then, you know… Organic reach tanked, making those likes mostly worthless. But I digress.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Customer Lifecycle Strategy Is Not What Advertisers Thought appeared first on Jon Loomer Digital.

What you do with default conversion reporting is what separates good advertisers from bad ones.

What do your results actually mean? How many of those results come from click-through, engage-through, and view-through attribution settings? How many came from the first reported conversion, and how many came from multiple actions by the same customers?

What additional information can you uncover that isn’t visible in your default reporting? Did more conversions happen? Did leads ultimately buy weeks down the road? How many of your conversions does Meta view as Incremental?

These are important questions that can be answered once you’ve mastered two important tools:

• Breakdown by Attribution (Settings and Conversion Count)
• Compare Attribution Settings

Let’s take a closer look at the specific ways that you can use these tools to make sense of your default reporting and uncover results that were hidden.

What Do Your Conversion Results Mean?

One of the most common mistakes advertisers make is that they take their results at face value. This leads to either an overconfidence in what is reported or frustration with Meta when inconsistencies are discovered.

That meaning can be found with the Breakdown by Attribution Settings option.

Assuming your events are set up properly and you’ve deduplicated events that are sent from multiple sources, what appears in the Results column reflects reality. But the meaning of that reality is what you’re often missing.

If you stick with the attribution defaults in the ad set, as most advertisers do, you use the Standard attribution model.

This means that Meta will give credit to your ads for conversions based on rules tied to clicks on links, clicks on other elements, and impressions that happened within designated windows prior to the conversion.

The default attribution setting under the Standard attribution model is 7-day click-through, 1-day engage-through, and 1-day view-through.

This means that Meta will report on conversions if…

1. Click-Through: Someone clicked a link on your ad and performed your desired conversion event within seven days.

2. Engage-Through: Someone clicked a non-link (like a reaction, comment, or share) or watched your video ad for at least five seconds and converted within a day.

3. View-Through: Someone was served an impression of your ad without performing a click-through or engage-through action and converted within a day.

When you use the Breakdown by Attribution Settings, Meta will generate four rows:

• 1-day click
• 2-7 day click
• 1-day engagement
• 1-day view

It doesn’t matter whether you get any results for one of these settings. The rows will appear. So even if you define attribution in the ad set as 7-day click only, you will still get rows for the other settings.

Of course, you have to use the Standard attribution model to use this breakdown. If you use the Incremental attribution model, the extra rows won’t appear. The reason for this is that breakdowns segment your results to help you understand how budget or results are distributed. If you don’t use the Standard attribution model, the attribution settings do not apply to your results.

The concentration of these attribution settings in your conversion results matters. Click-through conversions are those that we can say most confidently tie to your ad. Someone clicked your ad and went to your website before converting, either within a day or up to seven days. Clearly, the sooner that happens, the more likely that it was your ad alone that drove the action.

But neither engage-through nor view-through conversions require a click on a link. While it’s possible the targeted person will go directly to your website or Google you later to complete the conversion, the more likely explanation for these results comes from other marketing channels. That doesn’t make these results worthless, but your ad needed help.

This is why it’s so important to use the Breakdown by Attribution Settings. Without it, you have no context for what your results mean.

How Many Repeat Conversions Did You Have?

Another attribution setting that often goes missed is for Conversion Count. By default, Meta will report on all conversions that happen after someone engages with your ad.

What this means is that one person can perform multiple, separate conversions that can each be counted in your results. For example, someone can click your ad and immediately make a purchase on your website. If the attribution setting includes 7-day click, they can come back and make another purchase three days later. It will also be counted.

While I wouldn’t see this as a problem when the conversion event is a purchase (all purchases are valid, and you should value customers who perform multiple purchases), such purchase behavior could create reporting discrepancies.

Meta may report 50 purchases in Ads Manager, but you only know of 40 unique customers. That may seem like Meta is overreporting results.

This is one useful way to use the Breakdown by Conversion Count.

When used, Meta will generate rows for First Conversion and All Other Conversions.

First Conversion will count only the first conversion someone completes. All Other includes the additional conversions that those people made within the attribution settings.

The example above is for a lead magnet. New registrants are given repeated options to register for another lead magnet after signing up. That’s why only 55% of the results are First Conversions.

This context obviously impacts how we view results. While it’s valuable that many of these people signed up for multiple lead magnets, the Cost Per Total Registrants is important information to know. You could even call the default results misleading without knowing that context.

What Attribution Is Hidden?

The value of Breakdowns is that they provide important context behind your conversion results. But they won’t uncover conversions based on attribution models or settings that weren’t chosen in the ad set.

That role is carried by the Compare Attribution Settings feature, which is found within the Columns dropdown menu.

Checking boxes will add columns for individual attribution settings, models, or conversion counts.

It doesn’t matter how you defined attribution in your ad set. Columns will reflect conversions that may or may not be contained in the default results.

Here are some examples of how you might use this…

1. When you limited the Standard attribution model.

Let’s say that you defined attribution in the ad set as 1-day click only.

If you were to use the Breakdown by Attribution Settings, you would only see results that came from 1-day click. But that doesn’t mean that conversions didn’t happen from view-through, engage-through, and 7-day click attribution. They just weren’t part of your default reporting.

Using Compare Attribution Settings, check the boxes for 7-day click, 1-day view, and 1-day engagement. You could also check the box for 1-day click if you want, though the default Results column will be for that setting.

Then separate columns will be added for those attribution settings, which could reveal other conversions.

In the example above, 315 conversions were reported by default, using the 1-day click attribution setting. But 20 more were found from view-through (13) and engage-through (7). And one more conversion happened between 2 and 7 days (subtract the 1-day click total from 7-day click).

2. Uncover 28-day click conversions.

The younger advertisers may not remember this, but Meta’s original attribution setting years ago was 28-day click and 1-day view. The iOS 14 changes put an end to that, but Meta did bring 28-day click attribution back for reporting purposes.

I find this extremely valuable for situations where the goal conversion event of an ad set wasn’t a purchase. For example, you may be running a leads campaign and want to get a sense of how many leads purchased within 28 days. Adding a column for 28-day click can help uncover that information.

3. View an alternate attribution model.

As explained earlier in this post, the Breakdown by Attribution Settings won’t work if you defined the attribution model as Incremental in the ad set.

The reason for this is that Incremental Attribution doesn’t utilize click-through, engage-through, and view-through attribution settings. Instead, Meta uses complex models to determine whether an ad drove a conversion.

But that doesn’t mean that Meta can’t report on those attribution settings. When using Compare Attribution Settings, select the windows that you want to view.

The same can be done if you chose the default Standard attribution model in the ad set. You can view results that came from the Incremental attribution model if it’s selected when using this feature.

Incremental results tend to be more conservative, if not more accurate, than those found using the default Standard attribution model. These steps can help provide that alternate view.

4. View an alternate Conversion Count.

Meta also allows you to define the Conversion Count in the ad set for optimization and default reporting purposes. If you don’t touch it, Meta will report all conversions that qualify. You could then make use of the Breakdown by Conversion Count, as discussed earlier, to view how many of those results were First conversions versus “all other” conversions.

But if you defined Conversion Count in the ad set to only count First conversions, that breakdown option won’t reveal any data for the second row. You could find it using Compare Attribution Settings.

Once you check any of the attribution settings boxes, options will come up to “Choose Conversion Count.”

Meta will then generate separate columns for each selected attribution setting and First conversion, All conversions, or both.

More Data, Better Data, and Better Decisions

One of the biggest mistakes you can make as an advertiser is to rely solely on Meta’s default reporting. It’s not that these results are wrong or that Meta is lying, despite what many will claim. It’s just that those results require important context.

You need to know what your results actually mean. Breakdown by Attribution can help spot problems related to a heavy concentration of results in view-through or engage-through attribution. When you know this, you can make adjustments and report on your results more accurately.

The Compare Attribution Settings feature allows you to highlight data that may not appear in your default results. This can help you tell a broader story about the impact of your ads.

Use these tools daily. You will make better decisions based on more accurate information.

Your Turn

How do you use these tools to highlight better data?

Let me know in the comments below!

The post Meta Ads Attribution Reporting: What Your Results Really Mean appeared first on Jon Loomer Digital.

I just had back-to-back calls with advertisers who were thrown by this, and I know they’re not alone.

Meta made some changes related to Leads campaigns recently that are creating lots of confusion.

The Problem

When you create a campaign with the Leads objective, the default conversion location is now Website and Instant Forms.

This is easy to miss, especially if you normally stick with the default.

But you’ll quickly realize something is different when you create your ad. You will be asked for both a website URL and an instant form to use.

The two advertisers I spoke with were both confused because they only wanted to use one. But when you click the dropdown to select the conversion location, it seems like you can’t. The options are Website and Instant Forms, Website and Calls, and Website and Messenger.

Hidden at the bottom is an option for “Single.” This allows you to send people to one location where you want them to convert. But you can’t immediately see the choices available.

Expand “Single,” and then you’ll be able to choose one location, which includes only Instant Forms or Website.

It’s Buried for a Reason

Meta is working really hard to hide this. They clearly want advertisers to let the algorithm choose between two conversion locations.

The post Website and Instant Forms Confusion appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

This week’s updates cover a mix of useful reporting additions, early movement from ChatGPT ads, and a couple of new Meta ad creation options that are worth watching.

1. A new Grouping reporting feature
2. Conversions campaigns coming to ChatGPT
3. Instagram Live Video conversion location
4. Do AI-generated videos deliver?

Let’s get to it…

1. A New Grouping Reporting Feature

Some advertisers are seeing a new Grouping reporting feature in Ads Manager next to the Breakdowns menu.

Grouping is only available when viewing campaigns or ad sets. It allows you to group your available campaigns or ad sets by variables that fall under four different categories:

• Ad Settings
• Attribution
• Budget and Bidding
• Business Goals

It’s similar to Breakdowns. But while Breakdowns will break down results for each campaign or ad set, Grouping will group similar campaigns and ad sets by variable.

Ad Settings allows groupings by the following variables:

• Campaign Name
• Delivery
• End Date
• Recently Changed
• Special Ad Category
• Start Date

Attribution allows groupings by the following:

• Attribution Setting
• Conversion Count
• Conversion Event
• Conversion Location

While a Breakdown by Attribution would generate separate rows to break down how each attribution setting contributes to an individual campaign or ad set, the Grouping by Attribution will group all ad sets that used the same attribution setting.

Budget and Bidding allows you to group assets by these variables:

• Ad Set Spending Limit
• Bid Strategy
• Budget Scheduling
• Budget Strategy
• Buying Type

So Meta will group all campaigns or ad sets by the bid strategy used, for example.

And finally, Business Goals allows these groupings:

• Evergreen Campaign
• Funnel Stage
• Objective
• Performance Goal

While Objective and Performance Goal are reasonably obvious, Evergreen Campaign and Funnel Stage are particularly interesting. The Evergreen Campaign option will group assets by “Evergreen” and “Not Evergreen,” but I’m not immediately clear how each is determined. Funnel Stage presumably groups by Upper Funnel, Middle Funnel, and Lower Funnel.

I only have this for one lightly used account, so I’m limited on applications. But it’s something to look for.

2. Conversions Campaigns Coming to ChatGPT

I’ve joined the OpenAI Ads Manager Beta for ChatGPT ads. I covered the basics recently, and the takeaway was just that: Basic.

While simplicity is somewhat refreshing, the biggest issue was an inability to optimize for conversions. According to an email I received today, that’s changing.

The most important paragraph:

In early June, we’ll begin offering conversion-optimized campaigns, with early access for customers who set up Pixel or Conversions API in advance. Any account that has conversions set up by Monday, June 1st, will be granted early access by June 5th.

At the moment, you can set up the OpenAI pixel, Conversions API, and conversion events, but the events are only used for tracking. Soon, you’ll be able to direct delivery to focus on these events.

I’ve set up my OpenAI pixel using Google Tag Manager, and it’s mostly set up the way I set up manual tags and triggers for the Meta pixel. The code and events, however, are quite different.

Like the Meta pixel, the OpenAI pixel should appear near the top of the <head> section and fire before event code. Examples of conversion events include…

• appointment_scheduled
• checkout_started
• contents_viewed
• custom
• items_added
• lead_created
• order_created
• page_viewed
• registration_completed
• subscription_created
• trial_started

If you add your pixel and start sending events by June 1st, you’ll get early access to conversion-optimized campaigns on June 5th.

You can learn more about the technical specifications of the OpenAI pixel, Conversions API, and conversion events here:

• JavaScript Pixel
• Conversions API
• Supported Events

3. Instagram Live Video Conversion Location

A couple of weeks ago, I shared how some advertisers were seeing a new Live Video Ad toggle in the ad set.

It’s not clear if Meta’s experimenting with different flows, but that option disappeared for me and was replaced with a new conversion location for Instagram Live Video.

This conversion location can be used with any available performance goal. So you can promote a live video and optimize for purchases, for example.

You then need to choose whether you’re promoting a current live video (if there is one) or an upcoming live video.

Next, you set a start and end date and time to schedule the ads to run at the same time as your live video.

Ad creative is super basic. It will automatically use your live video as the creative, and you only need to provide primary text.

4. Do AI-Generated Videos Deliver?

Meta has been putting resources behind an impressive AI-generated video tool. It allows you to generate video ads based on still images.

I’ve experimented with it quite a bit. While it’s not perfect, it’s getting better, and I’ve even used some of these videos in my ads.

Or at least, so I thought…

I seem to have stumbled on a very annoying bug. I’ve gone through the process of generating videos for multiple ads now. It takes multiple minutes. But when I’ve gone back to look at which videos I created, they no longer appear.

In fact, I still see a “Get Started” button, as if I never created any videos at all.

This isn’t user error. I’ve tried it multiple times now. Generate a video (or multiple videos). Publish the ad. Open the ad, and they briefly appear in the creative tiles at the top right. Then they disappear.

I have no idea if this is just a display error or if the AI-generated videos aren’t saving at all. There’s no way to see for sure.

I assume this could fall under the Image Generation creative breakdown that I had months ago, but that disappeared. So now, I’m just assuming that the videos don’t save.

Have you had this issue?

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Meta Ads Grouping Reports, ChatGPT Conversions, and More appeared first on Jon Loomer Digital.

Meta advertising best practices today look nothing like what they did a few years ago. It’s obvious that most agencies and advertisers have struggled to keep up.

In some cases, it’s because they’re trying to apply strategies that were relevant years ago but aren’t now. In others, they’re confused because of conflicting information shouted on the internet. More often than not, they’re doing things that make results worse.

I’ve seen so many troubled ad accounts and campaigns. While every account is unique, there are seven common mistakes that I see over and over again. They’re so common that when I know results are suboptimal, I look for these things first.

Here are the seven most common mistakes agencies and advertisers make. Their bad setup and old strategies are only hurting their clients.

1. Incorrect Access to Account Assets

While the mistakes listed below can all impact performance, this first problem may be the most frustrating to business owners and the future agencies that manage them.

When setting up an agency’s access to assets, it’s critical that it happens like this…

1. The client has a business portfolio (previously known as a business manager).
2. The agency has a separate business portfolio.
3. The client owns all related assets (ad account, datasets, Facebook pages, Instagram accounts).
4. The agency is added to the client’s business portfolio as a partner.
5. The agency is granted relevant access to each asset.

When set up this way, the client owns their assets and the agency gets access to what they need in order to manage the ads. When the relationship ends (it will), the agency is simply removed as a partner. The new agency takes their place.

One of the most common mistakes that agencies make here is that they’ll own some of the most critical assets, like a Facebook page. When the relationship ends, that agency goes silent. And then the client is unable to manage that page on their own or give access to another agency.

This is hell for the client and their future agency partner. Do not do this. It doesn’t matter if it’s easier to set everything up yourself. The client needs ownership.

While an argument can be made for the agency setting up a separate ad account for a client that the agency owns, it’s typically motivated by an irrational feeling that strategy and results are proprietary. Hiding the ad account from clients buries deeper issues in your process.

Can the client simply pick up where you left off and get the same results? Could another agency do the same? If you think this, you’re putting far too much value in campaign construction and strategy. Your value, if you offer any, is found in how you adjust to day-to-day problems as they happen.

2. Audience Segments Not Defined or Incomplete

This is the first thing I like to look at. What an agency or advertiser has already done will say a whole lot about their quality of work.

If I go to Advertising Settings and see this for Audience Segments, I know we’re in trouble…

This means that Audience Segments weren’t defined. And if Audience Segments weren’t defined, it means that the business has no idea how much remarketing happens by default when running a Sales campaign.

The most obvious result is that we’ll see a heavy emphasis on remarketing strategies (see the next item below). But the reality is that the breakdown by Audience Segments can also highlight a delivery problem, where Meta dedicates too much of your budget to remarketing using algorithmic targeting. If you don’t know that, you can’t adjust for it.

Even if Audience Segments are defined, I often see that they’re defined incorrectly. There is no indication that these definitions impact delivery (they’re for reporting transparency). But we’ll often see advertisers use random audiences to define the Engaged Audience group in particular, thinking that this will help direct delivery to that group.

I recommend being as thorough as possible when defining Audience Segments.

Engaged Audience: All Website Visitors (180 Days), Entire Email List, Lead Form Custom Audiences (90 Days), etc. (whatever else is applicable)

Existing Customers: All Purchase Events (730 Days), All Segmented Purchases From Email List, etc. (whatever else is applicable)

Limiting these segments to only recent website visitors or your most valuable customers won’t change how your ads are delivered. All it does is mislead you into believing that there’s less remarketing happening by default.

3. Heavy Lean on Remarketing

Numbers 2 and 3 on this list are often connected, but they don’t need to be.

If you don’t know that remarketing happens automatically through algorithmic targeting, you’re likely to believe that Meta needs you to isolate your remarketing audiences to get the best results. And those results will make you believe that you’re making the right decision (more on that in a moment).

There are rare exceptions when remarketing makes sense. The most obvious examples…

Selling High-Ticket Products: You aren’t spending enough to exit the learning phase, or get even close. You’re not getting results by targeting broadly. But you can isolate a small group who are deep in your funnel with ads to reinforce messages they are also getting via email.

When using this approach, though, you understand that your ads aren’t working alone. High-ticket sales require reinforcement, so your ads are making your email messaging more effective while your email messaging will make your ads more effective.

Optimizing for Top-of-Funnel Actions: I don’t generally recommend ad sets optimized for link clicks, landing page views, ThruPlay views, or any other engagement-level action. Since Meta’s delivery algorithm is only focused on getting you as many of those actions as possible, there’s no consideration for getting quality results.

So if you’re going to optimize for this type of action, remarketing can make sense — particularly if it’s restricted to your email list or engaged website visitors. This helps prevent spend that results in empty actions.

But these two are exceptions and should be used sparingly. If I see a heavy emphasis on remarketing, it’s likely doing a bad job of driving incremental results.

4. Inflated Conversion Results

And that takes us here.

You want an agency that obsesses over generating high-quality, meaningful results. They’re not looking to impress clients with inflated results that don’t reflect reality.

Use the Breakdown by Attribution Settings to see what percentage of the reported conversions are 1-day click versus 1-day view, 1-day engage-through, or even 7-day click. The 1-day click results are those we can most confidently say reflected the impact of our ads. While customers clicked an ad for 7-day click results, there was likely a secondary marketing effort that closed it.

But 1-day view and 1-day engage-through are lowest on the impact scale. View-through requires only that an impression was served before a conversion happened within a day. Engage-through requires some sort of engagement or click other than a click on the link. In either case, they may not have looked at your ad at all.

This is where remarketing can inflate your results. You’re showing ads to people who are getting emails from you. While we can’t say for sure that the ad didn’t drive the conversion, we do know that they didn’t click through to your website. And if that didn’t happen, it’s questionable how much your ad did.

You should also use the Breakdown by Conversion Count. This will help you get a better sense of unique customers versus total conversions. And if you have deduplication issues, you’re likely to see a doubling of results.

Does the agency care more about results that look good or results that are actually good? Those who care will use these breakdowns. If they uncover these types of problems, they will make adjustments that will lead to incremental conversions.

5. Demographic and Placement Restrictions

Demographic and placement restrictions tend to be a sign that the agency is applying unnecessary control that could hurt performance.

I’m not suggesting that delivery is always perfect or that there aren’t very real problems that can happen related to demographic and placement delivery. But we shouldn’t approach either of these in a universal manner.

There is one exception. If you promote age-restricted goods, you should restrict by age. There’s no reason to take any other approach.

But if you’re running a Sales campaign optimized for purchases, there’s rarely a reason to restrict by age, gender, or placement. Many advertisers assume they need to restrict by age or gender to match their ideal customer. “People under 25 don’t buy from me.” If that’s true, Meta likely knows it.

The same tends to be the case with placements. Audience Network is a common source of low-quality traffic, but that’s usually when you’re optimizing for a link click or landing page view. Meta thinks you want clicks. But if you optimize for a purchase, you’re likely to see delivery to that placement disappear.

Sometimes, you have information that Meta doesn’t. Here are examples:

1. The performance goal is “maximize number of conversions” where the conversion event is a purchase. You get a high volume of purchases from people under 30, but these customers represent a much lower customer lifetime value.

2. The performance goal is “maximize number of conversions” where the conversion event is a lead. You get a high volume of leads from people over the age of 65, but they’re low quality.

3. The performance goal is “maximize number of ThruPlay views.” You get a high volume of ThruPlay views from the Audience Network Rewarded Video placement because people are incentivized to watch your video in exchange for virtual currency.

4. The performance goal is “maximize number of post engagements.” You get a high volume of post engagements from men, even though 99% of your customers are women. The men who engage don’t buy from you.

These are all examples of a weakness related to Meta’s ad delivery algorithm being “literal.” You said you want a specific action, and Meta knows that you can get lots of those actions from a specific group or placement. You actually want people who will end up buying from you, but Meta doesn’t know that.

You have information that Meta doesn’t. In those cases, the instinct may be to restrict by age, gender, or placement. Instead, use a value rule to lower the bid on that problematic segment.

Remember that these segments aren’t problematic universally. It depends on the situation, your performance goal, and whether you can get a high volume of cheap and low-quality actions that satisfy that performance goal.

Prioritize avoiding restrictions whenever possible. Use value rules to keep Advantage+ settings on.

If you or your agency restrict by age, gender, or placement often and don’t use value rules, I know that it’s likely that performance will suffer.

6. Complicated Campaign Construction

The more budget that’s available to you, the more flexibility I give for this one.

If you’re spending $50 per day and have three active campaigns, each with multiple ad sets, you are seriously watering down your results. None of those ad sets is able to perform optimally.

Most of the reasons for additional ad sets are misguided. You should rarely create separate ad sets for different detailed targeting or lookalike audiences because those inputs are almost always used as suggestions now. You’re going to create multiple ad sets that can reach the same people.

Your priority should be running a campaign for sales in most cases. If you can’t get the results from sales, focus on leads. For low-budget businesses, you should concentrate as much of that budget as possible to a single goal.

Within the campaign for that single goal, you should limit unnecessary ad sets. You know that each time you create a new ad set, less spend can go to each one. You’re spreading a $50 or $100 or $200 budget across four ad sets, rather than focusing it entirely on one.

Exiting the learning phase isn’t a requirement, but it should be a goal. If you’re able to get 50 conversions in a week for an ad set, Meta believes delivery is stable and predictable. Less than that, and expect the unexpected.

I’m not saying that every business should use only a single campaign with a single ad set. But it’s a great place to start for the vast majority of them. It consolidates your budget, prevents auction overlap, and gives you the best opportunity to get good results.

Make sure you have a good business reason for each new campaign and ad set you create. Will pulling budget from your most important ad set be worthwhile?

Higher budget businesses have more flexibility here. They can split up their budget across multiple campaigns and ad sets and still exit the learning phase in all of them. They can create Awareness and Engagement campaigns and not pull money from the campaigns focused on conversions.

But even then, it doesn’t mean that it’s advisable. Big or small budget, your ultra-complicated campaign structure may be making your results worse. It rarely makes them better, and it also makes it difficult to figure out what’s working and what’s not.

When an account is struggling, a complicated campaign structure is common, regardless of available ad spend. It becomes so complex that no one is sure where to start to fix it. The first step, in most cases, is scrapping the complicated approach and starting simple.

7. Obsession With Ad Copy and Creative Control

A simplified campaign construction is recommended now. While all of the various levers and settings could have been the secret to performance a decade ago, they aren’t now. And that puts the weight of performance on your ads themselves.

If you’re not getting the results you want, start with the ads. Your ads aren’t good enough, and they could always be better.

But the way many agencies approach ad creation can be counterproductive. You can’t look at ad components in 2026 through the same lens we used five and 10 years ago.

There was a time when finding the winning combination of ad copy and creative made sense. One ad was once defined as a single primary text block, one headline, and one image or video. When one ad performed well and one didn’t, it was obvious which components contributed to those results.

You shouldn’t be creating ads this way now, though. One ad should consist of up to five primary text blocks, five headlines, and as many as 10 initial images and videos. But this is truly just getting started.

One ad can also consist of AI-generated text, images, and videos. Different cropped variations. Versions that are changed with Advantage+ enhancements. If you have the new creative workflow, you can even customize the landing page by creative.

You can’t look at ads the same now. A single ad behaves more like the old ad set once did. It consists of thousands and thousands (if not millions) of possible combinations of ad copy and creative.

And yet, some agencies still obsess over finding winning combinations. They create ads that only consist of a single primary text, headline, and creative, like it was 10 years ago. They create separate testing ad sets just for these ads, often optimize for clicks, and then move the “winners” over to the main ad set.

And then they’re surprised when that winning ad doesn’t perform the way they expected. So they turn off other ads and try to force budget to the winner.

You no longer have control. Forcing your control over ads is not a sustainable strategy to get better results. You’re only looking busy for the sake of trying to appear valuable.

You should know what’s working and what’s not, at least to the best of your ability. You should learn from that information so that you can create new ads that leverage the strengths of the old ones.

But those who are obsessed with creative control, particularly at modest budgets, are often making results worse. And far too many agencies are doing exactly that.

Get Your Account Audited

These are the first problems I’ll look for when auditing an account.

My Strategic Meta Ads Audit isn’t for everyone. If you or a client spend at least $10,000 per month on ads and need a second opinion to help isolate your current performance problems, you’re a likely fit.

Learn more about how my Strategic Audits work here.

Your Turn

Are there any other common mistakes I didn’t list here?

Let me know in the comments below!

The post 7 Common Meta Ads Mistakes Agencies Make appeared first on Jon Loomer Digital.

Are your ads sending quality traffic to your website? There’s a way to find out.

Misleading Results

Ads Manager results can be misleading. When running Sales ads, we tend to focus primarily on sales and revenue numbers. But when sales aren’t happening, it can be difficult to diagnose the problem.

Advertisers tend to obsess over Click-Thru Rate and Cost Per Click in these situations. They think that if those numbers are good and sales aren’t happening, the problem must be the landing page.

But that makes the classic assumption that we’re sending quality traffic in the first place. There’s a way to verify it.

The Custom Metric

I created a custom metric called Quality Traffic Percentage. It starts with a custom event that tracks when someone spends at least 15 seconds on the page (here’s how to create a similar event). This is done using a timer trigger and GTM.

What you use for this “quality” definition is up to you. Don’t get lost in isolating bottom-of-the-funnel actions. My main priority with this metric is to eliminate immediate abandons and bots that could potentially be blamed on optimization or distribution issues.

I then find the percentage of total landing page views that result in a 15-second visit.

You’ll need to use the compare attribution settings feature to isolate 1-day click and first conversions before finding the percentage.

Otherwise, this number will be inflated. Unlike a purchase or lead, an action like a 15-second visit might be performed repeatedly. We don’t care about any of these conversions after the first one.

Using the Percentage

The percentage, by itself, doesn’t have much meaning. It gains meaning when you begin comparing campaigns, ad sets, and ads. You’ll start to find that some ads never had a chance because of the low-quality traffic that was going to your website.

I highlighted this metric a few weeks ago when discussing the low-quality traffic sent when optimizing for landing page views (despite the Meta AI Business Assistant claiming the opposite).

In that test, I found that the Quality Traffic Percentage was similar, on average, for my ad sets optimized for purchases and leads. But they were 7X higher than the Quality Traffic Percentage when optimizing for landing page views.

The post How to Track Quality Traffic Percentage appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

I spent a week with ChatGPT ads, and my initial reaction is mostly underwhelmed. Meanwhile, Meta rolled out or tested several updates that caught my attention.

1. A week with ChatGPT ads
2. AI-powered instant form creation
3. 730-day purchase retention goes live
4. Media translation
5. Image generation update
6. Business AI prompts and responses

Let’s get to it…

1. A Week with ChatGPT Ads

Now that I’ve been working with ChatGPT ads for about a week, I have some thoughts. And I’m mostly underwhelmed.

The simplicity is mostly refreshing when compared to Meta ads, but there’s a point where it’s just way too basic. Each ad has a single headline that truncates after 24 characters, a description that truncates after 48 characters, and a small square image.

That’s not a lot to work with. If you’re getting bad results with Meta ads, the options are endless for improving them with your ads. Different formats, aspect ratios, text lengths, text angles, visuals, you name it. But these tiny ads are barely visible in the first place, and you have so few characters and pixels that your options are limited.

I know that a conversion objective is coming soon, but that’s an absolute must at the cost of these ads. Otherwise, we’re paying an average CPC of $3 to $5 and CPM around $50 to optimize for clicks or impressions. If the underlying delivery system doesn’t care whether people convert, that’s a lot to pay.

It seems unlikely that ChatGPT ads will be a good fit for the basic lead magnet or low-ticket offer for that reason. The potential is there for these ads to be very effective due to the intent of a conversation, but there are too many limitations in the current version of these ads to get particularly excited.

The biggest issue for tech companies is that our most likely ideal customer (those who subscribed to a paid tier of ChatGPT) is unreachable with ads, even though they didn’t sign up for that subscription with that benefit in mind.

The conversion tracking is also a mess, though it’s always possible I’ve set something up incorrectly, but it doesn’t help that there’s a generic “Conversions” column and there is no explanation of what goes into that. No idea what the attribution setting is or whether it includes more than the conversion event that I selected in the campaign.

Anyway, I keep reminding myself that this is the beta version. It’s going to look completely different eventually, but OpenAI has a long way to go. The biggest limitation is likely to be inventory and effective placements, and I’m not sure how they’re going to get around that.

2. AI-Powered Instant Form Creation

Some ad accounts have access to a new feature that uses AI to generate an instant form from a landing page form. If you have it, you’ll see this when you click to create a form…

You need to provide the URL to a landing page where you embed a form. Meta will use AI to learn from that page and generate an instant form.

It can take up to a minute…

When I did this, Meta didn’t try to generate an image from the page. That’s not necessarily bad, since it will just use the image from your ad if you don’t upload one specifically for the form.

Otherwise, it pulled some very basic text for a headline and description.

It seemed to mimic my landing page form and ask for only first name and email address.

And then it generated a VERY generic ending screen with text that I wouldn’t use (referring to “our” website instead of “my”). And for some ridiculous reason, it uses the URL I provided for the landing page as the destination URL.

Why would I send people to my landing page to complete a form after subscribing to that same thing with an instant form? It’s nonsense.

Overall, this feels like an excuse to use AI that doesn’t necessarily do anything that’s particularly helpful. It may be a starting point, but that’s it.

3. 730-Day Purchase Retention Goes Live

A couple of weeks ago, I wrote about how Meta is expanding the maximum retention for purchase event custom audiences from 180 to 730 days (or from six months to two years). At the time, Meta said that any website or app custom audience based on a purchase event with a 180-day retention would automatically expand to 730 days on May 18th. That is, unless you opted out of that change.

Well, I didn’t opt out, and it doesn’t appear this change was forced after all. I have a 180-day website custom audience based on the purchase event that remains at that length.

There’s been some discussion in my Power Hitters Club – Elite community about how much this matters. Even if the retention for your custom audience was expanded to 730 days, I’m not sure what the risk or harm would be. It all depends on how you were using that custom audience.

But if you determined that this is something you don’t want, you could easily edit that audience or create a new custom audience at the original 180 days. It will update and generate immediately.

4. Media Translation Example

Last week, I covered the new AI translations feature. This feature will translate all of your text (primary text and text on media) into one of 11 languages.

At the time, I couldn’t get a preview of the translation of text on media to work. But now I have one.

I still can’t get it to work all the time. It seems that simpler images with bold text like this one are more likely to be translated than others.

5. Image Generation Update

I noticed that image generation has been further refined in my creative workflow. And the images are starting to look more usable.

When you get to the image generation step, Meta will use AI to generate several images, inspired by the original image and text. I’ve found most of these AI-generated images to be slop, and there hasn’t tended to be much of a common theme.

But that’s changing. Now I’m seeing three categories of AI-generated images:

• Refined Original Style
• Popular in Professional Services
• High Return on Ad Spend

While I’m not ready to pick all of these, they’re way better than what Meta was generating before. There are several that are passable and could be good alternatives to the primary creative options.

It would be nice if there were a way to control the creative direction of these generated images. But we’re seeing progress.

6. Business AI Prompts and Responses

Meta is really pushing the Business AI feature now. Originally, it was showing up as one of the many enhancements you could turn on and off, but now there’s a separate section for it in the ad creation flow.

If you turn this on, simple prompts may appear below your ad. Meta will use AI to develop them based on your product. There’s no way to control them, unfortunately.

Meta provided two sample prompts for this ad. If you hover over one, you can get a sense of what people will see if the prompt is clicked.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post ChatGPT Ads Limitations, AI Instant Forms, and More appeared first on Jon Loomer Digital.

OpenAI launched the self-serve ads beta to advertisers in the United States a couple of weeks ago. I finally got access late last week, and I’m eager to share what I’m seeing so far.

Be ready for a theme: BASIC. This is truly a beta product. OpenAI Ads Manager is basic in every way. I don’t think you can appreciate just how complicated Meta advertising has become until you create an OpenAI campaign.

The main point of this post is to share my initial hands-on experience. What we see today is likely to look different tomorrow. But I want you to know what you should expect when you create a campaign for the first time.

Within this post, I will share what the campaign creation process looks like today using OpenAI Ads Manager, and I’ll compare it to what we’ve come to expect with Meta ads.

We Know Nothing

That’s right. Nothing. Beware of the opportunists. There is no reason to pay for a course about how to run ChatGPT ads right now.

No one really knows anything. And the context of what we know will keep shifting as features are added and functionality changes.

Most of our knowledge of ChatGPT ads is limited to official documentation. But as I’ve found, much of the wording there is top-level and generic. You can blow through all of that information and still come away with a whole lot of questions.

Should you use CPM or CPC bids? How should you approach bidding? How much do Context Hints matter? What makes a good or bad Context Hint? How does OpenAI approach conversion attribution? How effective are these ads? Do they make the most sense for high-ticket sales? What costs should we expect?

I don’t know, man. We have our own experiences. We have OpenAI’s documentation. But it’s all so early and limited, and so much of what we see now won’t be relevant as all of this expands months from now.

Just remember that.

Create a Campaign

When you create a campaign, you’ll get a flow that looks like this…

Name your campaign.

Choose an objective. For now, your only options are Reach and Clicks, though Conversions is “coming soon” (NOTE: OpenAI’s documentation currently refers to Reach as “Views”). Your objective will determine how you bid in the ad auction.

Location is selected at the campaign level. United States is selected by default. At the moment, you can also choose Canada, Australia, and New Zealand.

Regional DMAs are also available.

Set a daily or campaign (lifetime) budget. You can’t change the daily vs. campaign selection later, but you can change the amount.

You can optionally select a conversion event that’s used for reporting. You’ll need to add the OpenAI pixel or set up the Conversions API first before creating an event. I’ll cover that briefly later.

Finally, set a start and optional end date.

Compared to Meta Ads Manager

You can set an objective and budget at the campaign level. Those are about the only similarities between campaigns on OpenAI Ads Manager and Meta Ads Manager.

Objective options are limited, to say the least, with ChatGPT ads. Meta offers six objectives, whereas OpenAI offers only two. The objective you select is much more closely tied to how ads will be delivered using ChatGPT ads. The objective is mostly decorative with Meta ads.

Location is set at the campaign level with ChatGPT ads, whereas it’s at the ad set level for Meta ads. The locations available for ChatGPT ads are limited to four countries and DMA regions for now.

You can set a conversion event at the campaign level for ChatGPT ads, but this is a step that wouldn’t be necessary for Meta ads. Meta will report conversions without selecting them in the ad set. For ChatGPT ads, I assume the conversions will not be reported unless the events are selected in the campaign.

Create an Ad Group

Next is the ad group…

Name your ad group.

Next, you’ll set a maximum bid. If you chose the Clicks campaign objective, you’ll set a maximum CPC bid. OpenAI recommends starting at $3 or $5.

If you selected Reach as your campaign objective, you’ll set a maximum CPM bid. OpenAI recommends starting with $60.

While this feels like trial and error, OpenAI will give you an indication of whether your bid will result in strong delivery.

If not, you’ll get a message of “May Not Deliver.”

Provide a website URL. This is only used as a default destination for new ads in the ad group. Feels somewhat pointless since you’ll define the destination URL in the ad.

And finally, Context Hints. OpenAI says to “describe the conversations, topics, or keywords where your products or services may be relevant; these hints guide matching but aren’t exact-match targeting rules.”

This isn’t the only way OpenAI finds an audience. The other critical elements are in the ad itself.

Don’t look at Context Hints as a way of keyword stuffing. Instead, OpenAI says good Context Hints describe user intent. Your goal is to “describe the types of questions, needs, or situations a user might bring to ChatGPT.”

Since I feel pretty clueless on this one, I used ChatGPT to help me with it. And since ChatGPT knows my content and product well, I feel reasonably confident about the Context Hints it comes up with.

This particular Context Hint is four paragraphs long. It’s something I will certainly be experimenting with.

Compared to Meta Ads Manager

The ad set in Meta Ads Manager is far more complex than the ad group for ChatGPT ads. Meta advertisers concern themselves with conversion locations, performance goals, bidding, attribution settings, targeting (suggestions, restrictions, and exclusions by age, gender, custom audience, lookalike audience, and detailed targeting), and placements. OpenAI Ads Manager has almost none of that.

There are two primary things to set within the ad group, and they are likely both impactful. The bid will determine whether your ads get shown or not. Context Hints replace targeting and will help define who sees your ads.

Both will require significant trial and error. It’s not at all clear how much Context Hints impact who sees your ads. That would help determine what makes for “good” or “bad” Context Hints.

Create an Ad

And finally, create your ad. But it’s not nearly the process you may be used to…

Name your ad.

The website URL will pull from the ad group, but you can edit it for each ad. It’s highly recommended that you use UTM parameters, particularly during these early stages of ChatGPT ads where tracking accuracy and completeness are unknown.

The Ad Creative section is extremely basic. Provide a headline that is up to 50 characters. Just know that it may be truncated on some placements if it’s over 24 characters.

Provide a description that is up to 100 characters. It may be truncated if it’s over 48 characters.

And a square image that is at least 256 x 256 pixels.

Here’s an example of what an ad might look like…

Very little text. Image is small. Avoid images with small text and details that won’t be seen when viewed at this scale.

Compared to Meta Ads Manager

It’s as if Meta Ads Manager and OpenAI Ads Manager exist in different worlds when it comes to creating ads.

Meta Ads Manager: The ads are everything. Focus on creative diversity of format, visuals, and text. Big, bold images and videos. Different aspect ratios. Provide up to five primary text and headline options. AI-generated text, image, and video options. Lots and lots of enhancements.

OpenAI Ads Manager: One ad is a single combination of headline, description, and square image. It’s small. No videos. No enhancements. No comments on ads to manage. You can create multiple ads, but otherwise it doesn’t get much simpler than this.

Measure Results

This isn’t going to be shocking to you, but reporting in OpenAI Ads Manager is currently very basic. I am currently seeing the following columns:

• Active (On/Off)
• Name (Campaign, Ad Group, or Ad)
• Status (Serving, In Review)
• Actions (?)
• Impressions
• Clicks
• Spend
• CTR
• Average CPC
• Average CPM
• Budget
• Start Date
• End Date
• Conversions

A few thoughts…

The first campaign I published was approved immediately. Meta advertisers will appreciate that.

Unfortunately, it seems as though reporting is much less dynamic than in Meta Ads Manager. OpenAI says to expect reporting to be delayed up to seven hours. Combined with the importance of bids, it wasn’t clear whether reporting was just delayed or I wasn’t getting any delivery.

This is also why UTM parameters can be helpful. One of the ads I created was for a lead magnet that utilizes a unique form that’s only used for these ads. Together, they gave me additional ways to monitor whether anything was happening.

I’m pretty sure the “Actions” column was added within the past day or so. I’m not seeing any documentation on it, and it’s blank for me.

If your objective is Clicks, you won’t be charged unless there are clicks on your ads. One clear difference from Meta Ads Manager.

OpenAI Pixel and Events

You can track conversions with ChatGPT ads in mostly the same ways you do with Meta ads.

Go to the Conversions section under Tools on the left.

You’ll first need to create a data source.

Enter the data source name (like your company website) and type (only “web” is available right now).

I set up my pixel and events using GTM. While it seems rather straightforward and similar to the process of setting up the Meta pixel and events, I’m not confident enough to share those exact steps with you here. I don’t want to mislead you.

OpenAI’s developer documentation is the best place to start, especially because the setup details are likely to change while this is still in beta. Start with these links from OpenAI’s official documentation:

• JavaScript Pixel
• Supported events
• Conversions API

Initial Impressions

I hate repeating myself, but there isn’t a whole lot to say about my early impressions of ChatGPT ads. It’s all so basic. And it’s far too early to get excited or frustrated about results.

But as someone who has been conditioned to accept the layers of complexity that exist in Meta Ads Manager, ChatGPT ads feel like a nice return to basics. I don’t know if this was the intention as a clear alternative to Meta ads or if it’s more a reflection of a product getting thrown together. But it’s oddly nice.

If results aren’t what I want them to be, it’s nice that there are very few levers to focus on to improve things. Although, I guess that could also be a limiting factor.

I’m excited about the potential of these ads because I know that ChatGPT already refers people to my website. But I’m skeptical that the current version of these ads will be all that effective.

While the potential for highly contextualized ads is great, I’m not sure that the current format will inspire many clicks and actions. We’ll see, but these small ads and their simple format seem awfully limiting.

I also worry that ChatGPT ads may have been ruined for tech companies before even getting off the ground. The OpenAI paid tiers were created without ads in mind. People subscribed to those tiers to get access to additional functionality. These would be the ideal target customers for many of us. Unfortunately, they are automatically excluded from ads.

That’s different from most subscriptions. You pay to remove ads from YouTube (or even Facebook and Instagram). The people paying for the ChatGPT upgrades weren’t doing it under the assumption that it would remove ads. And if costs are higher to run these ads, it could present a problem where the most valuable potential customers are unreachable.

Of course, that’s a concern for another day. For now, it’s fun experimenting.

Your Turn

Have you started running your first ChatGPT ads? What do you think?

Let me know in the comments below!

The post Hands-On With ChatGPT Ads: Initial Impressions appeared first on Jon Loomer Digital.

A few weeks ago, Meta announced AI Connectors. This allows advertisers to connect their Ads Manager to AI tools like Claude and ChatGPT.

But most advertisers don’t understand the exposure this creates.

Business Portfolio Integration

When setting this up, you choose which businesses you want Claude to access.

Either a specific business or all and future businesses. By “business,” this means your business portfolio, which was previously called a business manager.

This includes all of the ad accounts you control under that business.

You cannot choose which ad accounts Claude can access. And that can result in exposure to risk that you or your clients do not want.

What Can You Do?

This feels like a major oversight on Meta’s part, and I’m not sure most advertisers realize that this is the case. Assuming your clients give your business access to their ad accounts to manage, Claude would have access to all of those accounts if you made this connection.

Right now, the only thing you can do is explicitly tell Claude which ad accounts it can touch. But that puts a lot of trust in Claude since there isn’t a setting that controls this.

Meta needs to fix this. Advertisers will surely want to limit which ad accounts AI tools can access, particularly until this technology becomes more mature.

The post AI Connectors Might Put Your Clients at Risk appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

I spotted a whole bunch of new features and design changes during the past week. It’s not always clear whether what I’m seeing is part of a global rollout or a test. I’m also excited to have one feature activated while access to another remains in limbo.

1. One-Click CAPI Activated
2. Format Display Options Update
3. Video Generation Returns
4. AI Translations
5. Text Generation Update
6. Value Rules Updates
7. Live Video Ads
8. ChatGPT Ads: Still Waiting

Let’s get to it…

1. One-Click CAPI Activated

About a month ago, Meta announced a new no-cost, one-click Conversions API solution for web events. A few weeks ago, the option finally appeared in my Events Manager, but there was a problem.

I saw the option in my Events Manager settings…

And the option was available when I went the route of clicking “Connect Data” in Events Manager…

And I even was given a green verification message that it was active…

But it wasn’t actually active. The button and option to turn it on never disappeared, and the data never came through my Events Manager.

Or at least, it took a couple of days. My understanding was that there was a bug where requests were getting stuck in a queue. A few days later, I’m happy to report that all is now working.

The button to “Connect Now” in Events Manager has been replaced with a display indicating that Conversions API is, indeed, active.

And I now see server events coming through my Events Manager for all web events.

If you weren’t getting this to work before, your web CAPI may be active now. If not, try connecting again.

My main setup includes the pixel, Conversions API Gateway (web events), and Conversions API (CRM events).

2. Format Display Options Update

We’ve known for a while now that Format Display Options is the future of format selection when creating an ad in Ads Manager.

The message above was appearing under Format when creating an ad previously. In my Ads Manager, it’s been replaced with this…

The Format Display Options section looks a little different for me, indicating whether or not certain formats (Single Media, Carousel, and Collection) are turned on.

Single Media

If you uploaded individual images and videos, you can’t unselect the first option. You can also choose to create a video slideshow that combines one of your single media and site links or product links.

Carousel

Carousel will combine your single media with your site links or product links.

Collection

And Collection combines… Well, you get it now.

Everything other than single media is controlled within Creative Setup.

Make sure to customize your site links (thumbnails, URLs, and labels) if you haven’t already.

3. Video Generation Returns

Back in January, I wrote about Meta’s Advantage+ Creative Video Generation feature. It created videos out of up to five static images. Within days of writing that post, the feature disappeared from my Ads Manager.

Well, it’s back, and with a bit of a facelift. Video Generation is now part of the new creative workflow.

First, Meta will generate video concepts that you can choose from.

Meta will start with three concepts built around your single media and up to four other scenes, which can be auto-generated.

You can make edits to the scenes if you’d like. Otherwise, generate the videos.

Here’s a short example…

Personally, I find it pretty cool! A creative way to add variation.

4. AI Translations

I’m seeing a new Translation step during the creative workflow.

You may also see this under Essential Enhancements, and above Creative Testing.

You can select from up to 12 languages for translations (all selected by default).

You can choose to translate all ad creatives and text (including text on images) or only parts of your ads.

A preview of the translations is on the right. Here’s the Instagram version.

Interesting to note that the image is not translated here.

Should you turn this on? It’s experimental and requires some trust. Maybe not for sensitive clients. But this is the type of enhancement I’m willing to try out and take the risk when running ads to promote my own business.

5. Text Generation Update

AI text generation isn’t new. But how it’s displayed is changing.

I’m currently only seeing this during the Text step of the creative workflow, which is odd. I can’t find it after this process is completed.

Here’s an example of one block of AI-generated text.

You have to expand it to see what’s within it.

So in this case, there were three AI-generated headlines and two primary text options that were hidden under the main block. All are selected by default. You can’t individually remove one headline or one primary text block. Turn them all on or off.

There is some good and bad here. It looks nicer, but that’s not all that important. I also feel like the quality of the AI-generated text is better than it used to be.

But the way these are all on by default and you have to expand individual blocks to see what was generated is misleading. Not a fan.

6. Value Rules Updates

Meta made a couple of organizational changes to value rules.

The first update was to the creation of value rules. Here’s how this looked previously (or may look now for others). Select from a dropdown menu of the available criteria…

But mine is broken into three categories.

Conversion Location includes Website, App, and Instant Forms…

Audience includes most of the other criteria…

And Placements currently only consists of what was already available for placements criteria.

On one hand, this feels unnecessary. But it tells me that Meta has plans to create many more value rules criteria. Eventually, that single dropdown menu will otherwise get too long.

The other update is that you can now add and create value rules from the Placements section in the ad set.

I understand the thought process behind this. Meta wants to discourage advertisers from removing placements. Putting value rules here gives them a quicker alternative.

The execution, though, is messy. It would make sense if you could only select from value rules that utilize the placement criteria. But it’s instead just a new place to select any value rule.

Why doesn’t a similar section for value rules appear in the Audience section? I’m not saying it’s a good idea, necessarily. That’s messy, too. I’m just not so sure about the best way to do this.

7. Live Video Ads

I had heard rumblings about Live Video Ads during the past few months, but I finally have access.

From what I can tell, it’s available for most campaign objectives. If you have it, a Live Video Ad toggle appears above Campaign Details.

I only have Facebook as an option, but I assume others may have Instagram, too. It appears that a lifetime budget is required because it defaults to lifetime. When I switch to daily, it switches back when I move to the ad set.

Within Budget & Schedule at the ad set level, you can choose to promote either a “current live video” or an “upcoming live video.”

For a current live video, you’d be running ads for a video that’s happening right now. Meta says you’d need to select an existing post that features your live video for your ad under Ad Setup. It’s confusing since ad creation defaults to creating an ad from scratch, at least for me.

If you choose to promote an upcoming live video, you’ll need to schedule your ad at least three hours before your video starts. Meta says the start and end time of your ad should match that of your live video.

At the ad level, “Use Scheduled Live Video” is selected by default in Ad Setup.

Then provide a title and description of your live video.

This feels incomplete. But Meta says that after publishing your ad, go to Live Producer via computer or mobile device to complete the setup.

I haven’t hosted a live video from my Facebook page in years, so I can’t confirm much of this. But it could certainly be useful for those selling with live videos.

The most interesting part to me is that you can select a performance goal to maximize the number of purchase conversions, for example. You’re not forced to optimize for video views or engagement. So instead of getting random, quick views, you may actually get people likely to convert.

8. ChatGPT Ads: Still Waiting

Just last week, I wrote on these pages to tell you that ChatGPT ads are now self-serve and available to all advertisers in the US (I missed that country-specific detail). I walked you through how to create your account and submit your information for business verification.

Well, a week later, and I’m still waiting.

I get that there’s a high volume of signups, but this is ridiculous. The volume should have been expected. The problem with the delay is you’re bound to have a decent number of advertisers who were excited to get started who will just lose interest.

That won’t be me, but I’m still annoyed.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post One-Click CAPI Activated, New Meta Ads Features, and More appeared first on Jon Loomer Digital.

It’s been more than a year and a half now since Mark Zuckerberg said this on a Meta earnings call:

Over the longer term, advertisers will basically just be able to tell us a business objective and a budget, and we’re going to do the rest for them.

It was controversial at the time, mostly because it felt like fantasy based on Meta’s integrations of AI to that point. But with each new advancement, both within Meta Ads Manager and outside of it, we get closer to the statement being a potential reality.

It’s not abstract anymore. Fully automated campaigns are Meta’s goal, and we’re beginning to see the building blocks that will make it possible.

The question isn’t whether some version of Zuckerberg’s vision is coming. You have to bury your head in the sand to deny it. The biggest question is how this evolution will impact the future roles of advertisers.

I’ve grappled with these questions myself. I’ve never claimed to be particularly good at predicting the future, but I try to pay attention to where the current trends are taking us. It’s critical we begin to figure out how we contribute to this future, if we haven’t already.

Admittedly, my brain cramps while thinking through some of the scenarios. They’re scary and mind-bending, but also counterintuitive and contradictory.

I won’t have answers in this post. What I do know is that most of the takes I’ve seen on this fall into one of two camps, and both of them are likely wrong.

Two Extreme Reactions

The impact of AI on the future of media buying, Meta advertising, and the digital agency is a common discussion these days. The vast majority of opinions fall into one of two camps.

1. The modern advertiser is dead.

This opinion assumes that AI will make advertisers obsolete. We’ll all be replaced by AI, either through Meta’s enhancements or third-party integrations. It’s time to begin looking for a backup occupation.

2. AI is flawed and people will always be needed.

This opinion assumes the noticeable flaws of today’s AI will exist in a similar way a year or more from now. People on this extreme cite the many limitations of AI in its current state. They may even call it a fad.

By nature, we like to simplify things. When thinking of the future, it’s nearly impossible to think about the nuance related to factors that do not currently exist. So it’s easy, as a way of dealing with the unknown, to approach the future of AI in advertising as either devastating or not that big of a deal.

Two Extremes of the Market

While I try to predict where this will go, I like to think of two very opposite types of brands as a thought exercise.

1. The biggest brand.

Whenever I’m asked about using AI to create and manage a campaign from start to finish, I like to think of a company like Nike. Would they trust their brand to an AI agent? Fully automated copy and creative without legal approval? Unlikely.

The most bullish AI supporters may say that there will simply be an AI agent with branding or legal expertise (and that’s not crazy), but it’s hard to believe the biggest brands would be okay with that, at least early.

A lot of what looks like ads management at that level is really the work of translating brand strategy into media decisions. Sponsorships, talent, and brand campaigns are all work that is unlikely to transition to AI agents. Big brands like Nike will certainly invest in AI, but there are areas where they’re likely to be slowest to adapt, if they do at all.

2. The smallest brand.

The more I think about the prospects of automated campaign creation and management, the more I think it applies most to the smallest brands. It’s the “easy” button for small businesses that don’t have the resources to hire someone. It’s the “Boost” button of the future.

These brands have limited budget and either weren’t going to hire someone to manage their ads anyway, or they’re unlikely to commit to that relationship long-term. They haven’t seen the results when someone has managed their ads before (mainly because it’s so hard with small budgets), but they can at least replicate that performance without hiring someone.

While there may be flaws in this thinking, I can see why the smallest brands may be the best market for fully automated campaigns and ad management. No humans, no third-party integrations, just Meta’s built-in “tell us what you want and we’ll create it for you” feature.

These seem like reasonably likely scenarios, but they both touch on extremes. Most advertisers, and most of the brands they work with, will live somewhere in between.

The Middle

Extreme predictions are the simplest explanations, without nuance. Brands at both extremes will rarely reflect the majority. Much of the truth is going to be found in the middle.

It makes sense that the smallest brands would prefer the easy button and the largest brands would resist it. In the middle is a vast range of mid-sized brands with enough complexity that the easy button isn’t enough, but not enough scale to justify a full team.

“The middle” is where the future of the Meta advertiser and agency will be defined. Much of that will depend on how they make themselves useful, valuable, and irreplaceable as everything evolves. What more can they provide that the easy button can’t?

There are several questions that need to be answered that will help determine what’s next for this industry. And I don’t have clear answers for them.

First is what I call “The Auction Paradox.” Meta ads are delivered using an auction. This system relies on some ads being better than others to help determine what wins. If every brand has an AI agent optimizing against the same signals, where does the advantage come from? It would seem there needs to be something outside the optimization layer to do the work.

There’s also the access question, which might ultimately solve The Auction Paradox. It would be easy to assume that, thanks to capitalism, the better AI agents will cost more. Brands with the most sophisticated and expensive AI agents may have a structural advantage over the smaller advertisers and brands.

AI tools are largely accessible to most right now. It’s more a question of education. But it wouldn’t be surprising if these costs accelerated into a tiered system that would price out the majority.

Another question I have is about how the future advertiser works. Is the future advertiser managing multiple AI agents? Maybe they’ll be more efficient as a result, so they’re able to manage more clients. This could, conceivably, raise the bar on expectations related to quality and volume of work an advertiser takes on.

And finally, the future of Meta advertisers and agencies is found in the space in between that can answer this question: What new advertising-related needs and occupations are developed because of AI? Do they make up for the ones that are replaced?

What My Own AI Use Shows Me

I won’t claim to have all of the answers, or even any of them. But the clearest version of an answer keeps coming from my own work with AI.

My use of AI continues to expand with advancements in available tools and my own understanding of how I can use them. AI lets me do things I wouldn’t have considered before, and it lets me do them quickly.

This mainly includes the small things, like landing pages and featured images for blog posts. The output I’m able to generate is better than anything I could have created myself before, while taking far less time and effort.

I fully understand that the things I’m creating with AI still aren’t as good as the output a design expert could generate, particularly a design expert with the same or better tools. And that’s where we get to an important point that can be tied to the future of advertising.

If you work in design and ignore AI, the bar is being raised. If you previously produced average or below-average work, you will be easily replaced. But if you embrace AI to make your pre-AI work better and more efficient, you can improve both the quality and efficiency of your output.

The pattern doesn’t stop at design. Don’t ignore AI. It will replace many tasks. If you produce average results without AI, you will be replaced.

Those who survive will be the ones most educated about what’s changing and how to take advantage of it. Maybe most importantly, observe where new opportunities with AI appear.

Occupations related to online advertising are unlikely to disappear. But they’ll surely look different. The social media marketer who specialized in MySpace 20 years ago was forced to adjust. The same will be true here. If you don’t, you’ll become obsolete.

Some will use AI to become more employable. Some will move from ad management to software development to manage your ads. Instead of hiring people, you are buying software that was created by those who were displaced.

There’s one final layer to consider. The world of AI agents will be its own web of complexity. Which AI should you use? What produces the best output?

That requires an expert in itself.

AI Is Inevitable

It’s natural to assume the simple answer, and to predict the extremes. It’s easy to claim nuance and expect something in between. What you can’t do is ignore it altogether.

The ones who don’t make it will hope AI is a fad, or assume that the human touch, doing exactly what they’re doing now, will always be useful. That’s a dangerous game.

The fact that you are reading this post and care about how AI impacts your future puts you a step ahead. The majority are either unaware or unwilling to think about it.

Your Turn

How do you think AI will impact the future of Meta advertising, specifically, and the ad space, generally?

Let me know in the comments below!

The post The Future of the Meta Advertiser appeared first on Jon Loomer Digital.

Self-serve ChatGPT ads are here…

After a limited test beginning in February, advertisers in the US can now sign up for the self-serve beta.

Beta Features

Because it’s beta, expect the options to be limited.

The only campaign objectives available are Clicks and Views. No conversions yet.

Targeting? Not really. Instead, use “context hints.” Here’s how OpenAI defines it:

Describe the conversations, topics, or keywords where your products or services may be relevant.

The ad itself only includes title, copy, landing page, and image. No multiple options per ad.

Track impressions, clicks, and spend. Is that it?

Only the Beginning

So it’s… basic. While it would be easy to be greedy, we shouldn’t expect much more than this at this early stage.

Of course, that’s assuming you’ve been accepted into the beta. I applied last Wednesday, and I’m still waiting for approval. I created my account and provided my info to verify my business. And that’s where I’m stuck.

Truthfully, it shouldn’t be surprising that there’s a rush to sign up. There’s been curiosity about this ever since it was announced that there would be ads on ChatGPT.

While the ChatGPT ads self-serve beta won’t offer much in terms of control or special features, there’s mystery regarding how well these ads will perform. LLMs like ChatGPT are eating into search, and I’ve certainly seen both for my site. Way fewer search referrals, but way more LLM referrals.

While the volume still can’t compete with Google, the quality of referrals I’m receiving from LLMs is significant. I regularly hear from new customers who found me through Claude or ChatGPT.

Needless to say, I’m looking forward to trying out the beta once I finally have it. Are you part of the beta?

The post Self-Serve ChatGPT Ads Beta Details appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

ChatGPT ads are opening up to self-serve advertisers. This week’s notes also cover Meta’s AI connector rollout, longer purchase audience retention, “higher-quality clicks,” and partner request scams.

1. ChatGPT ads expand to self-serve
2. Meta AI Connector for Claude approval process
3. 730-day retention maximum for purchase events
4. What does Meta mean by “higher-quality clicks”?
5. Partner request scams

Let’s get to it…

1. ChatGPT Ads Expand to Self-Serve

Earlier this year, OpenAI announced a test of ads on ChatGPT. It would be a limited test of select partners, of course, with a minimum spend of $200,000.

And now OpenAI is making ChatGPT ads available self-serve. Go here to get started.

The setup process is generally pretty simple.

First, provide basic information about your business (legal business name, business website, favicon, and industry). Note that the favicon needs to be at least 128×128 pixels.

Next, provide country, currency, time zone, advertiser type (business or personal), and whether you’re an agency. It appears there are some additional steps required if you indicate that you’re an agency.

That creates your account. But then you’ll need to verify your business.

OpenAI says it’s “quick and easy,” and it mostly is easy. But not necessarily quick.

You’ll need to dig up your Business ID and then provide a physical address (no PO boxes).

And then verification will enter review.

That was mostly quick and easy. But how long will that review take?

Not surprisingly, demand is high right now, so I’m assuming waits will be longer than normal. I’ve been waiting several hours so far.

All indications are that the initial Ads Manager is rather basic for now, which shouldn’t be surprising. But it’s certainly worth testing. If we learned anything from those initial years of Facebook advertising, early adopters benefit before the rush.

2. Meta AI Connector for Claude Approval Process

Even though I was able to complete the process of setting up my AI Connector to Claude without issues, I’m not yet able to use it. Claude says it’s getting the following message from Meta:

Ads MCP is gradually being rolled out. Please check back at a later date to use Ads MCP with this Ad Account.

So once you get this set up, Meta still needs to enable it. I apparently have access for a couple of the ad accounts in my business portfolio, but not the one I want to use.

Based on a little Googling, it appears this is a common problem. If you’re running into it, you’re not alone.

3. 730-Day Maximum Retention for Purchase Events

This is weird.

At the top of the Audiences page, you may see this message…

It reads…

Maximum audience retention has increased for purchase events

Starting May 18, 2026, you’ll be able to set audience retention to a maximum of 730 days when you select purchase events in your website and app activity custom audiences. Your existing custom audiences with purchase events and 180 days audience retention will automatically update to 730 days.

If you would prefer to keep your current audience retention settings, you can opt out before May 18, 2026.

When you create a custom audience based on the purchase event (or any website custom audience, for that matter), the maximum retention has always been 180 days.

For whatever reason, Meta is extending this maximum to 730 days, or two years. It’s mostly odd due to the timing. Why now?

Like many advertisers, I contend that remarketing is rarely necessary now — especially general remarketing for the broadest audiences. Meta should already prioritize these people by default.

But where this could be most useful is for exclusions. One issue with excluding custom audiences is that they’re incomplete. The 6-month maximum retention was a big part of that weakness. Combined with a customer list audience, this could help make such exclusions more effective.

Presumably, Meta already has the data to complete these 730-day audiences, and they’ll fill on May 18th (assuming the event was firing that long). This provides further evidence that Meta has more data than simply that which we can define with custom audiences. We just don’t always see it.

If you’ve created website or app activity custom audiences based on the purchase event using a 180-day retention window in the past, know that they will automatically switch to 730 days on May 18th. If you don’t want that, you can opt out.

4. What Does Meta Mean by “Higher-Quality Clicks”?

I recently stumbled on an interesting note in Meta’s documentation about performance goals that I haven’t seen before. When discussing performance goals to maximize the number of link clicks or landing page views, it says this…

Note: To improve performance, we may prioritize higher-quality clicks.

This is news to me since the optimizations for link clicks and landing page views are notorious for driving low-quality traffic, often from Audience Network.

I decided to make use of the Meta AI Business Assistant to get more information, so I highlighted that note and prompted it for details:

Provide the specific ways that Meta prioritizes higher-quality clicks when using performance goals to maximize number of link clicks or landing page views, given that these are notorious for driving low-quality traffic.

The answer wasn’t particularly convincing. It said that Meta’s delivery system uses the following methods to ensure the traffic you receive is of higher quality:

Intent Verification: By optimizing for landing page views, we show ads to Accounts Center accounts who have a history of waiting for pages to load, rather than those who frequently click and immediately exit.

Pixel Confirmation: Unlike link clicks, which only track the action on the ad, landing page views require your website to successfully load. This ensures that the “Time on Page” you track is based on actual site visits.

Statistical Modeling: In cases where data is partial, we use statistical modeling (using mathematical patterns to estimate results) to account for views that our system identifies as high-quality based on user behavior.

That’s some paper-thin support for “prioritizing higher-quality clicks.” A landing page view is, by definition, measured when a page loads. And Meta recently eliminated the pixel requirement.

I almost never run Traffic campaigns, but I figured we should always test our assumptions. I’ve begged Meta to improve this type of optimization by prioritizing high-quality traffic signals (like dwell time), or at least provide that option.

I created a Traffic campaign with the performance goal to maximize the number of landing page views. I then created several ads to promote different blog posts that were shared to my page.

Years ago, I created custom events to help measure and optimize for traffic quality signals. For this experiment, I used two of them:

• 15-second visit
• 60-second visit

These should be pretty easy bars to pass for blog posts, especially at 15 seconds. And if higher-quality traffic is prioritized, we should see that in the results.

Custom events like these are easy to inflate because they can happen multiple times. So I used the results from 1-day click and first conversion, then divided by the total number of landing page views.

Traffic Quality for Landing Page Views Performance Goal

• 15-second visit/LPV: 8.4%
• 60-second visit/LPV: 4.4%

That seems bad. But we don’t have anything to compare these results to. So I also found the average of my numbers from the past 30 days when using performance goals that were optimized for website leads and purchases.

Traffic Quality for Leads Performance Goal

• 15-second visit/LPV: 54.0%
• 60-second visit/LPV: 27.5%

While the people Meta shows ads to for this performance goal may be higher intent, I also wouldn’t expect a long dwell time. The landing pages these ads send people to are very short, so even a successful conversion would likely happen in under 60 seconds. Even under 15 seconds isn’t impossible.

Traffic Quality for Purchases Performance Goal

• 15-second visit/LPV: 54.4%
• 60-second visit/LPV: 34.4%

Pretty much the same dwell time as traffic to my lead magnets, but with more 60-second visits.

What’s surprising about the results for the landing page views optimization is that a blog post, by nature, should generate longer dwell time. That is, of course, assuming Meta is prioritizing quality traffic and sending people who are likely to actually read it.

This experiment was hardly scientific, but it verifies my assumptions. The note that Meta “may prioritize higher-quality clicks” for performance goals related to link clicks and landing page views is flimsy, at best. They’re only “higher quality” because they’re landing page views. But Meta doesn’t seem to care at all about the quality of the visit.

And that, of course, verifies what I’ve long assumed about the deep flaw in Meta ads optimization. The algorithm is literal.

5. Partner Request Scams

Be on the lookout for the latest scam targeted at advertisers. I’m one of many advertisers who have started receiving an increase in partner requests via email.

The email above came from Meta, which will immediately throw some advertisers off. One of the most common flags is the email address a scam comes from.

This isn’t a situation of someone creating an email that attempts to look like it came from Meta. It actually came from Meta, and that’s part of what makes it a more effective scam.

What’s happening here is that the scammers are making (likely thousands or more) requests to random Business IDs to be added as a partner. Once they’re added as a partner, you can then grant them access to specific assets.

My hunch is that the true impact of the spam doesn’t only come from advertisers approving the partner request. Even if you go to your business portfolio and grant the request, it would require them to then convince you to provide the access necessary to do some damage.

There’s something else here. The email indicates the partner request comes from “Agency Pro Network.” The part I’ve fuzzed out is a link. This scammer has apparently included a URL in their business portfolio name, which is then hot linked when included in Meta’s emails.

What happens when you click that link? I don’t want to find out. But I’m sure nothing good.

Meta clearly knows about these scams since a large percentage of this email is warning you about them. It aims to clarify that the request is not coming from anyone affiliated with Meta, and then goes on to warn about fraud.

The first thing Meta could do is deactivate links that appear in business portfolio names. Maybe I’m wrong about how that’s working, but it’s otherwise a very critical security flaw.

This has become a commonly reported scam recently, so be on the lookout. As always, never click these links in emails that appear to come from Meta. Go directly to your account instead. And always be suspicious of anyone wanting access to your assets.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post ChatGPT Ads Go Self-Serve, Purchase Retention Expands, and More appeared first on Jon Loomer Digital.

Last week, Meta announced AI Connectors, which allow advertisers to manage campaigns directly through their existing AI tools, using Meta’s ads MCP server. At the moment, AI Connectors work for Claude, ChatGPT, and Perplexity.

Today, let’s focus specifically on using an AI Connector for Claude. In this post, I’ll provide details on how to set up the connector in Claude to get started before discussing some specific use cases.

A couple of notes before we proceed:

First, since Meta announced AI Connectors as an open beta, not all advertisers currently have access. You may be able to complete the steps below, but getting data from the connection may not yet be available to you.

Second, there are very real risks involved, including the possibility that connecting AI tools to your account could draw the wrong kind of Meta attention. You should use caution, and I’ll close this post by reiterating those potential risks.

Let’s get to it…

Connect to Claude

You don’t need a premium Claude plan to set this up. Custom connectors are available on every plan, including the free tier. Free users, though, are limited to one custom connector at a time. If you’re on a Team or Enterprise plan, your organization Owner needs to add the connector first before you can connect to it.

While logged into Claude, click “Customize” in the left menu.

Click “Connectors.”

Click the “+” to add a new connector and select “Add custom connector.”

Name your custom connector and provide Meta’s remote MCP server URL (https://mcp.facebook.com/ads). Then click “Add.”

Your custom connector will be added to the list.

But you won’t be connected yet.

Click “Connect.”

You’ll then be taken through Facebook Login and a series of prompts to select the business portfolios that you want connected to Claude.

When you start a new chat, click “+” > Connectors and make sure that the connector you just set up is toggled on.

Tools Available

The capabilities of this connector are driven by the tools available to it. Once you’ve completed the connection, there’s a list of the tools Meta allows Claude to use. I’m currently seeing 29 of these tools.

I won’t list them all, but here are a few examples:

ads_get_ad_entities — Pulls performance data at the account, campaign, ad set, or ad level for any time period, with filters and breakdowns.

ads_insights_advertiser_context — Provides a high-level overview of your business and marketing funnel to give Claude context for the rest of the conversation.

ads_insights_performance_trend — Analyzes the direction of your key metrics over time.

ads_insights_anomaly_signal — Surfaces unusual patterns or deviations in performance that may warrant investigation.

ads_insights_industry_benchmark — Compares your ad set performance against aggregated benchmarks from similar advertisers.

ads_insights_auction_ranking_benchmarks — Highlights which of your ads are winning the auction and which are getting under-delivered.

ads_get_errors — Surfaces delivery-blocking errors at the campaign, ad set, or ad level.

Use Cases

All interesting, but what can you actually do with this? What makes it worth using? I’d understand if this remains unclear to you.

Whenever you’re unsure of the best ways to use Claude for something like this, there’s no better way to uncover what’s possible than by simply asking it. And so I asked Claude for a handful of ways this AI Connector could be used, including a specific example for each.

Here’s what Claude returned (with some light edits)…

1. Account Exploration and Reporting

This is the most obvious starting point. You can ask Claude to pull performance data for any time period, at any level (account, campaign, ad set, or individual ad). Sort by spend, ROAS, CTR, frequency, or any other metric. Filter by status, objective, or thresholds you set. Apply breakdowns by age, gender, placement, country, device, and more.

Where this differs from Ads Manager is the back-and-forth. Ads Manager is fast if you already know which columns and filters you want. The connector replaces the moments where you’d normally export to a spreadsheet and build a pivot table to answer a slightly more nuanced question. Ask questions using natural language to get quick answers.

Specific example: Pull your top 10 ad sets by ROAS over the last 30 days, but only ones with at least 50 conversions, broken out by placement.

2. Diagnostic Work

Use this connector for troubleshooting. Claude can surface anomalies like sudden CPM spikes, frequency creep, and conversion drops that you might not normally catch. It can analyze auction performance to identify which of your ads are winning or losing in delivery against competing advertisers. It can also pull benchmarks so you can determine whether your performance is in line with similar advertisers in your category.

Specific example: Ask Claude what’s unusual about your account this week, then dig into any anomalies it finds.

3. Pre-Flight Error Checking

Meta surfaces delivery-blocking errors across campaigns, ad sets, and ads — the kind of red-flag issues that prevent things from running. The connector lets you ask for a sweep across your entire account in one prompt instead of clicking into each entity to check. This only catches configuration errors, not policy rejections or account-level restrictions.

Specific example: Ask Claude to surface any delivery-blocking errors across all active campaigns and ad sets.

4. Signal and Account Health

Three tools cover dataset and pixel health: configuration metadata, signal quality scores, and event volume statistics. This is the audit work that’s normally tedious, like checking whether your pixel is firing properly, whether match quality is good, or whether events are coming through at the volume you’d expect. This pairs well with Conversions API setup as a way to verify it’s working.

Specific example: Ask Claude to review your pixel health and event match quality, then summarize anything that looks off.

5. Campaign and Ad Creation

Claude can create campaigns, ad sets, and ads, all of which are created in PAUSED state by default. Nothing goes live without you flipping the switch in Ads Manager. It can also update existing entities (budgets, names, targeting) and activate or pause things on request.

Specific example: Ask Claude to duplicate your best-performing ad set with three new creative variations, all paused for review.

6. Catalog Creation and Management

If you run ads for e-commerce, the connector can create new product catalogs, browse and search products and product sets, and pull diagnostic information on data feeds and item visibility issues. It can also surface details on individual products, feed configuration, and ingestion rules.

Specific example: Ask Claude to identify any products in your catalog with feed errors or visibility issues, then summarize what’s causing them.

The Risks

The risks are real, and they start with whether you should set up this AI Connector in the first place.

It was only about two months ago when advertisers started reporting unexplained account shutdowns in response to unapproved integrations between Meta and AI tools. And now Meta’s inviting us to do it?

Granted, this shouldn’t be a trap. The account shutdowns were likely in error (though I haven’t heard what’s become of those accounts), and Meta has now announced these official integrations. We should be safe now, but I’m just as nervous as anyone.

There are other potential risks related to automations. New campaigns, ad sets, and ads created through the connector are put in a paused state by default, and you’d need to activate them from Ads Manager. Any tool that has full read and write access should make you nervous. How much damage can be done from a bug?

Another potential risk is the connector inherits all your Facebook user’s account access. Your business portfolio likely has access to ad accounts given to you by clients. That could generate additional business risks.

And finally, there’s a risk related to expectations. This is amazingly powerful and impressive technology. There are countless possibilities of what you could do with it. Using it for reporting, troubleshooting, and generally uncovering information is low risk. It gets riskier once you think that automating your processes, ad creation, and edits with little oversight means results will definitely be better.

The potential is huge. But, a whole lot of advertisers are learning some valuable lessons about AI. And those lessons can be expensive.

Some of these possible risks are related to unknowns because this is new technology. Some of this will likely be clarified shortly. Use caution. Be smart. And if it means avoiding the implementation of AI Connectors until you feel comfortable, wait.

FOMO is better than losing your shorts.

Your Turn

Have you set up a connection to your business portfolio using AI Connectors yet? How are you using it?

Let me know in the comments below!

The post Meta Ads AI Connectors and Claude: Setup, Uses, and Risks appeared first on Jon Loomer Digital.

Good riddance, Flexible Format!

Flexible Format is going away and being replaced by the new creative workflow.

This is a very good thing.

The Problems with Flexible Format

Look, I definitely used Flexible Format over the years, but it created so much frustration. You couldn’t customize creative by placement or alter the text by creative.

And most importantly, there was no transparency into how each creative variation performed. None of the breakdowns by creative worked.

New Creative Workflow as the Solution

The new creative workflow allows us to submit up to 10 images and videos for a single ad. Just like Flexible Format.

But unlike Flexible Format, this new workflow allows us to customize just about everything. Provide three aspect ratios by creative.

Customize the text and URL destination by creative.

Add or remove placements by creative.

And most importantly, there’s a new creative breakdown for this workflow.

It generates separate rows for each creative variation used. Not just the different images and videos, but the aspect ratio.

The new creative workflow is a huge step forward for advertisers, who won’t miss Flexible Format when it’s gone.

The post New Creative Workflow Replaces Flexible Format appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

Meta’s new AI connectors are the big story this week, especially after recent concerns about AI-related ad account shutdowns. We also got a rare drop in Meta app usage, more clarity on the one-click CAPI rollout, and a couple of helpful updates to the new creative workflow.

1. Meta announces AI connectors to manage ads
2. Meta forced to divest Manus?
3. Meta app usage falls for the first time
4. Updates to new Creative Workflow
5. How one-click CAPI works

Let’s get to it…

1. Meta Announces AI Connectors to Manage Ads

Talk about whiplash.

Less than two months ago, I documented how advertisers were reporting unexplained account bans due to presumably unapproved AI integrations. And then yesterday, Meta made a major announcement in the opposite direction.

Meta has launched AI connectors that let advertisers manage campaigns directly through their existing AI tools, without needing developer credentials or API setup. Built on Meta’s ads MCP server, the connectors provide authenticated access to campaign data, ad creation, catalog management, and audience insights.

The remote MCP server URL for all currently supported AI agents is https://mcp.facebook.com/ads. Meta links to instructions for setting up with ChatGPT and Claude.

I don’t know, man. Sounds great. But just a few weeks ago, I was terrified of attempting any of these integrations, for fear of losing my ad account forever (what happened to those who were banned, by the way?). And now I’m just supposed to trust that all’s good now?

I’ll let others try it first. Report back.

2. Meta Forced to Divest Manus?

In potentially related news, The Wall Street Journal reports that Meta is preparing to undo its Manus acquisition after it was blocked by China. Since Manus was founded in China, Meta may have little recourse.

All indications are that Meta is planning to divest. What that means for the integrations of Manus that Meta has already started rolling out is unknown. But it would make sense that this uncertainty led to Meta’s apparent shift related to other AI integrations via the MCP.

3. Meta App Usage Falls for the First Time

Well, this is different.

Meta released 2026 Q1 earnings, and the family daily active people dropped for the first time.

While the drop from 3.58 to 3.56 billion may not seem like much, any decrease in users across the family of apps is notable. Meta suggests that this is an anomaly due to world events:

The slight quarter-over-quarter decline in DAP in the first quarter of 2026 was driven by internet disruptions in Iran, as well as a restriction on access to WhatsApp in Russia.

Sure, current events in Iran and restrictions on WhatsApp are certainly challenging. But Meta has been confronted with difficult circumstances in the past, and they’ve never prevented user growth. Until now.

Since Meta consolidates usage data under the entire family of apps, it’s not clear where the biggest hits are coming from. But it’s especially interesting given the continued growth of Threads.

Of course, you shouldn’t cry for Meta. This drop didn’t seem to have any noticeable impact on ad revenues, which were up 33% year over year.

This was driven by both an increase in ad impressions and price per ad. Despite the drop in usage, ad impressions were up 19% worldwide.

And a 12% increase in price per ad.

So, revenue-wise, Meta is in no immediate danger. And a drop was inevitable eventually. But this is certainly a test to keep an eye on for future quarters.

4. Updates to New Creative Workflow

The test of Meta’s new creative workflow isn’t available to everyone, but there have already been a couple of big improvements.

One of my early complaints was that you couldn’t customize video thumbnail images. This is a major problem with flexible format, and it was something that needed to be corrected for the new workflow to be useful. Well, that update was made.

Now when you provide a video, you’re able to customize the thumbnail image.

The only remaining issue here is that it’s one thumbnail, regardless of the aspect ratio. That could result in some awkward crops.

Speaking of crops, another request I had was to simplify the process of providing the three crops for an image or video. Meta’s now made it so all three versions appear in the same view, which is a minor improvement.

Maybe I’m greedy, but can’t Meta use AI to scan these images and suggest initial crops? Then we can adjust them if necessary. Is that asking for too much?

Overall, I’ve been really happy with the new creative workflow. In the past, I was creating multiple ads for different formats and creative variations, but it often felt like a duplication of effort. Now I can consolidate ads and have all creative variations contribute to the same ad.

5. How One-Click Conversions API Works

Indications are that about half of advertisers now have access to Meta’s no-cost, one-click Conversions API option (I’m part of the unlucky 50%). So, what is it?

One popular method for setting up the Conversions API for web events is the API Gateway. I’ve used it for several years. When set up manually, it requires paying for an AWS server. This created opportunities for cheaper alternatives from companies like Stape.

At some point, Meta also made an API Gateway Demo option available. It uses the AWS server but is hosted by Meta. That demo lasted one year before costs would be incurred.

My understanding is that the new no-cost, one-click Conversions API version is essentially the API Gateway Demo, but without an expiration on the free offer. It’s also a single click, unlike the demo.

To find out if you have the new option, click “Connect Data” at the top left of Events Manager.

Select “Web.”

Select your dataset.

Google Tag Manager may be recommended. Click “Set up with Conversions API Gateway.”

If you have the update, you’ll see an option to “Set up with Meta.” I don’t have it, and I still see the demo.

Word is that if you sign up with the demo, you’ll eventually be upgraded to the no-cost, one-click option. If you trust that info, go with the demo. Otherwise, you can wait.

Is this right for you? There are certainly more sophisticated web API options that you may already be using. But if you’re otherwise paying for the API Gateway, you should check to see if you lose anything by switching.

More than anything, this could be a great option for small businesses that resist setting up CAPI due to technical challenges or costs.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Meta Ads AI Connectors, App Usage Drops, and More appeared first on Jon Loomer Digital.

While sales may be the ultimate goal, Meta ads for lead generation are often the best use of ad dollars for brands wanting to make a difference. That is, of course, as long as you prioritize building your list with quality leads.

The cost per lead tends to be a fraction of the expected cost for a purchase. While that cost will vary based on industry and offer, a steady focus on lead generation and list building can be an excellent long-term strategy.

Such a strategy doesn’t come without pitfalls, of course. While it is generally easy to tell when e-commerce ads are failing (they aren’t leading to sales), it tends to be far less obvious with lead generation.

This is what makes Meta advertising for lead generation such a science. You want leads, but not the cheapest leads. You want quality leads, but without jacking up your costs in the meantime.

An inexperienced advertiser is most likely to miss the signs of problems, and they’re also most likely to overlook the actual cause and misapply solutions. In this post, I hope to clarify what to prioritize and how to approach lead generation to get the most out of it.

Lead Costs

Do not chase the lowest cost per lead. The reality is that a lead might cost less than a dollar or it could cost hundreds of dollars. What matters is what that lead is worth to your business.

Lead generation will look different depending on the business. If you offer a professional service and a sales team will contact new leads in an attempt to convert them, it will be easy to assign lead value. What percentage of your leads convert? What’s the average revenue generated by those conversions?

It’s less obvious when building a list with lead magnets. You are offering something of value in exchange for joining your email list. The amount of lifetime revenue produced by a lead may not be known for months or even years. This makes the measurement of lead value challenging.

And some offers fall in between. The lead magnet may be a free webinar. On that webinar, there will be a sales pitch. What percentage of leads attended the webinar? What percentage made the eventual purchase? Of course, even in this case, the door may not be closed on sales following that pitch.

The main point is that you should do some work to determine what a lead is worth to you. Track these leads the best you can to see what they do after filling out your form. Try to apply an average lead value so that you have a goal for what a lead should cost you.

If the quality of those leads is an issue, you’re more likely to run into a problem with the costs not matching up with the returned value. But there are solutions you should consider.

Website vs. Instant Form

Should you run lead ads that utilize instant forms or send people to a landing page on your website? Is one more likely to result in low-quality leads?

The general assumption is that instant forms generate more and cheaper leads, but at a lower quality. This would make sense since they provide less friction and are easier to complete. But I wouldn’t assume it to be the case (my own tests do not reflect this quality drop-off).

Each approach has its own benefits. Instant forms are ideal when either website performance is a problem or you want to skip the step of designing a landing page. Website forms may be preferred when you want the traffic on your website.

The reality is that quality can be controlled from each one. Instant forms actually provide more built-in options for controlling quality that would require custom coding on a website. Don’t assume instant forms generically provide lower quality leads without trying these options.

My personal preference is website forms. The main reason is that my CRM will not accept opt-ins via third-party apps from people who previously opted out. Since my business is 15 years old, it’s a relatively common scenario. So I need an embedded form on my website to ensure they’ll get my emails.

The other reason I prefer website forms is control. I want people on my website. I want that traffic. I can also control the experience after completing the form and easily direct them to other resources.

Lead Quality Diagnosis

Maybe you’re lucky. Maybe the quality of your leads is amazing and you’re regularly able to get leads at a cost that is lower than their value. But even if this is the case, it’s quite possible that you can address inefficiencies to increase the value of your leads.

Evaluate whether there’s a common theme regarding the leads you get that makes them low quality. A failure to convert may not be a quality issue at all.

Are leads from a location you do not serve? Are they men when your customers are women? Over 65 when your product is irrelevant to people over 54? Not in the right industry? Unreachable by phone or email?

Use both internal and external data to diagnose. For example, you have information that Meta doesn’t from leads after they’ve completed the form. But you can also see how form completions and ad spend are distributed by age, gender, and placement by using breakdowns.

Make note of the similarities between the leads you’ve categorized as low quality. These are things you can address.

Location, Age, and Gender

There are ways to address quality issues related to location, age, and gender.

Location

Let’s assume you only serve people in the United States. If you get leads from people in Canada, the UK, or India, that’s a problem. If you’re not already targeting those countries, the problem isn’t easy to solve.

First, make sure the out-of-market leads are coming from your ads. If they are, Meta probably doesn’t know that they are out-of-market. They may be using a VPN, and there’s not much you can do.

If you can only serve people in a certain state or region, there are similar challenges, but also some things to look for. If you target people in the Denver area, you will automatically reach people living in and recently in that location. That will include people traveling there from other cities and states.

But you should also make sure that you haven’t checked this box when targeting regionally…

Meta will automatically expand targeting to reach people interested in this location.

And finally, consider value rules for location if you don’t want to eliminate certain countries, states, or cities entirely. If you have convincing data showing that the value of a lead varies by location, you can adjust bids accordingly.

Age

If you offer age-restricted goods, you should restrict by age accordingly. That can happen at the account level or in the ad set.

But there may also be a quality issue related to age.

I’ve found that age range can be an inefficiency related to lead generation. For whatever reason, people over the age of 65 often complete forms at a lower cost per lead. And as far as Meta knows, that’s a good thing.

But it isn’t necessarily if those cheap leads are low quality. While restricting by age may seem like the logical solution, it can backfire. You also may not want to restrict an age group completely.

Use breakdowns by age to see if you have a problem related to age distribution in the first place. And make sure you have data on the back end that proves these leads are lower quality. If they are, consider applying a value rule by age range.

Gender

Even if you serve predominantly men or women, don’t assume that you need to restrict by gender. Use the breakdown by gender to see how results and ad spend are distributed. Also verify how lead value differs by gender before acting.

Avoid restricting by gender if you can. If there’s a problem to be solved, prioritize using a value rule to adjust bids by gender.

Form Quality Control Options

Go back to your analysis of lead quality that we discussed near the beginning. What are the common themes?

The reality is that we want to limit unnecessary friction generated with our lead forms. The reason for adding friction would be to improve lead quality.

The fewer the questions you ask, the more leads you’re likely to get. If you only ask for a first name and email address, it will likely cost less than if you also ask for last name and a phone number. So if you ask for that additional information, make sure that there’s an exchange of value.

Here are a few ways to address lead quality with your forms…

• Ask qualifying questions
• Add SMS verification for phone calls
• Use conditional logic to filter out bad leads
• Require a work email if you are B2B

You can also get creative with website forms, though it may require some programming. If people answer questions in ways that are disqualifying, you can have the form completion redirect to a different confirmation page. That way, you can have the lead event only fire for qualified leads.

Lead Magnet Options

Maybe this is obvious, but the lead magnet you offer will almost always impact the quality of your leads. It needs to be desirable, but it should also be highly relevant to your business offerings.

This is why many contests and giveaways miss the mark. A business may offer an iPad or some other irrelevant, but desirable, product in exchange for contact information. But if that iPad has nothing to do with the business services, the lead is likely to be worthless.

Maybe you can offer the first lesson of a 10-lesson course for free. Or the first chapter of a book. The more obviously linked the upsell is to the lead magnet, the higher the success rate is likely to be.

Track and Optimize for Qualified Actions

What action can a new lead take that proves they are a quality lead? While this can be a purchase, such an action may not always happen right away. You don’t need to start that far down the funnel.

Something I track is whether a lead clicks a link in my emails. It doesn’t have to be the first email. A new lead will have multiple opportunities to open my emails during the first month. If they don’t click a link in any of them, they aren’t a qualified lead.

This is especially easy when the link is to access the lead magnet. While I may not know the value of a lead for months or years, I can at least make some assumptions based on those initial actions.

I created CRM automations and apply tags when new leads click a link within the first 30 days. If they do, they are marked as a “Quality Lead” and an event is sent to Meta. That allows me to add a column to my reporting to see how many of the people subscribing are quality leads.

This also gives me the option of optimizing for the Quality Lead custom event, rather than the lead itself.

This is something I experiment with often. But it doesn’t always improve the ultimate cost per quality lead.

Lead Outreach Process

Finally, the quality of your leads may not be the problem at all. Maybe you pass the leads on to a sales team, and they’re telling you that they are garbage.

But maybe it’s the outreach process that is the problem.

What automations do you have in place once someone completes your form? Are they emailed immediately? Texted to let them know you’ll be calling? How long until they’re contacted?

How is the sales script? What’s the quality of the salespeople? Is the problem actually with those who are tasked to close the sale?

It’s easy to label leads as being low quality because they’re unreachable, but this often misses the reality of the challenges that exist today. There are many things you can do to address these challenges.

Your marketing emails may go to spam and “other” folders. Review your emails and sender to limit your risk of triggering these filters. Make it clear on the confirmation page that they may need to check spam and other folders for your important email.

Many people don’t even take phone calls from unknown senders. These calls are silenced and go straight to voicemail. Send an initial text that asks the lead to respond. Even better, have them add you to their contact list so that they’ll receive your call.

It’s a Science

Lead generation is a science, and how it’s approached will vary depending on the business. But there are some common themes:

1. Know the average value of a lead
2. Know what makes a lead low quality
3. Address the low quality factors the best you can
4. Only add friction when it’s in exchange for lead quality
5. Do not ignore the post-lead outreach

You may never come up with a “final” approach to lead generation. You’ll analyze, tweak, and adjust to try to maintain that perfect balance of cost and value.

And that’s part of the fun of it.

Your Turn

How do you approach lead generation?

Let me know in the comments below!

The post The Science of Lead Generation With Meta Ads appeared first on Jon Loomer Digital.

Advertisers know they need to focus on creative diversification. So they’ll prioritize using different formats and uniquely different visuals in their ads.

But until now, that has resulted in a problem, and it’s a problem that can be solved with the new creative workflow.

The Problem

In the past, if you wanted to create ads with uniquely different images or formats, you had to create different ads. When you did that, you got put in a weird place regarding text.

Do you really need unique text for different formats or visuals? In some cases, it feels unnecessary.

But if you want to truly lean into creative diversity, that’s what Meta would seem to want. Meta discourages creating different ads based on minor tweaks. So you’d either come up with five completely different primary text options and headlines for each ad or you’d stick with the same text anyway.

The New Creative Workflow as a Solution

The new creative workflow makes this so much easier when unique text isn’t necessary.

Just submit up to 10 images and videos for a single ad. You can leverage the same text for all of them, or you can customize the text by creative.

One of the first things I did when I received access to the new creative workflow was to consolidate many of my ads. I went from 36 ads to six.

While fewer ads isn’t necessarily better (just as more ads isn’t either), consolidating ads eliminates unnecessary repetition that can make delivery optimization less efficient.

The post Consolidate Ads with the New Creative Workflow appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

Meta made a couple of announcements this week that could have a real impact, especially if the one-click, no-cost Conversions API rollout works as advertised. There were also several smaller updates worth watching, from a pixel upgrade and deeper AI integration to more signs that Dynamic Creative is finally on its way out.

1. ‘Analyze’ with Meta AI Business Assistant
2. Thoughts on audience segments
3. One-click CAPI spotted
4. Upgraded Meta pixel spotted
5. Dynamic Creative is going away
6. AI Voiceover test enhancement

Let’s get to it…

1. ‘Analyze’ with Meta AI Business Assistant

Meta has started to further roll out the Meta AI Business Assistant, integrating it with Business Suite and Ads Manager.

Some have noticed it on the bottom left Menu in Ads Manager.

But others have seen a much deeper integration. For example, here’s an example within the Ads Manager interface where advertisers are prompted to “Analyze” their results.

When clicked, it suggests some specific ways that you can analyze your data (28-day performance, creative engagement, industry benchmarks, and top and bottom performers).

What’s beginning to get confusing is how Meta views the future of Business Assistant versus Manus. Are there plans to further integrate Manus into Ads Manager? What are the future plans of Business Assistant? How are they different?

It would seem that there’s a decent amount of overlap.

2. Thoughts on Audience Segments

I’m a big fan of audience segments because of the information they can reveal in reporting. But something I’ve mostly glossed over is considerations about whether audience segments are useful for more than reporting. Do they influence ad delivery?

Let’s back up for a moment. Audience segments allow you to define your existing customers and engaged audience using custom audiences in Advertising Settings.

Once defined, you can break down results of sales campaigns to see how ad spend and performance differed by engaged audience, existing customers, and new audience (people not in either group).

Discussion of whether your definition of audience segments impacts delivery could devolve into a bit of a chicken and an egg-type scenario. Meta shows that 20% of your budget was spent on reaching people in your audience segments. But does that mean Meta prioritized those people because they were defined as audience segments or is the breakdown simply revealing what would have happened regardless?

It’s difficult to prove. So when something is difficult to prove, I prioritize what Meta says and doesn’t say. There are two important elements from Meta’s documentation that inform my interpretation of how audience segments are treated.

First, Meta’s official documentation only refers to audience segments as a helpful aid for reporting:

You can use audience segments to enable breakdown reporting for your sales campaign reporting in Meta Ads Manager.

There’s nothing else I’ve found from Meta that suggests your definitions of audience segments will influence delivery. This certainly appears to be intentional.

The question is whether Meta needs your audience segments to know whom to prioritize. You need to look no further than Meta’s original definition of Advantage+ Audience to realize they don’t.

Meta’s been prioritizing those groups for years, and that definition suggests that it doesn’t even matter if you’ve created custom audiences. Algorithmic remarketing will be prioritized based on pixel and event data.

Anyway, this is simply something I’ve been thinking about lately. It’s something I’ve long assumed, but I decided to look more closely to see if Meta makes any suggestions anywhere that audience segments would have additional powers related to delivery.

That doesn’t appear to be the case, but it doesn’t make them any less valuable. I still find audience segments hugely valuable as a reporting tool.

3. One-Click Conversions API Spotted

I’ve seen various reports of the new no-cost, one-click Conversions API spotted in the wild. My understanding is that the rollout is 50% complete, so not all advertisers have it.

I’ve seen some examples of it in the Events Manager settings tab that work exactly as Meta claims. A simple button. One click, and Conversions API is connected. Unfortunately, not everyone has been so lucky.

Another version looks like this…

But it was grayed out and needed workarounds to get set up.

I saw this in my Events Manager overview, which made me think I was a click away…

It wasn’t one-click either. I was asked for a bunch of information that I didn’t have, and the additional instructions led me to a dead end. I thought it was because there was a conflict since I already had CAPI set up, so I cancelled my integration. And then the one-click option disappeared.

In other words, learn from my mistake. Don’t cancel your integration until you know you can get this set up successfully.

4. Upgraded Meta Pixel Spotted

Another update that Meta announced last week was an upgrade to the Meta pixel. Some advertisers are already seeing it.

Within the Events Manager settings tab, you may see this section for “Meta Pixel Settings.”

It says that Meta will use “AI to identify and send optimal website details that are most likely to maximize your ad performance.” It’s on by default, but it will not go into effect until May 15th.

Expand it to get website and product details. This is the information that Meta can uncover related to the page or products using AI.

I originally interpreted this as being a replacement for a product catalog when running Catalog Ads. Well, maybe, but not necessarily.

Event data that includes richer page and product details can support better optimization and reporting. This also supports product based ads, helping you build a product catalog if you do not have one yet, or improve the quality and freshness of an existing catalog.

Should you leave this on? We have a month to figure it out, but the risks remain unknown. Meta says that turning it on can help improve ad performance and result in a higher quality catalog. But without knowing the risks and how to avoid them, it’s tough to say what each business should do.

Read more about this update here.

5. Dynamic Creative Going Away

In what shouldn’t be a surprise, Dynamic Creative is finally going away. And Meta may mean it this time.

If you feel like you’ve heard this message before, I get it. Dynamic Creative was supposedly going away a few years ago, replaced by Flexible Format. But then Meta decided, for some unknown reason, to keep Dynamic Creative for some objectives and Flexible Format for others.

You also may have seen a similar message related to Flexible Format.

Neither feature will be relevant once the new creative workflow is available to everyone (assuming it will be soon). Like both features, the creative workflow allows you to include up to 10 images and videos for a single ad. But it’s far more customizable, and it offers breakdown reporting transparency.

I’m a big fan of the new workflow and find it’s a huge upgrade over the two older features.

6. AI Voiceover Test Enhancement

It’s been a while since Meta’s introduced any new test enhancements (you can spot these in Advertising Settings under “Creative Features”). But let’s welcome AI Voiceover.

Here’s how Meta defines it:

Add a voiceover to your image to help people understand your messages. We’ll automatically select a voice that fits your image. If your ad has music, the voiceover will be layered on top.

Here’s an example…

I’m not sure why Meta believes this voice “fits” my image, but it’s still interesting. It’s also not clear how or when the voiceover is played, or if it simply turns an image into a video.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Meta AI Business Assistant, Tracking Updates Spotted, and More appeared first on Jon Loomer Digital.

Some advertisers have access to a new set of value rule criteria: Audiences.

This is the most robust application of value rules yet. When creating a value rule for audiences, you can adjust your bid for people who belong to specific custom audiences. Bid more or less for people who are qualified leads, disqualified leads, high-value customers, low-value customers, and more.

While a sizable amount of effort is required to label these audiences, there could also be equal or greater payoff for specific types of businesses.

Let’s take a closer look…

General Overview

If you have access to audiences within value rules, you will see the following categories of options…

If you are seeing this option for the first time, it’s likely that all but “New Audiences” are grayed out.

New audiences: Audiences that are not included in your custom audiences.

Engaged audiences: Prospective customers who engaged with your business but haven’t converted yet. All available options are labels you added to custom audiences.

Engaged Audiences includes the following groups:

• Qualified leads: Leads that meet your qualification criteria
• Disqualified leads: Leads that don’t meet your qualification criteria
• App installers: People who have installed your app
• Trial users: People who started a trial of your product
• Cart abandoners: People who added a product to their cart but didn’t check out
• Other engaged users: People who showed interest but didn’t make a purchase

Customers: Audiences that have made a purchase from your business. All available options are labels you added to custom audiences.

The Customers category includes the following:

• High value: Customers you consider valuable to your business
• Low value: Customers who are of low or negative value to your business
• Recent purchasers: Customers who have made a purchase or converted recently
• At risk: Customers who are showing signs of disengaging or churning
• Disengaged: Customers who have not made a purchase recently or stopped subscribing
• Customers: People who have converted and become customers

Other audiences: Audiences who don’t fit into the other categories.

The Other Audiences category includes these final groups:

• Personas: People who fit your business’ customer personas
• Other 1: People who don’t fit into the other categories
• Other 2: People who don’t fit into the other categories
• Other 3: People who don’t fit into the other categories

How Are These Groups Defined?

When I first heard about value rules for audiences, I logically assumed that they’d leverage audience segments. These value rules specifically call out “customers” and “engaged audiences,” after all, which is how audience segments are defined.

But once I was granted access to this feature, I quickly discovered that this couldn’t be the case. Audience segments are broadly defined as engaged audiences and existing customers. Value rules rely on further segmenting those groups, in addition to defining an “other” category.

A couple of things jump out to make it clear that this isn’t about audience segments. First, the definitions all mention “custom audiences.” The New Audiences group is defined as those who aren’t in any of your custom audiences — not your audience segments, but any custom audience.

Second was the mention of “labels.” And that’s when I realized how this worked: Advertisers need to define these groups by adding labels to custom audiences.

What Are Labels?

Most advertisers have a basic version of audience labels for customer list custom audiences. When you create or edit a customer list custom audience, you will see the option to add a label.

But even in Meta’s documentation, this version of labels is only intended to categorize your custom audiences to make them easier to find. There’s no functional application. And the label options are limited to a handful: Qualified Leads, Disqualified Leads, Customers, High Value, and Low Value.

If you have this update, labels will extend to website custom audiences (and possibly others, though I’ve confirmed they don’t apply to video, page, and lead ad custom audiences). And most importantly, you’ll get to apply additional granularity to these labels.

These labels mostly match the options available for creating value rules by audience above. The main exception is “Restricted Audiences.” But this label is for people you can’t show ads to for legal or other reasons. There would be no reason to apply a value rule to this group since you’d want to exclude them entirely.

The other is “New Audiences,” which would imply a person not only doesn’t have a label, but isn’t in any of your custom audiences.

Labeling Your Custom Audiences

This all sounds interesting, but remember that some significant effort is required on your part. You now need to go through and label your custom audiences in order to use these options with value rules.

And that’s where things can get tricky. As long as you’ve determined that a website custom audience can be accurately defined with a label, that definition will be dynamic and update regularly.

But where you need to be careful is with customer list custom audiences. You need to be sure that they are updated regularly so that a person who has progressed from “low value” to “high value,” for example, is accurately moved from one label to the next. That will require technical effort and automation.

It’s also not clear what happens, if anything, if someone is given conflicting labels. While we know that Meta handles the overlap of engaged audiences and existing customer audience segments (Meta will only count a person as being an existing customer if found in both), there’s no clear indication that the same would apply here. And if someone is counted in two conflicting groups where you’re increasing and decreasing bids, that would obviously create a problem.

The Rule of Value Rules

While this all sounds amazing, there are three critical elements for assessing whether value rules for audiences would be worthwhile for you…

1. Use to solve a problem.

Possibly most importantly, I worry that advertisers will use value rules for audiences when they aren’t necessary. Advertisers will assume that they need to bid more on high-value customers or less on low-value customers. But Meta will be doing some of this automatically already.

The basic rule of value rules is to use them when you have information that Meta doesn’t and there’s a problem to be solved. A prime example is related to qualified and disqualified leads.

If you’re not getting the results you want and Meta is leaning heavily into remarketing to your leads, you may consider bidding more on qualified leads or less on disqualified leads. Since Meta doesn’t know that you value these people differently, they may both be targeted aggressively.

2. Data volume.

This just isn’t likely to be a small business solution. You need to have the volume of data required that would allow Meta to have an impactful audience of people to raise and lower bids on.

I’d consider my email list and website traffic to be well above average for most small to medium-sized businesses. But even I have doubts whether I have the volume that would make this worthwhile.

I could see potentially using this for the broader labels (all customers or qualified leads), but most of these refinements get too granular to be helpful.

3. Accuracy of labels.

As detailed above, you’ll need to dedicate a decent amount of effort to defining these audiences to make use of them. The accuracy of your work will have a significant impact on the output. That also includes dynamic updating and potentially the handling of conflicts.

Overall, I do think that the potential is significant for larger businesses with substantial data that can be parsed and segmented — especially when you can label information that gives Meta a better understanding of customer value. This could help Meta be more efficient in the delivery of your ads.

But truly, only in the cases where you are giving Meta information they don’t already have and there’s a problem to be solved.

How Would You Use This?

When I’ve spoken about this feature with others, I often need to add a clarifier regarding how and why you might use value rules for audiences. Otherwise, it’s easy to assume that this does something it doesn’t.

This isn’t about optimizing for higher value customers or qualified leads. It’s about spending more or less to reach the people who are already defined as that.

Now, there could be some spillover effect. If you bid less on disqualified leads, Meta will show your ads less to the people you’ve previously disqualified. By doing so, it’s reasonable to assume that this could also impact how much you reach people like them. Reasonable to assume, but all theory, of course.

With that out of the way, let’s think through ways you might use the new value rules for audiences…

1. Bid less on customers.

Let’s say you’ve determined that you want to lean more into prospecting. This could be because of company goals or you’ve seen that new customers are more valuable than existing customers. You could accomplish this without excluding current customers.

2. Bid less on disqualified leads.

Because they’re on your email list and have visited your website, Meta’s likely to keep targeting them. But if you’ve already disqualified them based on their answers on a form, you can bid less to reach them less often.

3. Bid more on cart abandoners.

You have a limited time offer that’s available to all new and existing customers. Bidding more on cart abandoners could make sense to make sure they know of this offer since they were seemingly a step away from completing the purchase. That assumes, of course, that Meta wouldn’t prioritize them already.

4. Bid more on high value customers.

You have a new high-ticket offer to announce, and you are going to target people broadly. But you want Meta to bid more on those who have proven to be loyal, high-value lifetime customers. Since lifetime value may be data that is hidden from Meta, it can help focus on what is otherwise low-hanging fruit.

Your Turn

Do you have access to value rules by audiences? How would you use them?

Let me know in the comments below!

The post Meta Value Rules for Audiences Explained appeared first on Jon Loomer Digital.

This could be a big deal.

Meta made a couple of announcements about upgrades to tracking recently, and one really got my attention.

The first was an upgraded Meta pixel that will use AI to automatically pull in page and product information. This should especially help e-commerce companies that don’t want to fumble with product catalogs.

That’s an interesting update, but this one is even better…

Conversions API Update

Meta announced a no-cost, one-click Conversions API solution. That’s right, free. And easy. Conversion API setup tends to be anything but free and easy.

Meta says it will require “no technical expertise, no costs, no ongoing maintenance, and will allow businesses to get set up in minutes.”

Why This Matters

The Conversions API for web makes a difference. Meta says that it results in an average 17.8% lower cost per result.

But most small businesses don’t bother due to costs and potential technical complexities. If this is accurate, Meta’s new solution could change all of that.

It may also cut into the profits of third-party apps that have provided API integration at a cost. It’s not clear what limitations, if any, this free solution will have.

This is a development that’s worth watching. I don’t believe it is available yet, but it could be a helpful option, depending on your current setup.

The post No-Cost, One-Click Conversions API Solution appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

Meta made a couple of announcements this week that could have a real impact, especially if the new one-click, no-cost Conversions API is as simple as it sounds. Beyond that, there were several smaller updates that are easy to overlook now, but could matter later.

1. No-cost, one-click Conversions API
2. Meta pixel upgrade adds AI-powered data
3. Expanded custom audience labels and value rules
4. What are Dataset Attributes?
5. Live video ads
6. New creative workflow helps consolidate ads

Let’s get to it…

1. No-Cost, One-Click Conversions API

Meta made a couple of announcements recently, and this one could be the most impactful.

Meta announced a simpler way to set up Conversions API for web:

…we’re also making it significantly easier to set up the Conversions API for web, with the introduction of a Meta-enabled Conversions API setup – a one-click option that requires no technical expertise, no costs, no ongoing maintenance, and will allow businesses to get set up in minutes.

This could be a really big deal. Meta has pushed the benefits of the Conversions API for years (this announcement mentions a 17.8% lower cost when implemented), but technical hurdles and costs have always been obstacles, particularly for smaller businesses. Meta doesn’t provide details of this solution, but they used all of the magic words: “one-click,” “no technical expertise,” “no ongoing maintenance,” “set up in minutes,” and “NO COSTS.”

No costs? Really? In addition to being easy to set up? Sign me up.

The main question will be what downsides, if any, there are related to this new option. Otherwise, third-party vendors that generate revenue from these integrations could see a noticeable hit to their profits.

2. Meta Pixel Upgrade Adds AI-Powered Data

Meta actually led that announcement with what I believe is a less interesting update, though that may depend on your needs: An upgraded Meta pixel.

We’re introducing a new Meta Pixel feature that allows businesses to use AI to automatically include additional page and product information – such as product names, availability, and business details – with the events you share with Meta.

If you run ads for an e-commerce company, you surely know all about the importance of product catalogs. Advertisers can run catalog ads that dynamically pull product information from a feed provided by the company. The accuracy of that feed, of course, is critical.

Will this new automated way be more accurate? Will it be as accurate? Without more details, it’s difficult to grasp how exactly this will work and what potential downfalls there may be.

Apparently, this change will be forced on all advertisers in the future, though it can be turned off. From Meta:

Existing Meta Pixel users will receive a notification with a 30-day window to review the feature before it’s enabled. Advertisers can adjust or turn it off at any time through Events Manager, and they can also review and manage which specific data categories are shared so that they can optimize best for their business and compliance needs.

How does this impact advertisers who don’t already utilize a product catalog? Will it be a good thing or a detriment?

Look for this addition soon in your Events Manager.

3. Expanded Custom Audience Labels and Value Rules

Some advertisers have the option to add labels to custom audiences. Up until now, labels have only been for customer list custom audiences, and options included the following:

• Qualified Leads
• Disqualified Leads
• Customers
• High Value Customers
• Low Value Customers

I’m part of a test that expands these labels. Not only are there more options, but they also apply to website custom audiences.

There are now four categories of labels: Engaged Audiences, Customers, Restricted Audiences, and Other Audiences.

Engaged Audiences include labels for Qualified Leads, Disqualified Leads, App Installers, Trial Users, Cart Abandoners, and Other Engaged Users.

Customers include High Value, Low Value, Recent Purchasers, At Risk, Disengaged, and All Customers.

Restricted Audiences would only be those you can’t legally serve ads to.

And then Other includes Personas and three “Other” groups.

These labels don’t seem to have anything to do with Audience Segments. I’ve added labels and they did not impact the definitions of my Engaged Audience and Existing Customers in Advertising Settings.

So, what are labels for? That’s where the fun stuff starts. If you have these expanded labels, you probably also have a new value rule for Audiences.

These value rules then allow you to bid more or less based on the label.

This is something I’m currently digging into, and I’ll be covering it on my blog shortly. Stay tuned!

4. What Are Dataset Attributes?

Earlier this week, a field for Dataset Attributes appeared in my ad sets when using the Website conversion location and optimizing for conversions.

The tooltip says that this allows you to “track and measure the performance of the ad delivery for prioritized attributes from your dataset.”

What does that mean, exactly? The dropdown menu to select said attributes was empty. And based on every example I saw, others ran into the same thing.

Might this be related to event parameters? So, instead of focusing only on a purchase, it’s the purchase of a specific product, product category, price, or something else? While you could do this with custom conversions, Dataset Attributes would make that process easier so that custom conversions wouldn’t be required.

I’m only speculating since I can’t find any documentation and the feature has since disappeared. But I’m sure we’ll see this again soon.

5. Create Live Video Ads

If you host live Videos to your Facebook page, you can run Live Video Ads that promote your video while it broadcasts or after it ends.

To run Live Video Ads, you first need to have created a live video on your page and satisfy eligibility requirements. Then you may see the following option when in the ad set…

While similar to a regular video ad, Live Video Ads will highlight a “Live” icon while the video is being broadcast.

I don’t actively host live videos on my Facebook page, so I don’t have anything to share from personal experience. But you can learn more about how to set these up here.

6. New Creative Workflow Helps Consolidate Ads

I’m part of the test for Meta’s new creative workflow, and I think it will solve a lot of structural problems. The biggest opportunity is that it can help consolidate ads that may have otherwise resulted in inefficiencies. Let me explain…

Advertisers have become hyper-focused on creating diverse ads to lean into the power of Andromeda. That includes leveraging different formats. While prioritizing diverse formats is a solid goal, the old workflow presented some obstacles.

Single image or video ads gave advertisers the ability to build ads around one image or video. If you wanted to utilize both an image and video, you generally had to create separate ads. You could try to work around this by creating one ad, but you’d need to customize it to showcase an image in some placements and a video in others.

Neither option was ideal. If you created separate ads, would you utilize the same primary text and headlines in each ad? That conflicts with creative diversification, but it may also feel unnecessary and burdensome to create five primary text and headline options for an image and a completely different set for a video when the same text could have worked for both.

And customizing by placement for one ad wasn’t the best option either. You may assume that certain placements may be better for images and videos, but we’re not giving Meta much flexibility.

What’s funny is that Meta has been taunting us with a solution for a while. Related Media was the precursor to the new creative workflow. It allowed us to add multiple images and videos to a single ad.

The main problem was that we were limited to the creative Meta recommended. And the customization of that Related Media is nonexistent.

The new creative workflow is exactly what I’ve been begging for. I want to create one ad that gives Meta multiple options of images and videos that could work with a specific set of text options. Then Meta can figure out which format and version would work for each person in a particular placement.

I’ve already consolidated my ads quite drastically. I’ve combined 36 ads from a single ad set into six.

And that includes two carousels that are different only by color. If I could combine them, I would (I’ll likely eventually turn one off as I see more results).

At minimum, this is much cleaner from an organizational standpoint. Will it lead to better results? That’s the part we’ll find out next.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post One-Click Conversions API, Pixel Updates, and More appeared first on Jon Loomer Digital.

Some advertisers have access to a new creative workflow that allows them to upload up to 10 images and videos for a single ad. I am part of this test, and I believe it will be a helpful improvement.

This workflow allows advertisers to do the following:

• Upload up to 10 images and videos for a single ad
• Provide three cropped aspect ratios for each creative
• Customize text and URL destination by creative
• Opt in or out of placements by creative variation
• View performance by creative asset

In this post, we’ll walk through the following:

1. Requirements
2. Initial Setup
3. Customizations
4. Switch Back to Original Workflow
5. View Performance
6. How It’s Different
7. How I’m Using It
8. What Could Be Better?

Let’s get to it…

1. Requirements

First, know that not all advertisers have access to this test. If you have the new creative workflow, it will apply to all new and duplicated ads, not to existing ads. You’ll see this for image and video creatives only, not for Advantage+ catalog ads or carousels.

While this workflow applies to all campaign objectives, there are a few exceptions by conversion location…

• Awareness: All conversion locations
• Traffic: All except Facebook or Instagram
• Engagement: All except Facebook Page
• Leads: All except Website + Instant Forms or Website + Calls
• App Promotion: All
• Sales: All except Website + Calls or Website + In Store

2. Initial Setup

When you create a new ad following the requirements above, click to create an image or video ad.

Complete the Creative Setup step as you normally would and click “Next.”

The screen to add or upload images and videos will look mostly the same as the original workflow.

Normally, you can only upload or select one image or video. But now you can select up to 10. You can toggle between the Videos and Images filters to select from both.

When you do, those selected creatives will appear at the bottom.

If you hover over a thumbnail, you can remove or crop the creative.

Let’s assume you didn’t click the crop icon. We can still do that later. Click “Next” and then complete the steps in the creative setup wizard for text, image generation, and enhancements as you normally would.

3. Customizations

Once you complete the wizard, here’s how the Media section looks…

Recall that normally your creative would be displayed in placement groups, and you could customize it for each individual placement. But now we can customize all of the creative we provide.

Click “Customize Media.” It will look like this…

Using the old workflow, Meta normally asks for square, vertical, and horizontal versions of an individual creative. Now you can customize this for each image or video that you provide.

You can choose to keep the original or crop for each selected size.

If you choose to crop, Meta will automatically adjust the crop tool to fit the aspect ratio that is requested. You can then move or shrink the tool to focus on the area you want to represent the cropped version.

Once you’re done, click on Placements. Here, you can opt in or out of a placement for the chosen creative version.

Note that Text and Destination are grayed out when the first creative is selected. This version will automatically use the text and URL destination provided for your ad.

But if you select any of the other creatives, you can edit these things. If you don’t make any changes, Meta will use the text and destination from your original setup.

But this is where you could use very different text depending on the image or video. One drawback is that you don’t have the five options that you have for the first creative option.

You can also customize the destination URL by creative. One obvious use case for this would be if you want to use different images to promote different products. But you could also send people to different landing pages, depending on the image. I’m doing this so that my ad creative matches the creative on the landing page.

Otherwise, finish setting up your ad the way you normally would with Format Display Options, tracking, and URL parameters. Then preview and publish.

4. Switch Back to Original Workflow

I realize some people hate change, and this is likely to be confusing the first time you go through it. If you’d like, you can revert back to the original workflow.

Of course, it’s not easy to find. Hover over “Media” and then you’ll see a link to “Return to original single media experience” at the bottom of the tooltip.

5. View Performance

One very big addition to this new creative workflow is a way to break down results by creative asset. When live, you’ll see an option for “Media” under the Creative breakdown.

I didn’t have this option until I published my first ad with the new workflow.

6. How It’s Different

Maybe you’re thinking, “Oh, great. I can add up to 10 images and videos for a single ad. But can’t I do that with Dynamic Creative and Flexible Format?”

Yeah, you can. But whether you can use Dynamic Creative or Flexible Format will depend on the objective. And based on reports I’m seeing, Flexible Format is going away.

Another big element is the level of customization. With the new creative workflow, you can do the following with up to 10 images and videos:

1. Crop three sizes for each
2. Customize primary text, headline, and description for each
3. Customize URL destination for each
4. Opt into or out of placements for each

You are now able to do things that would have required creating multiple ads in the past.

But the biggest benefit of this new workflow over Flexible Format is the reporting granularity. Advertisers have complained about the lack of this element ever since Flexible Format became available.

7. How I’m Using It

It’s funny how this became available to me only a few days after I had a specific need for it. I immediately applied it to make my current ad setup more efficient.

I previously created six separate ads to promote my Cornerstone Advertising Tips lead magnet. Each ad used a different image and sent people to a different landing page that featured an image that matched the one they’d see in the ad. While each ad used slightly different text, they certainly could have been the same.

So I consolidated this approach with the new creative workflow. One ad with the same images I was using for those six separate ads. They all share the same five primary text and headline options, but I’ve customized the destination URL for each one.

The screenshots in this blog post are actually from the process I went through to create it.

Ultimately, this new workflow gives advertisers the tools to create ads that are truly in the spirit of creative diversification. In the past, you would have created separate ads for images and videos, but now you can combine them into one without having to choose specific placements for each format.

8. What Could Be Better?

Early on, my complaints are minimal. I like this a lot, and more issues are likely to pop up as I use it. But I have a couple of specific problems with it.

First, you can provide three crops for each image or video. But this is more limited than it needs to be.

For example, Meta says it supports either 5:4 or 1:1 for the square crop. Is this right? I often use 4:5 for Feeds, not 5:4. But regardless, you can only crop by 1:1.

Same thing for vertical. Meta says that 4:5, 3:4, 2:3, and 9:16 are all supported. But you can only crop to 9:16.

And, of course, the horizontal crop. Meta says that 3:2, 5:4, 1.91:1, and 16:9 are all supported. But you can only crop to 1.91:1 (I think?).

Why not offer the ability to adjust the crop to different aspect ratios if they qualify?

My only other request is related to ways to make this process faster and easier. If you have 10 creatives, you have to go into each image individually to crop each one.

I’m no designer, so I know that this has to be a challenging problem. I have no idea if an alternative that would allow you to crop multiple images at once is realistic. But that feels like a potential bottleneck.

Your Turn

Do you have the new creative workflow? What do you think?

Let me know in the comments below!

The post New Creative Workflow for Meta Ads appeared first on Jon Loomer Digital.

This is a common mistake advertisers make…

The reporting breakdown feature allows advertisers to break down results by things like age, gender, placement, platform, location, and more.

It’s incredibly valuable, but most advertisers incorrectly apply what they learn from these results.

The Mistake

They’ll see cheap results based on age, and they’ll remove other ages. They’ll see fewer results from stories and reels, so they’ll remove those placements. They think that this will improve results, but it almost never does.

This is an old school “optimization” approach, which assumes that we can manually trim away waste to force Meta to spend more efficiently. It may have worked at one time, but those days are in the past.

Meta wants you to get more results based on your performance goal. You don’t need to remove ages or placements to get more results.

A Troubleshooting Tool

Instead, breakdowns should be used as a troubleshooting tool. This is where they’re most valuable.

Because Meta wants you to get more results, they’ll exploit weaknesses related to your performance goal. And this is where you can uncover a high number of cheap, low-quality results by demographic group or placement.

Thanks to breakdowns, you’ll be able to detect a concentration of ad spend on the audience network because it’s leading to a high number of cheap clicks.

Or on people aged 65 and up, because they’re leading to a high number of cheap leads.

I also recently documented how I used breakdowns to troubleshoot a weird drop in performance. Read about this weird problem and how I solved it here.

The post Use Breakdowns for Troubleshooting appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A handful of smaller updates caught my attention this week, but several of them could have a meaningful impact on how you approach delivery, tracking, and creative. From new value rules for audiences to Flexible Format going away and an updated Meta pixel GTM template, there’s plenty here that’s worth a closer look.

1. New value rules for audiences
2. Flexible Format is going away
3. Updated Meta pixel GTM template
4. How I’m using AI to create videos
5. Ad posts moving from Page Posts
6. Ad performance troubleshooting

Let’s get to it…

1. New Value Rules for Audiences

Some advertisers are seeing a new option to create value rules by Audiences.

I’ve seen other versions of this related to customer value. It’s not clear what’s a test and what is an official rollout. I have neither. But let’s attempt to make sense of the screenshot above.

In this example, there are four initial options:

New Audiences: Audiences that are not included in your custom audiences.

Engaged Audiences: Prospective customers who engaged with your business but haven’t converted yet. All available options are labels you added to custom audiences.

Customers: Audiences that have made a purchase from your business. All available options are labels you added to custom audiences.

Other Audiences: Audiences who don’t fit into the other categories.

Since only the New Audiences option is available here and the others are grayed out, I’m not able to tell you what happens if you select the other three. But let’s try to make sense of what we’re seeing here.

On the one hand, it would make a whole lot of sense if these were based on audience segments for engaged audience and existing customers, as defined in Advertising Settings. These definitions reference engaged audiences and customers, and those who aren’t in either segment are considered “new audiences.”

But there are a couple of other things here that caught my attention.

1. Custom Audiences. Yes, audience segments are defined with custom audiences. But these definitions seem to imply that if you aren’t in any custom audience, you fall under “New Audiences.”

2. Labels. The definitions for both engaged audiences and customers refer to “labels” that were added to custom audiences. This sounds familiar, but I don’t believe I have this.

So, my interpretation is that advertisers will need to somehow label custom audiences that reflect either the “Engaged Audiences” or “Customers” group to then bid differently on those people. Anyone in a custom audience that hasn’t been labeled will be considered “Other.” And anyone not in any custom audiences at all will be “New.”

Until I have any of this functionality, it’s only an educated guess. But it feels like a lot of moving parts that advertisers need to take care of to do this right. And I worry that some will micromanage delivery when it’s not necessary.

Actually, I don’t worry about that. I know that will happen.

2. Flexible Format is Going Away

An alert at the top of Meta’s official documentation on Flexible Format indicates that it’s going away.

Or, more literally, it will “no longer be available in Ad Setup.” So I guess it’s possible that it would still be available in another way. It also appears it’s already started going away for some people, which is consistent with some of the confused questions I’ve received.

Of course, I still have Flexible Format…

But we also know that format selection is changing, and Format Display Options will be the way to choose formats.

It would be a whole lot easier if I had any of these changes to communicate them with you. But one of the screenshots I’ve seen suggests that you can add at least three images or videos for a given ad — and maybe it was for a single placement or placement group. This would be similar to how Flexible Format works, though it allows up to 10 creatives for an ad.

I don’t want to speculate too much at this point because it’s difficult without access to any of these changes. But maybe you have them?

3. Updated Meta Pixel GTM Template

Meta has updated the official Meta Pixel Google Tag Manager template to use for tracking.

This replaces the old Facebook Pixel template. One added benefit is GA4 integration. It will automatically reuse your existing GA4 dataLayer, meaning an event like purchase is mapped without manual configuration.

I admittedly need to switch to this. I’ve been using GTM to manage the pixel for so long that it was set up with manual tags, instead of a template. That surely means I’m missing out on some functionality.

4. How I’m Using AI to Create Videos

I have a complicated relationship with AI. I hate AI slop, but I like using AI to simplify my processes so that I can create more valuable stuff more quickly. I wanted to share one way I’ve started to use AI to help with videos that can be used in ads.

I’m sure I’m not alone in saying that video creation is a bottleneck. I like creating videos, but they take time. I currently publish a short-form video per week, but I’ve done as much as one per day. I’ve essentially reached peak efficiency there, and it still takes time.

Because of this, I pick my spots to create new videos. I rarely record videos to promote my new content. When I do create videos for my ads, they often fall into similar templates. They lack creativity and the styles are repetitive.

So, I was having this conversation with Claude (which has asked that I call it “Chip”), not thinking that it could help me in this area. It surprised me with a solution.

That solution was very basic to start. I’d give it details of a piece of content that I want to promote, complete with the title and full body. It would then suggest some video concepts. I’d pick one, and it would create an HTML video.

When I say it’s an HTML video, I mean it’s very basic. No animated characters or AI-generated people, characters, or scenes. Animated text, colorful backgrounds, and sound effects. More slideshow than video, if we’re to be technical.

That may sound boring to you, but that’s how I already create a lot of my videos now. What made this especially attractive was that I could generate these videos very quickly and I’d be far less limited in terms of style.

Most importantly, because of the nature of these simple videos, they don’t look like AI-generated videos. They don’t stand out as the typical slop that I hate so much.

But that initial version had issues. I could play the video, but there was no file or export. I had to screen record it, which created issues with aspect ratios and video quality.

And then we moved to Claude Code. I took everything we did with that initial version and had specs drawn up as a foundation for a new dashboard. After a day of iterations, I now have a dashboard that helps me quickly and easily generate new videos in 9×16, 4×5, and 1×1 aspect ratios.

At the top, I provide the title of the piece of content I’m promoting and the body of that content itself. This gives it something to review and generate a relevant video.

I worked out several styles and variations with Claude that fit formatting and dimensions that I wanted. So, I can either manually select a style and variation, or hit “Surprise Me” and it will choose one randomly.

It starts building the video for me…

It generates 9×16, 4×5, and 1×1 versions of a video.

I can chat with Claude to make refinements if necessary.

When I’ve approved the final output, I can export and then download the files.

Here’s an example of a video that was generated with this dashboard that promotes my Beginner’s Checklist for Meta Advertising blog post.

Is it rocket science? No. And I know there are all kinds of tools out there that you can pay for that do something similar.

I just find it super cool to be able to customize something to my exact needs. I’ll continue to refine it, and hopefully it will become a part of my process.

5. Ad Posts Moving From Page Posts

If you publish ads to your page, know that this process is moving from the Page Posts section to Meta Business Suite.

One of the reasons I did this before was that you couldn’t create videos organically that include a CTA button and send people to a page on your website. So I’d take an ad that does this and publish it to my page.

According to this message, this functionality should be available in Business Suite now. Unfortunately, I’m not seeing it, so I’ll continue to use Page Posts until that happens.

6. Ad Performance Troubleshooting

One of the challenges I have is that I’ve published endless content on every topic imaginable. If someone wants to know how I approach something, they can find it on my website. But some of it will be old and outdated, and there isn’t a single post that consolidates my current recommendations.

That’s why I developed Ad Briefs. Ad Briefs are summaries on all of the most important topics related to Meta advertising. They are based on my most recent content and are updated monthly to remain current and relevant.

There are currently 23 Ad Briefs on topics related to targeting, attribution, reporting, campaign construction, and a whole lot more. While Ad Briefs are only available to my Power Hitters Club – Elite members, I’ve created a sample ad brief on Ad Performance Troubleshooting that you can read for free.

Here are the key takeaways:

• Most performance drops have logical explanations: randomness, recent changes, competition, website issues, or event delays—investigate these first.
• Evaluate results in seven-day windows at least two weeks in; don’t overreact to daily fluctuations, especially at low volume.
• Dig beyond surface metrics—use attribution comparisons, audience segment breakdowns, and backend validation to understand what’s really happening.
• When ads consistently underperform, simplify campaign structure and focus on improving the ads, offer, and landing page rather than chasing tactical hacks.
• Early strong performance followed by decline is often natural audience exhaustion, not a sign something is broken.

Read the sample Ad Brief here.

If you want access to this and 22 other Ad Briefs, as well as other benefits like weekly strategy sessions, weekly webinars, a private community, and more access to me, go here to learn more.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post New Value Rules for Audiences, GTM Updates, and More appeared first on Jon Loomer Digital.

I’ve been doing this for a very long time. In August, this website will be 15 years old. I’ve been talking about and writing about and practicing Facebook-then-Meta advertising longer than almost any walking human.

But that also gives me lots of chances to be wrong. And I’ve been wrong a lot.

In some cases, I was right about something for a time. But as things changed, I clung too tightly to that old perspective. Correction was more difficult in some situations than others.

It’s why I’d argue that it’s easier to be a new advertiser than an experienced one. If you’re new, you don’t walk in with expectations about how things should work. But the experienced advertiser has years of “proof” that you should do things one way, and that other way is wrong.

Meta advertising is constantly changing. The pages of this website are proof of that. Hundreds and hundreds of pages of new features and best practices that may conflict with those discussed a few weeks, months, or years earlier.

Below are examples of when I was the most wrong about Meta advertising. This, of course, is not an exhaustive list.

1. Remarketing

The thing about remarketing is that I was absolutely right about it for a while. I was obsessed with it, and for good reason. Targeting was possibly the most important factor for getting results.

I was all over every type of remarketing. The minute we could create website custom audiences, it became almost all I did. I’d segment and micro-target and do some of the craziest stuff.

And it worked. Part of the reason it worked so well was that I was lucky enough to have large and engaged audiences to target. Facebook was still surfacing my content organically, leading to lots of social traffic to my website. Google loved me, resulting in plenty of referral traffic. And my email list was large and active, leading to a big chunk of email-driven traffic.

It’s not that I don’t still benefit from these things, but I took for granted how easy it all was back then. And, combined with the low costs, it made remarketing all the more effective.

My sweet spot was from 2015 to 2018. A large percentage of my strategic-focused content was about remarketing. When I spoke at a conference, I’d talk about remarketing. It became what I’d be known for.

Remarketing strategies became part of my identity. And that, my friend, is dangerous. Because if remarketing were to become less relevant, that would mean I was less relevant.

And I became less relevant.

It probably started with iOS 14 changes and the drop in reliability and completeness of remarketing audiences. But costs also increased while my audiences began to shrink. And Meta, faced with challenges related to privacy and charges of discrimination in targeting, focused on improving algorithmic targeting.

And this is when I started to be wrong. I would not allow myself to believe that Meta could reach the people I wanted to reach without me explicitly targeting them.

It wasn’t until Meta introduced Advantage+ Audience that it began to click. Even if you didn’t provide audience suggestions, Meta would prioritize people based on pixel activity, conversion data, and prior engagement with ads. You know… REMARKETING.

But I clung to my old strategies until Meta made audience segments available for manual sales campaigns. It wasn’t until then that I could actually prove that Meta was doing the things that were claimed.

And once I could prove it, I first had to admit to myself that I was wrong. And it was time to change course.

2. Targeting Control

Of course, I wasn’t just wrong about remarketing. I was wrong about all targeting control.

If I wasn’t remarketing, I was using interests and behaviors to reach a new audience. And if I wasn’t using interests and behaviors, I was using lookalike audiences. To make things really fun, I’d layer them together.

And I’d restrict by age, usually focusing on people aged 25 to 54. It didn’t matter what I optimized for.

This was the accepted way to reach a new, but relevant, audience. It wasn’t wrong years ago. But then Meta started making changes.

The first big moment was the introduction of audience expansion. It started as a checkbox to allow Meta to reach people beyond your chosen interests and behaviors if it would improve performance.

Then it was offered for lookalike audiences. And eventually, under the labels of Advantage+ Detailed Targeting and Advantage+ Lookalikes, these were forced for many of the most common performance goals.

I wanted to restrict my audience. But Meta wouldn’t let me. I initially revolted against the idea that my chosen audience could be expanded, and I’d have no way to prevent it.

Eventually, I gave up the fight. If I can’t restrict by these audiences, why should I obsess over which ones I reach? And if algorithmic targeting is already proven to reach my remarketing audience automatically, what does that say about the necessity of detailed targeting and lookalikes?

These audience inputs ceased being part of my focus when creating ad sets. I was wrong to fight it for as long as I did.

And while restricting by age made sense in certain circumstances, it took me far too long to realize that context matters. The only reason to restrict by age would be if Meta could find cheap, low-quality optimized actions from a certain age group.

Eventually, even this scenario wouldn’t require a restriction. Value rules became the solution.

3. Ads Focus

When my priority was targeting, I spent very little time worrying about the ads. I was convinced that great targeting provided the benefit of making even below-average ads effective.

And this was a nice perk for someone who didn’t love the ads side of advertising. I know that sounds ridiculous, but I’m an accidental marketer. I enjoy the stats and strategy and mechanics of advertising, but I’ve always hated the selling part.

Because I wasn’t trained in sales, I wasn’t particularly concerned about the psychology of sales copy. I convinced myself that my soft sales approach with low-quality creative that I whipped up in two minutes was good enough.

They weren’t. I was wrong. And it wasn’t until the craze over Andromeda that I started to figure it out.

I can actually thank my initial overreaction to Andromeda in the first half of 2025. To test what Andromeda meant to performance, I’d create 10, 20, and even 50 ads in an ad set. It forced me to dive deep into the copy and creative — things that never made me all that comfortable.

In the meantime, I started thinking about customer personas, pain points, solutions, and sales psychology — all things I had mostly avoided until that point.

While I can also say I was initially wrong about needing so many ads, that experience helped me get more comfortable with the importance of ads. I started to understand that not every potential customer was the same or would respond to the same text or creative.

These realizations coincided with embracing a hands-off approach to targeting. So once targeting took a back seat, my focus shifted to the ads themselves. If I wasn’t getting the results I wanted, I knew where to look.

In the past, I may have focused on the targeting. But now, I understand: Improved performance can be found with better ads.

4. Complexity

This truly happened as a result of abandoning my long-held beliefs about targeting.

If remarketing happens algorithmically, why do I need to create a separate remarketing ad set? And it went so much deeper than this since I’d create “Evergreen Campaigns” that included separate ad sets and ads for separate stages of the funnel based on when someone engaged with me.

If detailed targeting and lookalike audiences are automatically expanded, why do I need separate ad sets for different groups of detailed targeting and lookalikes when they could all, technically, reach the same people?

I was wrong. And because I was wrong, I could approach campaign complexity differently.

I cut out these extra ad sets, which eliminated much of the complexity that became a common trait of my advertising in the early days.

5. Control

Years ago, it made sense. Control targeting. Control placements. Control the message.

Ultimately, we were trying to guarantee that a specific ad looked a specific way for a specific group of people. And we’d “optimize” performance by further controlling these things.

I turned off placements that metrics suggested were less effective. I obsessed over winning copy and creative combinations. These are the types of actions taken when you assume your target audience is static.

I was wrong. And it’s a lesson I’ve come to embrace as control has steadily disappeared.

I know now that, because the algorithm is literal, Meta is trying to get results that reflect my performance goal. That’s a balancing act of costs, impressions, and performance by placement. As long as a placement isn’t known for generating low-quality optimized results, there’s no reason to control it.

The entire concept of five primary text and five headlines conflicted with my original approach years ago. It didn’t make sense. I wanted this specific message to be shown to that group of people. Why would I let Meta choose from other options?

And what do you mean I should provide up to 10 images and videos for a single ad? And I can’t see which one performs best? That’s nonsense.

I now understand that different people respond to different creative and messaging. Meta has more data than I do and can adjust dynamically to show the right versions to the right people. And ultimately, micro-results by primary text or headline or even creative aren’t all that important. We should be viewing results in aggregate to determine success.

6. Automation

If you’ve been doing this long enough, you may remember when Campaign Budget Optimization was introduced in 2018. Then, in early 2019, it was reported that it would eventually become a fixed default that couldn’t be turned off (this would never happen).

I nearly rioted. If I set a $100 daily budget for one ad set and a $50 ad set for another, that’s what I want to spend. That’s why I set those budgets in the first place!

I spent some time understanding how CBO worked. I tested it. And I realized that my budget would be spent optimally to help get the most results. It took the thinking out of setting my budgets.

I was wrong.

And you can certainly connect so many other automations to my former need to control things. Allowing budget to be spent on all placements. Allowing Meta to find my audience algorithmically. These were all things that would have originally seemed like nonsense.

Because I knew better, dammit. But I was wrong.

7. Creative Enhancements

I was like most advertisers when Advantage+ Creative enhancements first became a thing. I despised them. And many advertisers still do.

But it wasn’t until all of that time spent trying to better understand Andromeda and creative diversification that everything clicked. These enhancements help me create more variations of my ad to give Meta a larger pool of options.

I was wrong. Or maybe I was wrong about most of the enhancements. Some of them are still trash. But I’ll leave most of these enhancements on now.

The main thing I’ve come to understand is that creative enhancements are there to help me. They won’t change my images and videos universally. Instead, they give Meta the option to enhance a version that will become one of many versions that could be shown.

8. The Learning Phase

Don’t touch a campaign that’s working!

My voice was one of many shouting this, and it was related to the learning phase. Avoid any change that will restart the learning phase because that will mean that performance is likely to go into the toilet.

And maybe there was some validity to that claim for a while. But even when it became less true, I clung to it for too long.

I was wrong. I was wrong that the learning phase was something to be feared. And I was wrong that exiting the learning phase is a necessary priority.

The truth is you may never exit the learning phase when selling an expensive product. That doesn’t mean that you can’t get good results. The concept of 50 optimized actions in a week is a rule of thumb. More is better, but it’s not required.

And delivery isn’t as sensitive as it once was. You’re getting good results now, not because of some magical combination of settings and luck that can be disrupted with a change. You’re getting results because you have good ads.

I’m no longer fixated on ways to avoid restarting the learning phase. I’m not going to tell you to only raise budget by 20 percent or avoid creating new ads when you’re getting decent results. If you’re getting good results now, you can get good results again.

It’s not magic or luck, and that’s the main thing that’s helped me understand that I was wrong.

We Will Be Wrong

If you’re ever going to have success with Meta advertising — or anything, really — you need to understand that you will be wrong. If you are unable to embrace this possibility, you’re likely to create a fantasy land where everyone else is to blame but you.

It’s the algorithm. Meta’s automations. Server outages. Creative enhancements. Attribution. Targeting. Magical campaign structure.

I was wrong. The fact that I’m still here today, writing on this website, is because I was willing to accept that fact. And truly, there was once a danger that I wouldn’t be able to do that.

If you’re truly committed to getting the best results possible, you’ll also be committed to fully understanding how things work. That involves researching Meta’s documentation and running tests that aim to disprove your beliefs, not confirm them.

Once you’re open to being wrong, you’re much more likely to survive what is an absolutely wild ride of constant changes to features and functionality. And that ride isn’t ending anytime soon.

Buckle up.

Your Turn

What have you been wrong about related to Meta advertising?

Let me know in the comments below!

The post I Was Wrong: How My Approach to Meta Ads Changed appeared first on Jon Loomer Digital.

Meta desperately needs this feature…

Meta introduced site links a couple of years ago now. They allow advertisers to generate a variation of their ads that will showcase other products and services.

But there’s an element of this that is incredibly frustrating, and Meta can fix it.

The Problem

This is one of the most annoying exercises for advertisers. Every time you create an ad, you need to define your site links.

In the example above, Meta includes site links I’ve customized previously but also automatically detects other links and generates new site links for me. So then I’ll need to go through and manually remove the ones I don’t want to feature the ones I do.

Site links require a thumbnail image and label, so each one needs some thought and customization. Meta isn’t helping by automatically detecting these links.

The Solution

So here’s what I request: Add an account-wide setting for site links in Advertising Settings. Allow advertisers to globally define them so that a default setting will be used for each new ad.

This shouldn’t be that complicated. We shouldn’t need to remember to re-customize these links every single time we create a new ad.

I’m fine with Meta being able to recommend other site links, but they shouldn’t be included by default. Can we do this, Meta??

The post Site Links Feature Request appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A handful of small but meaningful updates stood out this week, from customer acquisition and attribution to audience segments and delivery. None require a full deep dive, but they’re all worth your attention right now.

1. Customer Lifecycle Strategy
2. Engage-through attribution rollout
3. Audience segments use case
4. Incremental attribution windows
5. Adaptive Ranking Model for Instagram

Let’s get to it…

1. Customer Lifecycle Strategy

I normally refrain from sharing information about a new feature unless one of these three things is true:

1. I have the feature
2. Meta mentions the feature
3. Someone I know and trust has it

So while I’ve heard rumblings about this one for more than a week, I’ve been more than a little skeptical about the information I’ve seen. But I’ve heard enough now that it does appear to be a thing.

This screenshot has been shared repeatedly on LinkedIn…

From what I have been able to sort out, it appears at the top of an ad set when creating a Sales campaign. As you can see from the screenshot, you’re given two options:

Reach new and existing customers: Choose this if your strategy is to get conversions with a lower cost per result and you don’t need to exclude people based on if they made a purchase.

Acquire new customers: Choose this if conversions from people who have never made a purchase generate more value for your business. This may increase your cost per result.

“Reach New and Existing Customers” would be the default option, and it’s what happens naturally now anyway. This should be the chosen method for most advertisers, but there may be exceptions when you’d want to only reach completely new people.

I’ve seen all sorts of weird speculation that choosing “Acquire New Customers” will automate the process of excluding your customers. You no longer need to create custom audiences to exclude them, and Meta will use their data and magical wand to exclude those people for you. And that all sounded like a bunch of nonsense.

I had my theories about how this would work that were far less mysterious. And based on the latest screenshot I’ve seen shared, my theories were correct.

Take a close look at that screenshot under the selection of “Acquire New Customers.” Meta lists the custom audiences that will be excluded based on that selection.

In other words, Meta is only able to exclude your current customers because you’ve defined the Existing Customers audience segment in your Advertising Settings.

This is a bit of a 2026 twist on the Existing Customer Budget Cap that originally existed for Advantage+ Shopping Campaigns. ASC was the reason that defining your existing customers became necessary in the first place, and audience segments were born.

Back to the second screenshot, you’ll see there’s one more option: “Reach engaged or non-engaged audiences.” In the example, “Reach both” is selected.

Once again, this appears to be pulling from your audience segments. You can also choose to exclude people who have engaged with your business in some way but haven’t purchased. These people are defined in your audience segments as Engaged Audience.

There’s a whole lot to discuss here regarding whether it’s a reasonable strategy to exclude those who have bought or engaged with your business before. That’s a separate topic for another day.

For now, it does appear that this is a thing. But it’s also not as interesting as many are suggesting. If you wanted to exclude these people before, you could have manually. And, of course, the accuracy of those exclusions will be dependent on how you define your audience segments in the first place.

2. Engage-Through Attribution Rollout

A month ago, Meta announced some significant changes to attribution. Not only does click-through attribution now require a click on a link, but engage-through attribution replaces engaged-view. All of the social and other clicks that were previously counted under click-through attribution move to engage-through.

Earlier this week, advertisers started seeing the changes reflected in the ad set.

And now we’re also seeing it in the reporting tools. When using the breakdown by attribution settings, “1-day engagement” replaces “1-day engaged-view.”

And you’ll also see this reflected when using the Compare Attribution Settings feature.

What’s confusing about using these tools is what becomes of the data that was previously defined under the old definitions of click-through and engaged-view attribution. Just keep this in mind when viewing historical data prior to, or overlapping with, the rollout of these changes.

3. Audience Segments Use Case

Audience segments are one of my favorite features, and I know I talk about them a lot. While the primary use case of audience segments is to prove how much remarketing happens naturally, there’s another way to use this feature to help explain performance fluctuations.

Here’s an example…

I have a campaign that has been running for eight weeks and performance has decreased over time. I used the breakdown by audience segments, and I found it interesting how the percentage of ad spend dedicated to remarketing (as defined by the Engaged Audience and Existing Customers audience segments) declined over time.

• Week 1: 19.6%
• Week 2: 19.9%
• Week 3: 21.6%
• Week 4: 27.5%
• Week 5: 26.7%
• Week 6: 8.4%
• Week 7: 2.4%
• Week 8: 2.4%

So, Meta spent as much as 27.5% of my budget on remarketing in those initial weeks. But this slice dropped to 2.4% during the most recent weeks. Not coincidentally, the cost per conversion has risen during that time.

The natural reaction might be to assume something is wrong. Why would Meta shift so dramatically in remarketing? But the reality is that hidden beneath the surface was a 9.0 frequency for remarketing audiences during this time, which was double that of the new audience.

In other words, Meta leaned into remarketing early to get the best results. But as these audiences were exhausted, effectiveness decreased. Meta then relied less on remarketing, which meant new audiences became the primary source of conversions.

The decline in performance that I saw wasn’t much of a decline at all. The new audience performance remained mostly the same. It’s just that remarketing became less available to bring down my costs.

This doesn’t mean that my remarketing audiences are completely exhausted. I also haven’t created new ads in a few weeks. Once I do, those new ads are likely to be shown to my remarketing audiences again.

This example is a reminder that your performance fluctuations can often be explained by natural delivery shifts. Don’t forget to use the breakdown by audience segments to help diagnose it.

4. Incremental Attribution Windows

Beginning about a year ago, advertisers have had the option of Standard or Incremental attribution in the ad set when maximizing the number of conversions.

When Incremental attribution is selected, Meta will use models that predict whether a conversion was caused by your ad. This is different from Standard attribution, which focuses on conversions that happen after clicks or views within a defined window of time.

It’s important to remember this because the breakdown by attribution settings won’t work when using Incremental attribution. Typically, this breakdown will segment your results by attribution window.

But these rows will not appear if you used Incremental attribution. The reason for that is breakdowns will only segment results relevant to your settings. Since you aren’t using Standard attribution, rows for those windows wouldn’t be relevant.

Of course, you can uncover conversion results outside of your settings by using the Compare Attribution Settings feature. When used, you can add columns for the click-through, engage-through, and view-through windows and see how they compare to your Incremental results.

I’ve found that Incremental results tend to be consistent with click-through conversions, though they don’t match up exactly. View-through and engage-through can be incremental, too.

5. Adaptive Ranking Model for Instagram

Meta recently detailed improvements made to show more relevant ads to people on Instagram. This was shared on its Engineering Blog, so the audience understands the technical details that will make little sense to the rest of us normies.

Here’s why Meta says they developed the Adaptive Ranking Model:

This increase in scale & complexity exacerbates a fundamental ‘inference trilemma’: the challenge of balancing the increased model complexity and associated need for compute and memory with the low latency and cost efficiency required for a global service serving billions of people. To overcome this, we have developed the Meta Adaptive Ranking Model, which effectively bends the inference scaling curve with high ROI and industry-leading efficiency.

Uhh….

Thankfully, Andrew Hutchinson at Social Media Today helped interpret.

The bottom line is less compute power and more relevant ads. Meta says that since launching the Adaptive Ranking Model on Instagram in Q4 of 2025, advertisers have seen a 3% increase in conversions and a 5% increase in click-through rate for targeted users. Maybe not significant, but relevant.

We’ve heard a lot about Meta’s developments related to new hardware and software to improve ad performance. While Andromeda, GEM, and Lattice get a lot of attention, add Adaptive Ranking Model to the list of technologies that Meta is working on to help advertisers reach their ideal customers most efficiently.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post Customer Lifecycle Strategy, Attribution Updates, and More appeared first on Jon Loomer Digital.

Some Meta advertisers have spotted a new “Describe Your Audience” AI feature for detailed targeting. I finally have it in one ad account, so let’s walk through what it is and isn’t.

It’s not clear if what advertisers are seeing is a test or the early stages of an official rollout. I’m not seeing any updates to Meta’s official documentation related to detailed targeting that would indicate this is a new fixture going forward.

I have some thoughts. The vast majority of reactions I’ve seen exaggerate or misunderstand what this feature actually does. My hope is to clarify things here.

Let’s get to it…

How It Works

If you have this feature, you’ll see “Describe Your Audience” as a new, default tab for detailed targeting.

The other option is to “Use Original Options.” If you don’t have this feature, it will look like this…

With “Describe Your Audience” selected, click “Get Started.” You’ll see a window that looks like this…

The instructions say to “describe your ideal audience and our AI will find relevant interests and behaviors.” Once you enter a prompt in under 2,000 characters that describes your ideal audience, click “Find Options.”

After a few seconds, Meta will generate a list of interests and behaviors that match your prompt.

If you hover over an interest, you’ll see an explanation of why Meta believes it’s relevant to your ideal audience.

You can individually remove interests, remove them all, or revise your prompt and have Meta find more options. Once you’re ready, click “add all” to add the selected interests and behaviors to your audience.

Your detailed targeting will then be updated with those items.

Here’s a quick video that walks through the process…

Is This Anything?

I’ve seen a lot of excitement about this new feature from advertisers, particularly on LinkedIn. To sort out whether this is a big deal, we need to answer a few questions…

1. Does it offer anything new?

Not really.

Yes, it’s a cool way to find interests and behaviors that you would otherwise search out manually. But these aren’t new options. This might surface interests and behaviors you wouldn’t have considered before, but they were always there.

2. Does it add more control?

No.

Within the default audience creation view, all detailed targeting inputs are suggestions.

If you click the option to “further limit the reach of your ads,” detailed targeting remains a suggestion for 11 of the most commonly used performance goals.

The advertiser’s level of control remains the same, which is minimal.

3. Does it make detailed targeting more accurate or effective?

The natural response to technology like this is awe because it automates stuff and makes it easier. But does it make it better?

The quality of the outputs will rely heavily on the quality of your inputs. You are still responsible for writing out a prompt that will generate relevant interests and behaviors.

That doesn’t mean we should automatically accept all of Meta’s recommendations either. This feature still needs to accurately surface the right detailed targeting options, including those you may not have considered, to be useful.

To be truly valuable, it needs to provably impact performance. And that takes us here…

4. Does it even matter?

The main thing to remember here is that this does not substantively change anything about our inputs or control. This feature only surfaces options that were previously available. It’s a novelty.

Whether the inputs are selected manually or found using the “Describe Your Audience” tool, they’re treated the same way. They’re only audience suggestions in most cases. And we have no reliable way to prove whether Meta uses them or whether they make a positive impact on performance.

If you’re using one of the few performance goals that allow advertisers to restrict by detailed targeting, this could become more valuable. That assumes, of course, that the output generated by the AI leads to better results. And we won’t know that without testing.

Verdict: Meh

I hate to say it, but this feature is mostly pointless.

It would have been useful five to 10 years ago when control over detailed targeting was possible and useful. But, of course, such technology to automate this would not have existed then.

If you believe that the detailed targeting suggestions you provide make a significant impact on performance, you probably think this is a big deal. I just don’t think that’s the case. I doubt suggestions can hurt you, but there’s little evidence that they do much of anything.

Meta has even de-emphasized the use of audience suggestions. They’ve removed the button to provide suggestions in the default view. Language indicating audience suggestions are “prioritized” before going broader has seemingly disappeared.

I just don’t understand why time was spent developing this feature. If Meta actually wants advertisers to provide detailed targeting suggestions because those suggestions have proven to improve performance, why are all of the platform’s trends moving in the opposite direction?

Your Turn

Have you experimented with this feature yet? What do you think?

Let me know in the comments below!

The post Meta’s “Describe Your Audience” AI Feature for Detailed Targeting appeared first on Jon Loomer Digital.

This feature changes everything…

The problem is that far too few advertisers use it. And many who do aren’t using it correctly.

What are Audience Segments?

Meta introduced audience segments for manual sales campaigns in the summer of 2024. I credit this feature for completely transforming my understanding of algorithmic targeting.

Advertisers who accurately define their engaged audience and existing customers get incredible insights.

When you run sales campaigns using algorithmic targeting, you can break down results by audience segments. This helps you see how ad spend and conversions are distributed between remarketing and prospecting audiences.

The Impact

Before I had this information, I thought Meta needed me to explicitly target remarketing audiences. But once I used it, I understood how much remarketing happens by default.

The impact of having this information is that it validates a more hands-off approach. You can simplify targeting and ad set creation because you can see that your ads are already reaching these people.

I created a free mini-course to help you get the most out of audience segments. It takes you through how to define them, break down the results, and use that information.

The post Audience Segments Change Everything appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A handful of small updates stood out this week, from new delivery controls and audience filters to questions about location targeting, video translation limits, and Threads ads.

1. “Push delivery to this ad”
2. Engagement frequency filter
3. Did Meta change location targeting?
4. Video translation limitations
5. Threads ads updates

Let’s get to it…

1. “Push Delivery to This Ad”

Some advertisers have reported seeing a new “Push delivery to this ad” option.

I don’t seem to have it yet, but here’s what I’ve been able to figure out based on what I’ve seen so far…

For published ads that aren’t getting distribution.

I’ve seen versions of this where an advertiser sees an alert in the Delivery column that a published ad is getting “limited or no ad spend.” The advertiser is prompted to “push delivery” to that ad by dedicating a portion of their budget to it.

Similar to the creative testing tool.

The screenshot above may look familiar since it looks almost exactly like that of the creative testing tool. For example, here’s what that feature looks like…

In fact, it appears that Meta completely stole the functionality from the creative testing tool and applied it to “push delivery.” Meta even seems to have mistakenly included text related to it being a “test.” But this shouldn’t be a test since you’d only be pushing delivery to a single ad.

How it seems to work.

I can only make educated guesses about how this works, but I’ve seen enough that I think I can piece it together.

If an ad isn’t getting impressions, you can choose to dedicate a specific percentage of your campaign or ad set budget to the ad for a designated number of days. This will allow you to see what would happen if the ad were to actually get impressions so that you can decide whether any other changes need to be made.

The other difference between this and the creative testing tool is that it appears to be intended for ads that were already published and running (though it’s not clear if this can be applied for new ads when published). The creative testing tool is only for new ads, which is actually one of its annoying quirks.

This could change some things.

I have to admit I’m chuckling to myself a bit about how excited some advertisers have been about this. I’ve been pushing the creative testing tool for months for this precise purpose. The creative testing tool is part of my routine, as I start new ads with a test.

But the addition of “push delivery” could change that approach. One reason I used the test for all new ads is that we couldn’t apply the creative testing tool to active ads. But this new tool gives us flexibility.

Now you can publish an ad without an initial test and see how Meta distributes budget normally. If you have questions about why it’s not getting impressions, you can push budget to it for a few days and see what happens.

I’m looking forward to trying this out when I have it.

2. Engagement Frequency Filter

Here’s a change that’s only about eight years too late…

Advertisers who create engagement custom audiences for their Facebook page or Instagram profile will notice something new: A filter for either “in the past” or “at least.”

For example, you could create an audience of people who engaged with your Facebook page in the past 14-21 days.

Or you could isolate the people who engaged with your Facebook page at least 3 times in the past 90 days.

These filters apply to the following types of custom audiences for Facebook pages:

• People who engaged with your page
• People who visited your page
• People who engaged with any post or ad
• People who clicked any call-to-action button
• People who sent a message to your page
• People who saved any post

And they apply to the following types of custom audiences for your Instagram account:

• People who engaged with this professional account
• People who visited this professional account’s profile
• People who engaged with any post or ad
• People who sent a message to this professional account
• People who saved any post or ad

So… interesting. But like I said, it’s about eight years too late.

Why would Meta choose to do this now? Are you trying to tell me that Meta doesn’t already prioritize the people who engage most with our Facebook and Instagram accounts? This is the kind of remarketing that should be left mostly in the past.

Some people will be really excited about this. But it will likely be forgotten soon after.

3. Did Meta Change Location Targeting?

This is one of those very subtle changes that would be easy to miss, but it jumped out at me. It may not mean anything. Or maybe it means a lot.

Here’s what the location targeting box looks like today…

Do you see what’s missing? Here’s the exact same box from February 15th…

Meta removed the line that location targeting allowed you to “reach people living in or recently in this location.”

I’m not sure if Meta’s documentation changed, but it does mention “living in or recently in” within the explanation. But it’s quite possible that it’s been softened. Here’s how it reads now:

Location targeting lets you show ads to people who spend time in locations, such as people living there or recently there.

Maybe it’s always read that way. But the “such as” implies that it includes people beyond those “living there or recently there.”

It’s certainly vague. I have questions.

4. Video Translation Limitations

This isn’t necessarily ad-related, but I don’t know what this is all about…

I use Business Suite on desktop to schedule Reels. I first started using the video translation feature last September. It’s pretty awesome. It allows me to generate versions of my video that translate my voice into other languages.

Last year, I published three of these per week without issue. Now I’m only publishing one per week. Then, without warning, I started seeing this message.

It’s a “limited new feature” now? Even though I was previously using it three times per week, now I’ve suddenly exceeded my max while using it once per week?

It’s not clear what the maximum is or what this is all about. But pretty annoying.

5. Threads Ads Updates

Meta made a couple of Marketing API updates to Threads ads:

App Ads now supported: App ads support the Threads Feed placement through the Marketing API globally. This is for the App Promotion campaign objective with performance goals related to link clicks and conversions (or at least that’s how I interpret it).

Reply Moderation Tooling: View, hide, and reply to top-level replies on your Threads ads. Reply moderation lets you manage the conversation on your ads by reading replies, hiding content, and adding your own replies.

These are Marketing API updates, so they apply to third-party tools that allow you to manage your ads. Presumably, this will also carry over to the main interface if it hasn’t already.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post “Push Delivery to This Ad” (Plus 4 Updates) appeared first on Jon Loomer Digital.

Meta ads targeting is more confusing and controversial than ever. That is, at least, if you listen to all of the loud recommendations about strategies and control.

The reality is that targeting has never been simpler. It’s just confusing if you don’t take the time to understand your role in it, and when you should insert yourself.

And that’s the point of this post. Targeting inputs fall into three categories now:

1. Audience Controls
2. Audience Suggestions
3. Audience Restrictions

It’s your job to understand all three levels: What they mean, why they matter, when control is necessary, and when methods other than an audience restriction should be preferred.

Let’s get to it…

Age Range

Advertisers can set an audience control using a minimum age between 18 and 25.

You can also provide an age range suggestion that doesn’t act as a tight constraint.

Or you can turn Advantage+ off and set a strict age minimum and maximum.

How to Approach:

Avoid restricting by age when possible. The one exception would be when selling age-restricted goods, where you cannot serve people under or over a certain age. Such a restriction can also be done at the account level.

An age range suggestion shouldn’t hurt delivery, but it also won’t necessarily impact it. Meta will ignore suggestions if it means helping you get more optimized actions, as defined by your performance goal.

A common mistake that advertisers make is that they envision their ideal customer when defining the ad set audience. Whether based on data or not, advertisers might restrict by ages 25 to 54 because they believe this age range most accurately reflects their customers.

While that may indeed be a reflection of customer data, such a restriction is rarely necessary. Meta has mounds of data to help find the people who are most likely to perform the action that you want, and this is likely to be reflected in budget distribution. Restricting your audience also restricts the algorithm, preventing Meta from spending even a small amount of your budget on a group that could lead to results.

In some cases, Meta will concentrate budget on certain age ranges because it has proven to drive more of the results that you want. But you might find that this is because an age range is the source of cheap and low-quality results.

Consider lead generation as an example. I’ve found that when running ads for leads, Meta will concentrate a high percentage of my budget on people aged 65 and up. The reason for this is that, for whatever reason, this age group results in cheap leads at a high rate. I have information that Meta doesn’t, which is that these aren’t high-quality leads.

Don’t overthink this. Prioritize limiting restrictions. Know that any guidance Meta needs is most likely related to weaknesses that will lead to cheap and low-quality results. Know how to detect such behavior.

When it happens, prioritize using value rules to adjust bids, rather than removing age ranges entirely.

Gender

Advertisers cannot set an audience control for gender in the default setup. You can provide an audience suggestion for gender that doesn’t act as a tight constraint.

Or you can turn Advantage+ off and set a strict restriction by gender.

How to Approach:

Avoid restricting by gender when possible. A gender audience suggestion shouldn’t hurt delivery, but it also won’t necessarily impact it. Meta will ignore suggestions if it means helping you get more optimized actions, as defined by your performance goal.

As is the case with age ranges, advertisers often fall into the trap of restricting by gender to more accurately reflect their ideal customer. Such a restriction is rarely necessary, and it can prevent you from spending even a small amount of budget on a segment that could lead to customers.

In some cases, Meta will concentrate budget on a gender because it has proven to drive more of the results that you want. As is the case with age range, this can be because a gender is the source of cheap and low-quality results.

For example, consider a situation of a service focused on helping women-owned businesses. When optimizing for a purchase, gender restriction shouldn’t be necessary. But if they run ads with a performance goal related to an engagement-level action, Meta is likely to show their ads to men if they’re likely to engage. This could mean spending a lot of your budget on people not likely to purchase.

If there’s such a problem to be solved, prioritize using value rules to adjust bids, rather than removing a gender entirely.

Location

Location is an audience control. You can include specific countries, states/regions, cities, postal codes, addresses, DMAs, Comscore Markets, or congressional districts. You can also add worldwide or global regions.

Since location is an audience control and not a suggestion, Meta will not reach people outside of those locations. However, there’s a major caveat. Meta defines those who would qualify to be reached under this setting as “people who spend time in locations, such as people living there or recently there.” You cannot isolate residents or travelers.

When a city, state, or region within a country is selected, advertisers have the option to “reach more people likely to respond.”

When selected, you can also reach people who wouldn’t qualify under living in or recently in that location, but they live in that country and have provided signals that indicate they are interested in that location.

How to Approach:

The main thing to understand here is that advertisers cannot isolate local residents or travelers. The definition of a person who qualifies when a location is selected will automatically include both.

That’s frustrating, but it’s not something we can successfully overcome. I know that some advertisers will target one selection of cities while excluding the surrounding areas. By doing so, you’d also exclude some of the people you want to reach (for example, those who live in the area you’re targeting, but commute to the excluded area for work).

So if you’re hoping to reach people in a local area, include it in locations. Know that you are likely to reach some people you didn’t want to reach. But address this with ad copy and creative to clearly appeal most to the people you want to reach.

Otherwise, most businesses will be focused on country targeting. In most cases, you can and should combine multiple countries in the same ad set, particularly when costs are similar between them (for example, targeting the United States, Canada, United Kingdom, and Australia).

The ideal scenario might be to separate countries into their own ad sets to create ads that are most relevant to residents of each country. But you should only do this if you have the budget that can be adequately spent on each country to bring in a volume of results that is meaningful.

To keep with the theme of this post, do not overthink location targeting. In most cases, keep it simple. Target the relevant locations and keep it broad when possible. Consolidate locations in most cases, leaving opportunities to separate them as exceptions to the rule.

Custom Audience Exclusions

Custom audience exclusions are an audience control. Meta will respect such an exclusion as a tight constraint.

How to Approach:

Custom audience exclusions make the most sense when promoting a product that can only be purchased once. You may not want to waste budget on reaching people who cannot buy. That said, there are a couple of factors to consider.

First, it’s virtually impossible to successfully exclude everyone. This isn’t because Meta ignores your exclusions. It’s because custom audiences are, by nature, flawed and incomplete. Expect that you will always reach some people you do not intend to reach.

Second, an argument can be made for including prior customers, even if they cannot buy again. They are the people most likely to share positive comments. Your ad may also be a reminder to make a related purchase, though it may not be of that product.

The bottom line is this: You can exclude custom audiences. Limit such exclusions to situations that make the most sense to avoid wasted ad spend. But know that custom audience exclusions are imperfect, and you will always reach some of these people anyway.

Detailed Targeting

Detailed targeting is one of the oldest types of audience selection for Meta advertisers. It allows you to select categories of interests and behaviors that match your intended audience, based on the data that Meta has from both in-app and off-app signals.

Any detailed targeting inputs will be suggestions by default. Meta will go beyond suggestions to help you get more results.

Even if you attempt to turn Advantage+ off, you can rarely restrict by detailed targeting inputs. For 11 of the most common performance goals, detailed targeting will always be used as a suggestion. You can remove suggestions for the remaining performance goals, when applicable.

How to Approach:

There are multiple layers to answering the question of how you should approach detailed targeting.

1. Can you restrict by detailed targeting inputs? In most cases, no. Detailed targeting is almost always used as a suggestion.

2. Do detailed targeting suggestions positively impact distribution and performance? We don’t know for sure. Outside of using A/B tests, which could vary widely based on the account and detailed targeting suggestions used, Meta offers no way to prove how much suggestions are used and how much they matter.

Suggestions might matter sometimes. Meta might mostly ignore them at others. We have no idea how suggestions are applied from case to case.

Even Meta’s documentation has softened on how suggestions are applied. Meta previously said that suggestions were “prioritized” before going more broadly if it meant getting more results. But that mention of prioritization disappeared.

I wouldn’t tie my strategy to an audience input that has completely unknown impact. Detailed targeting suggestions can’t hurt, and there’s an argument (even if unproven) that they may be helpful in cases when Meta has less data.

But the biggest mistake you can make is to treat detailed targeting suggestions as tight constraints. They’re not. Do not create separate ad sets based on different detailed targeting inputs. You’re only creating multiple ad sets that have the capability of reaching all of the same people.

Keep it simple. Use detailed targeting suggestions or don’t. But don’t build around them.

Lookalike Audiences

Meta introduced lookalike audiences in 2013 as a way to simplify the process of isolating people most similar to our current customers. Lookalike audiences use custom audiences as the source audience to learn from when developing a larger pool of people.

By default, lookalike audiences are only used as audience suggestions.

Even if you attempt to turn Advantage+ off, you can rarely restrict by lookalike audience inputs. For nine of the most common performance goals, lookalike audiences will always be used as suggestions. You can remove suggestions for the remaining performance goals, when applicable.

How to Approach:

See the approach to detailed targeting above since you can apply the same to lookalike audiences. Your lookalike audience inputs are almost always used as suggestions now, and it is rarely possible to restrict by that audience. So the question is whether the suggestion matters.

You can make an argument it makes sense that Meta could benefit from lookalike audiences in certain circumstances. If you’re running ads for a new account and Meta doesn’t have any data, a lookalike audience suggestion might be beneficial.

But that’s only a theory. Meta also learns quickly in real-time, and we may be putting too much value on that historical data. And since we have no idea how much Meta leans on audience suggestions (and this may not be consistent from case to case), it’s impossible to truly know the impact.

Would a lookalike audience suggestion hurt performance? Probably not. I totally understand why you might use one, particularly with new accounts. But remember that these inputs are almost always suggestions, so don’t approach them as tight constraints.

Do not create multiple ad sets with different lookalike audience and detailed targeting inputs. Because these are only suggestions, all of the ad sets can reach the same people.

But feel free to use these suggestions as a little magic pixie dust.

Custom Audiences (Remarketing)

Custom audiences reflect groups of people connected to your business. Advertisers can create custom audiences based on website activity, customer lists, on-platform activity, and more.

By default, custom audiences are only used as audience suggestions.

If you turn Advantage+ off, you can restrict by custom audiences and remove the suggestion.

This is also known as remarketing.

How to Approach:

As I’ve said repeatedly throughout this post, audience suggestions shouldn’t hurt you. If you want to add a custom audience as a suggestion, knock yourself out. But I have doubts whether it will matter.

Meta already prioritizes remarketing audiences by default. This was made clear with the rollout of Advantage+ Audience in the fall of 2023. If you don’t provide an audience suggestion, Meta will prioritize pixel activity, conversion data, and prior engagement with your ads.

Remarketing happens by default, and you can prove it with audience segments and sales campaigns. Once you’ve defined your engaged audience and existing customers audience segments, you can break down results to see how conversions and ad spend are distributed between remarketing and prospecting groups.

Might custom audience suggestions be more beneficial for new accounts where Meta lacks data to prioritize remarketing groups automatically? It’s certainly possible. Feel free to use custom audiences as suggestions in those cases.

Should you restrict by custom audiences? That’s debatable.

Years ago, about 90% of my advertising leaned on remarketing strategies. I created custom audiences for everything imaginable in an effort to target the most relevant groups possible.

But these days, I understand that most of what I did then is no longer relevant or necessary. Meta doesn’t need me to isolate these people. Remarketing already happens by default. And by not restricting by custom audiences, Meta can dynamically balance budget and performance between remarketing and prospecting audiences.

Are there exceptions when remarketing can be beneficial? Yes, but they are rare. And I use them sometimes. Particularly for high-ticket products, it can make sense to remarket to the people who completed a low-ticket purchase or registration. But even then, these ads are used in addition to the other sales efforts that should be happening to convert these people.

The main thing to understand is that general remarketing should no longer be necessary. There’s rarely a reason to isolate all of your website visitors or entire email list and only target those people. Meta will already be prioritizing them, all while also reaching new people.

The only other exception may be if Meta is leaning too heavily into remarketing and you want to focus more on bringing completely new people into your funnel. Remarketing results are easier, so Meta will take advantage of that. If you need to control remarketing spend, you might create separate ad sets with strict budgets, one for remarketing and one for prospecting (excluding the remarketing audience).

A General Approach

Now let’s put together a general approach to targeting, simplifying the suggestions above and focusing on overarching themes.

1. Take mostly a hands-off approach.

This seems to be the most difficult concept for advertisers. You don’t need to control everything. In many cases now, you can’t. Stop fighting it.

2. Use suggestions — or don’t.

As long as you know they’re suggestions, feel free to use them. I have serious doubts whether they make a difference. Meta removed language indicating they’re prioritized. But as long as you don’t let them control what you do, suggestions can’t hurt. They might even be beneficial in some cases.

3. Avoid restricting by age or gender.

You should restrict by age when selling age-restricted products. But otherwise, you shouldn’t need to restrict by age or gender to match your ideal customer.

That doesn’t mean that Meta is perfect when it comes to distribution by age and gender. Meta can exploit weaknesses based on performance goal in an effort to get you more results. This can lead to low-quality leads or engagement because it’s information that you have and Meta doesn’t.

When this is proven to be a problem, use value rules to adjust bids by age or gender, rather than restricting by demographic group. Your priority should be to avoid restrictions and use other methods, when possible.

4. Remarketing happens by default.

You don’t need to isolate general remarketing groups because Meta already prioritizes them. You can prove this to yourself with sales campaigns and audience segments.

While remarketing is rarely necessary now, there are rare exceptions. Just make sure to approach remarketing as such an exception, rather than the rule.

5. Simple targeting means simple campaign construction.

When you’re following steps 1 to 4, you quickly eliminate many of the reasons you’d previously create separate ad sets. You’ll find that, in most cases, you can get by with only a single ad set. Prioritize that, and only add complexity when it’s completely necessary.

Your Turn

Is there anything else about targeting that you’d add?

Let me know in the comments below!

The post A Guide to Meta Ads Targeting in 2026 appeared first on Jon Loomer Digital.

Most targeting inputs are audience suggestions by default now. And if you use detailed targeting or lookalike audiences, they are only suggestions for the most common performance goals.

But whether Meta actually wants us to provide suggestions isn’t clear. Based on recent changes, you have to wonder.

The Signs

At some point, Meta removed language indicating audience suggestions are prioritized before going more broadly.

Previously, Meta said this:

“If you share an audience suggestion, we’ll prioritize audiences matching this profile before searching more widely.”

But now the language about how suggestions are used is much less committal.

“We’ll also reach people beyond any custom audience, age, gender and detailed targeting settings you apply when it’s likely to improve performance.”

That doesn’t sound like it’ll be prioritized. And Meta recently removed the button to provide suggestions, which draws less attention to that option.

No Clear Stand

Does Meta recommend providing a suggestion now?

I can’t find anything in documentation that it’s what Meta wants. And if suggestions were helpful, you’d think Meta would say that and provide a better way to prove it.

I discuss this topic in my free mini-course. Subscribe to Modern Targeting to learn more.

The post Does Meta Want Audience Suggestions? appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A mix of AI-related risk, new automation tools, and ad creation updates stood out this week. Some of these are small, some are potentially more serious, but together they say a lot about where Meta is headed when it comes to control, guidance, and AI inside its business tools.

1. AI-related ad account shutdowns
2. New Meta Manus connectors
3. Meta AI Business Assistant
4. Add Business AI enhancement
5. Format selection is changing
6. ComScore Markets for auto ads

Let’s get to it…

1. AI-Related Ad Account Shutdowns

If you’ve been using the Meta ads connector to connect your ad account to Claude or other AI agents, tread very carefully. Maybe disconnect entirely for now until Meta’s stand on this technology clarifies.

We’ve seen several reports during the past week or so of advertisers getting their accounts shut down without warning. Like this from Naman Bansal, who said on LinkedIn that they’d lost 17 ad accounts due to AI…

Or this, from Tomas Kliment, who tells a similar story…

It sounds like some have been able to get their ad accounts back and some haven’t. But as we all know, the appeal process can be frustrating, with no ability for a second review in some cases.

If I were to guess, I don’t believe this is intentional on Meta’s part. Sure, they want to encourage the use of Manus (coming in the next update below), but this would be overly aggressive even by Meta’s standards. It’s more than likely a scenario of not accounting for the large volume of API calls that are happening.

Does that mean you should continue using these tools? Heck no. Stop until you get the all clear. It’s just not worth the risk.

2. New Meta Manus Connectors

Coincidentally (maybe?), Meta put out an announcement yesterday about the various ways you can unlock the “Power of Manus and in Meta’s Business and Creator Tools.” Meta formally announced the connectors available across Ads Manager, Instagram, and WhatsApp.

Ads Manager: While we know about the problems mentioned above, you should be safe using Meta’s official Ads Manager connector to Manus. When connected, “Manus identifies actionable insights, and creates meaningful output formats such as websites, dashboards, slides and infographics.”

Instagram Creator Marketplace: Manus helps match businesses with creators whose audiences align with their marketing goals.

Instagram: Create, publish, and analyze all Instagram content types in one streamlined workflow.

WhatsApp: Beginning to roll out now, Manus handles tasks like checking calendars, drafting responses, and creating to-do lists to help small business owners manage their work directly within WhatsApp chats.

3. Meta AI Business Assistant

I know that this has been rolling out for a while, but I was finally granted access while writing this post. An alert popped up for Meta AI Business Assistant that looks like this…

Once you connect an ad account, Meta opens a panel on the right to initiate a chat with Meta AI.

You can either pick from some generic options or ask a custom question. I asked it how I could get better results from my active ads. While it does give an analysis and makes recommendations, I wouldn’t say there’s any added value. It just recommends the same stuff you’d get from Opportunity Score.

I’m just digging in, but it’s interesting, though I’m not sure it’s useful enough that I’ll make it part of my routine. It’s hopefully most useful for support-related situations where Meta’s human and automated support systems otherwise fail.

4. Add Business AI Enhancement

Speaking of Meta Business AI, I’m now seeing this enhancement when creating ads.

It’s not a new enhancement, but it’s been in testing or rolling out for the past several months. This is the first time I’ve seen it in one of my accounts.

The Add Business AI enhancement is available for Sales campaigns where you utilize any of the website-related conversion locations. It works within Facebook Feed, Facebook Reels, Facebook Stories, Instagram Feed, Instagram Reels, and Instagram Stories placements. An AI agent can instantly answer questions that people have about your product or business that are featured in the ad.

There are a couple of weird things about this enhancement so far…

First, the enhancement may be available even if you’re not currently eligible for the web, Messenger, or WhatsApp integrations of Business AI.

Second, I haven’t seen a way to preview how this would work with my ads. How can we trust that Meta AI will answer questions accurately? Is it based only on the ad copy and landing page information?

Certainly a risk, but maybe you’re adventurous. I turned it on for fun.

5. Format Selection is Changing

You may have spotted this message when creating an ad…

It reads:

Format selection is changing. Soon, Format display options in Ad creative will be the new way to choose formats. This also gives you the flexibility to show ads in multiple formats.

The Format Display Options feature is reasonably new, but I’ve been using it for several months now.

Admittedly, I haven’t seen it as a complete replacement for format selection to this point. I customize site links and they’re used to generate carousel and collection options. And if you choose carousel as the main format, you can use Format Display Options to generate ads that create variations featuring individual carousel cards.

What isn’t clear to me is whether anything else about the ad creation flow will update substantially when format selection goes away. That’s the biggest barrier to me understanding how this change will be executed.

Will we still be able to customize creative by placement? Provide creative options for different aspect ratios? I sure hope so.

6. ComScore Markets for Auto Ads

And finally, Meta announced that it is transitioning from Nielsen DMA to Comscore Markets for automotive mobile ads.

Per Meta, from the announcement:

This transition to Comscore Markets® allows us to provide advertisers with a partner solution that is more sustainable and scalable for long-term performance and measurement.

This change only applies to you if you run automotive ads. Here’s the developer rollout timeline…

• March 23: Advertisers will have the ability to update their vehicle offer feed to include Comscore Market IDs, mapping, and feeds instead of Nielsen DMA.
• April 20: A notification will be issued during feed ingestion for Catalog Ads if dma_codes is still included in the vehicle offer feed.
• June 22: Nielsen’s DMAs will be removed from all Meta targeting and reporting solutions.

Your Turn

What do you think about these updates?

Let me know in the comments below!

The post AI-Related Ad Account Shutdowns, Meta AI Chat, and More appeared first on Jon Loomer Digital.

Sometimes weird stuff happens.

One of the things Meta advertisers need to understand is that most of what happens with their ads is explainable. When you understand how things work, you see the ups and downs of performance as logical and not the work of some rogue algorithm.

Of course, sometimes weird stuff happens, too, and you have to account for it. An example of that is the focus of this post.

“Weird stuff” in this context includes the things that are either difficult to explain or completely oppose how things are expected to work. Performance shifted, but not due to reasonable explanations like creative fatigue, an increase in competition, seasonal behavior, website issues, event delays, or even randomness.

This weird stuff is reflected in data, and not on gut. And it’s rarely found by looking at the surface-level results. You need to dig deeper.

Let me share a story…

A Stable Campaign No More

I was working with a professional service company recently that enjoyed incredible stability and predictability. The goal for one campaign was leads with a budget of more than $20k per month.

You could set a clock to the results. In fact, the cost per lead was only getting better from month to month. Cost per lead hovered around $60 last summer but has slowly come down since. During the months of January and February, cost per lead had settled at a campaign best of about $50.

And then, on precisely March 1st, something happened.

Cost per lead for the first five days of March was $114, with four of the five days over $100. A bad day here or there can happen and is nothing to be all that concerned about. But this was a trend. And for context, there wasn’t a single day of January or February where cost per lead was over $100.

There had been no recent changes made to the campaign, ad set, or ads. But something had caused a drop in performance.

This had become a problem.

Considering Logical Explanations

Whenever you see a drop in performance, the initial focus should be on finding a logical explanation based on how things work. This will lead to a logical solution. You shouldn’t immediately jump to weird stuff since that will be the exception, rather than the rule.

CPM was up to $57 in those initial days of March, which was about 42% higher than normal. While CPM can fluctuate, it otherwise hadn’t been higher than $43, which was in December.

But that increased cost alone wouldn’t explain a more than doubling of cost per lead. The conversion rate (percentage of leads over landing page view) had also more than cut in half, from about 7% to just over 3%.

So the ads, which had been consistently and predictably effective, suddenly weren’t working. One of the ads Meta leaned into most had over 6,000 comments on it. I checked those comments to make sure that there was nothing there that would kill conversion rate.

And that’s when I noticed something. Comments stopped. They normally came in at a dozen or so per day, but now there were no new comments at all in March.

Why?

The Weird Stuff

The ad itself shouldn’t suddenly stop performing unless something changed. Conversion rate should remain stable, but there are reasons competition costs would increase. The ad wasn’t converting and it wasn’t even getting engagement anymore.

And this change wasn’t gradual or the result of some trend. It went from peak performance to falling off a cliff overnight.

This suggested a shift with the ads themselves. First, I performed a breakdown by platform to get a sense of where the ad was being shown in January and February to set a baseline.

Prior to the drop, Facebook made up nearly 92% of the ad spend, followed by Instagram at under 5%.

The ad set was using Advantage+ Placements, so all placements were eligible. But the older demographics on Facebook were the primary audience, so this split made sense.

This shifted completely in March. Instagram now made up 86% of the spend, followed by Facebook at 8%.

That’s not a minor variance in distribution that can be attributed to randomness. That’s a complete structural shift. And that structural shift will also mean reaching a different demographic.

In January and February, all four of the most-used placements occurred on Facebook, with 64% going to the in-app Facebook Feed. That’s where those 6,000+ comments were happening.

But during the first five days of March, both of the top two placements were on Instagram. The top Facebook-related placement was Marketplace, and that didn’t happen until later on the 5th (more on that in a moment).

The weirdest part? The in-app Facebook Feed stopped getting any impressions at all. It went from getting 64% of the ad spend to zero.

That’s undeniably weird.

So the problem wasn’t just that distribution shifted from Facebook to Instagram generally. It was that the one placement that Meta leaned into the most when things were working well was no longer used at all.

The First Attempted Fix

The hope at this point is to shift delivery back to Facebook and away from Instagram. While the old approach to fixing this might be to remove Instagram entirely, my preference was to use value rules. This way, we could adjust Meta’s bids to hopefully shift how ads were delivered.

So during late afternoon of the 5th, a value rule was applied to lower the bid on Instagram placements by 50%.

This would hopefully shift budget back to Facebook without removing Instagram entirely.

Now, data at this point is messy because changes made were happening in the middle of days. But while the value rule “worked” by shifting ad spend back to Facebook, it had unintended consequences.

Instead of shifting ad spend back to the in-app Facebook Feed, Meta sent all of the impressions to Facebook Marketplace. Meta spent a total of $404 on in-app Facebook Marketplace in January and February combined. But in under a 24-hour period, Meta had already spent more than half of that ($222).

And budget still wasn’t getting spent on the in-app Facebook Feed.

The Weird Solution

At this point, something felt structurally wrong. Meta refused to give impressions to a placement that had driven the vast majority of conversions for months. A value rule didn’t fix that, and I had doubts that removing placements would either.

On one hand, it would be easy to panic. But knowing that the performance drop wasn’t due to something explainable was strangely reassuring.

The question would be how to fix something so randomly broken. Because this felt like a bug instead of something explainable, there were a couple of weird options that are intended for such weird situations.

These “solutions” shouldn’t be solutions at all. They don’t make sense. They aren’t based on anything logical. There’s no reason they should work. But when nothing else does and it seems like something is buggy, they’re worth trying.

It seemed as though either the ad set or primary ads were corrupted. Our initial thought was to create a new ad set and start over. But first, we’d try something far more basic.

We duplicated the ad that Meta had previously given so many impressions. But instead of using the existing post, some minor changes were made to it. And then it was published.

When I say that the impact was immediate, I’m not exaggerating. The ad was quickly approved, and the initial wave of impressions made it obvious: Impressions were now flowing back to the in-app Facebook Feed. Marketplace was no longer prioritized.

After a couple of days of this, the value rule was removed. And here were the results from March 9th through the 13th:

• 87% spent on Facebook, 10% on Instagram
• 66% spent on in-app Facebook Feed
• $46 cost per lead

The biggest relief may be that cost per lead not only returned to normal, but has actually been lower than January and February. And that’s even though the original ad with 6,000 comments is no longer getting shown, even though it hasn’t been turned off.

Approach Weird Solutions as Exceptions

I was hesitant to share this story because I know what tends to happen. Anyone who hasn’t seen good results recently will assume they are part of this glitch. “It happened to me, too!”

All indications are that this was an isolated situation. We haven’t seen this across accounts.

That doesn’t mean that other advertisers haven’t seen unexpected shifts in performance. But always start under the assumption that there’s a logical explanation. Resist labeling it “weird stuff” until you’ve not only exhausted those explanations, but you’ve pinpointed what makes it especially weird.

In this case, we had months of data that highlighted expected performance and distribution. That shifted completely on the 1st of March, which impacted results. And despite the use of value rules, nothing could force Meta to show ads on what were previously the primary placements.

That was weird. It felt like a bug. And because of that, a weird solution was attempted as a last resort. Luckily, it worked.

Your Turn

Have you run into similar weird problems that required weird solutions?

Let me know in the comments below!

The post When Weird Stuff Happens to Meta Ads appeared first on Jon Loomer Digital.

The potential power of AI for advertisers is real, so this isn’t meant as an indictment of AI. But I have one primary concern about AI when it comes to Meta advertising.

Let me explain…

Make More Bad Decisions

But I fear AI’s main application for advertisers, particularly early, will be to make more bad decisions more quickly. Whether they are creating dashboards, subscribing to random apps, or even leveraging an AI agent or LLM, there is real risk here.

If you’ve obsessed over the wrong metrics before, you’ll have AI to do it for you automatically. But you’ll have that AI create colorful dashboards so that it looks more important.

If you micromanaged ad results based on small sample sizes before, you’ll have AI highlight these differences for you now. You’ll just have access to more of these meaningless results more quickly.

Advertisers will have AI generate countless ad copy and creative variations to “win in the era of Andromeda.” But too many won’t check the quality of those assets.

When AI Will Be Useful

AI is only useful when it’s based on best practices and good advice. Otherwise, it can do a really good job of creating convincing output that won’t actually help you.

Automating bad strategies will only help you lose money faster. AI will not fill your knowledge gaps if it doesn’t first have a foundation in best practices.

There will be excellent applications of AI for advertisers, but we’ll need to sort through a whole lot of bad ones first. Everywhere you look, someone will have an “Amazing AI App” that looks impressive at first glance. The proof will be in the long-term, sustainable results.

AI can be useful when in the right hands. But it’s otherwise going to be a cool shiny object that results in more frustration. It will be our job to discern value from pretenders.

We’ve been talking a lot about AI in our weekly strategy sessions. Join us at www.jonloomer.com/elite.

The post Concerns About AI for Meta Advertisers appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A handful of smaller updates this week focused on cost, control, and the ways Meta continues to guide advertiser behavior. None require a full deep dive, but they’re all worth paying attention to.

1. Meta announces location fees
2. Advertisers pushed to schedule budget increases
3. Describe your audience with AI
4. New audience restriction flow
5. Breakdown by attribution limitations

Let’s get to it…

1. Meta Announces Location Fees

Advertisers running ads in certain European countries were greeted with an unpleasant surprise on Tuesday. Meta announced location fees, “due to the evolving regulatory landscape, including but not limited to digital services tax legislation.”

Here’s Meta’s explanation:

When your ad is delivered to an audience in a jurisdiction with location-specific fees, such as a digital service tax (DST), a location fee will be added to your bill. This fee is separate from your campaign budget and will appear as a distinct line item on your invoice or transaction statement.

So, these are fees that Meta had previously absorbed, but now they’re passing them on to advertisers. And that fee will be itemized, rather than baked into your overall costs.

Fees are based on the location of the user receiving the ad impression, not the location of the business or advertiser. And fees will only apply to impressions hitting users in specific countries.

Location fees will apply in the following locations:

• Austria: 5%
• France: 3%
• Italy: 3%
• Spain: 3%
• Türkiye: 5%
• United Kingdom: 2%

If you spend $100 on ads to Austria, your total bill will be $105 ($100 + 5% location fee). Not a huge difference, but no one wants to give away money, especially as advertising in Europe gets increasingly difficult.

Users in Europe can already opt into less-personalized ads, which likely makes advertising less effective in the region. And now advertisers will need to spend more to get those results.

While there may be multiple reasons for this move, it could put consumer pressure on the regulatory commissions that are ultimately the source of those fees. Or at least, Meta likely hopes that advertisers see it that way.

2. Meta Wants Me to Schedule a Budget Increase

When I create a campaign with a $100 daily budget, I get this message…

It reads:

Schedule a budget increase and take advantage of higher opportunity for similar costs
​
Our model has identified 5 days with higher opportunity. Increasing your budget to $130.00 may help you achieve additional results at a better cost per result.

That’s interesting enough. So apparently Meta has determined that I can get better results by taking advantage of opportunities that are higher than normal on five days of the week.

When I hover over the tooltip, I get additional details…

In short, this is a high opportunity period, and Meta’s models think I can get a lower cost per result if I take advantage of it.

If I click the button to increase budget, Meta completes the schedule for me.

Meta fills out the schedule to increase my daily budget by $30 from March 14 through March 18. These are apparently my “high opportunity” days. I’m not sure why the increases are scheduled from 12:30 am to 11:30 pm, but I guess that’s what the models want.

For the record, it doesn’t matter what I set my original budget at. Meta is going to detect high opportunities to increase my budget. For fun, I tried a $1,000 daily budget. I bet you can guess what happened next…

I’m not sure what to make of this. It would be helpful if Meta provided more details about what the models are seeing. Otherwise, this feels like a transparent money grab.

3. Detailed Targeting AI Recommendations

While I don’t seem to have this in any accounts, I’ve seen a few advertisers share it now. It’s not clear if it’s a test or the start of a rollout.

If you have it, you’ll see a “Describe your audience” option under detailed targeting. Here’s how it works…

1. Using natural language, describe the audience you’re hoping to reach in under 2,000 characters.

2. Meta will automatically interpret that text and suggest interests and behaviors that match your suggested audience.

3. Keep the suggested audiences or remove them manually.

This is one of those options that would be a whole lot more interesting several years ago, back when our detailed targeting inputs meant more. But they are only suggestions when using 11 of the most popular performance goals (there’s no indication that this would change that).

But this does simplify and automate what can otherwise be a bumbling exercise of trial and error. Again, assuming you actually provide detailed targeting inputs these days, it’s helpful. Especially when using a performance goal where you can restrict by detailed targeting.

Personally, I find it odd that we’re seeing this at all given the direction we’re heading with algorithmic targeting. Does it mean that such suggestions could have more impact than Meta’s been letting on? Otherwise, Meta’s been doing all it can to de-emphasize audience suggestions.

But again, we don’t know whether this is simply a test or an actual rollout at this point. Stay tuned.

4. New Audience Restriction Flow

Some ad accounts are seeing this message after clicking the link to “further limit the reach of my ads.”

This is messaging that Meta has played with a lot over the years. They’ll display various warnings and even highlight stats to suggest that restricting your audience will hurt results. But this is different.

In addition to the seemingly ridiculous options of sticking with the current setting or restricting your audience and increasing costs, Meta provides one more option: Apply value rules with Advantage+ on.

I’ve been pushing value rules pretty hard the past several months, but I’ve been disappointed that Meta hasn’t done more to explain why they’re useful. This is actually a really good start.

While some advertisers surely restrict their audience when it’s unnecessary, there is another segment where there’s a clear problem to be solved. This is often due to low-priced, low-quality optimized actions.

For example, I’ve often talked about how Meta will spend a high percentage of my budget on people aged 65 and up when optimizing for a lead or registration. The reason for this is simple: Meta can get very cheap leads that way, and my goal is to get as many leads as possible.

But quality matters, too, and that demographic does not get me high-quality leads. In the past, I’d restrict my audience by age group. But value rules are a better solution.

With the help of value rules, you can increase or decrease your bid for a specific demographic. In my case, I lowered the bid on people aged 65 and up.

I’m glad that Meta is making this a clearer option here. You should rarely need to restrict by age or gender. If there’s a problem to be solved, Meta wants you to know that you can keep Advantage+ on and use a value rule to fix it instead.

5. Breakdown by Attribution Limitations

I’m a big fan of Meta’s new breakdown by attribution (setting and count) because it lets advertisers get a clear view of how their results are distributed.

It’s great for your optimized conversion event that is the focus of the Results column.

But it’s not so great for any other conversion event.

Why isn’t this data available when using breakdown by attribution? We know it exists because it appears when using the Compare Attribution Settings feature.

Maybe it was an oversight. Maybe it’s a bug. Or maybe we’re seeing the early version of this breakdown while they work out the details.

Whatever the reason, I hope that these breakdowns will be built out further to be even more helpful.

More to Come

I’ll keep sharing observations like this every week, as long as Meta keeps rolling out things that are interesting, confusing, or both.

The post Meta Announces Location Fees (Plus 4 Updates) appeared first on Jon Loomer Digital.

Attribution is how Meta gives credit to an ad for a conversion. While that’s generally what it is, there are several layers of attribution to understand if you’re going to truly know how it works.

When you create an ad set that utilizes the Website conversion location and performance goal to maximize the number of conversions, the default attribution setting is 7-day click-through, 1-day engage-through (changed from engaged-view), and 1-day view-through.

But that’s not it. The default attribution model is Standard…

…and the default conversion count is All Conversions.

That’s a lot, and these layers make attribution far more complicated now than it used to be. Once you thoroughly understand these layers, you can intelligently evaluate results and make better optimization decisions.

In this post, I’ll help you understand each of these elements of attribution as they pertain to Meta advertising and how to approach them in 2026.

1. Click-Through Attribution

Definition

Click-through is when someone clicks your ad link and converts within the period of time that you select. Meta learns from these results and will show your ads to people likely to click through within the same period of time.

Background

Meta changed the definition of click-through attribution in March of 2026. Prior to this change, Meta considered any click on your ad prior to converting under click-through attribution. The old definition included social clicks and other clicks on ad elements other than links.

What Are Link Clicks?

It would be easy to assume that a “link click” and an “outbound click” are the same thing. While an outbound click is a type of link click, a link click isn’t always an outbound click.

Meta’s definition of link clicks includes clicks on links to websites. But it also includes several other actions like click to call, click to message, lead forms, and more.

While it may seem like a long list of potential loopholes, these “other” link clicks should be exceedingly rare when using the Website conversion location and the performance goals to maximize the number or value of conversions.

Conversion Windows

The default click-through conversion window is 7 days. In other words, Meta will count a conversion if it happens within seven days of clicking your ad.

Advertisers have the option of 7 or 1-day click-through attribution in the ad set. This will impact both how results are reported by default as well as how ads are delivered.

Years ago, the default click-through attribution window was 28 days. While it’s no longer available as an optimization element, advertisers can use the Compare Attribution Settings feature to view conversions that happened within the 28-day click-through window.

How to Approach

If your goal conversion event is a purchase, it’s generally recommended to stick with the default 7-day click-through attribution window.

If offering something for free that doesn’t require deliberation, 1-day click-through is recommended to better reflect the typical customer journey. Though the impact on volume when switching from 7-day click to 1-day click should be considered.

2. Engage-Through Attribution

Definition

Meta counts an engage-through conversion when someone converts after clicking anything other than a link on your ad (likes, reactions, comments, shares, saves, and all “other” non-link clicks). It also includes engaged-views.

Background

Engage-through attribution replaced engage-view attribution when Meta made changes to click-through attribution in March of 2026. Prior to this change, the previously named engaged-view attribution only included the situation where someone watched a video for at least five seconds and then converted within a day.

Meta updated click-through attribution to only include conversions that happened after a click on an ad link. Conversions that resulted from social clicks and all other clicks not included under click-through attribution moved to the new engage-through attribution. This new attribution still includes engaged-views.

Conversion Windows

Engage-through attribution is on by default with a 1-day conversion window. Advertisers have the option of turning engage-through attribution off by selecting “none.”

The screenshot above was taken after engage-through attribution was announced, but prior to the completed rollout.

How to Approach

When the goal conversion event is a purchase, it’s recommended to keep the default 1-day engage-through attribution on. This action reflects interest and awareness. Even if the eventual purchase was driven by another channel, the initial click shows that the ad had impact.

On the other hand, it may not make sense to keep 1-day engage-through on when optimizing for any other type of event, like a lead. If the person didn’t click through to get the free thing, your ad’s impact is debatable. You may also want to remove engage-through when restricting by a remarketing audience.

3. View-Through Attribution

Definition

Meta counts a view-through conversion when someone converts within a day of being served an impression. This assumes they never clicked the ad, otherwise it would fall under either click-through or engage-through attribution.

Background

View-through is the most controversial of the attribution settings, and it leads to the most confusion. It’s also the most likely source of inflated results.

While click-through attribution reflects intent and engage-through suggests interest, the value of view-through results is cloudy. While view-through conversions can suggest that your ad impacted a customer decision, we also can’t prove that the impression was even seen.

The most common scenario where this comes up is via remarketing. Someone who is on a customer email list or who engages with a brand on other channels is served an ad that they may not have seen. Within a day, they act on another marketing effort and convert.

Did the ad have an impact? In some cases, even if the impact is less than if a click were made. But in others, that impression may have had no impact at all.

Conversion Windows

View-through attribution is on by default with a 1-day conversion window. This can be turned off by selecting “none.”

How to Approach

The most common example of a view-through conversion that has value is when someone is shown your ad for a product, the person doesn’t click, but it inspires them to later Google your business or product to complete the purchase. While your ad doesn’t deserve full credit, it positively impacted results.

But such an example is limited to purchases and would be rare for any other event. Might someone be interested in your free lead magnet, not click, and then Google it later to complete the form? Sure. But, it’s not normal behavior.

Consider removing 1-day view-through attribution for any non-purchase event. And since inflated results are most often caused by this type of conversion when remarketing, remove view-through attribution in that case.

4. Attribution Model

Attribution Model is set to “Standard” by default, but you do have another option…

Definitions

Standard: Optimize ad delivery for a selected time window and engagement type based on your attribution settings. For example, 7-day click, 1-day engage-through, and 1-day view-through.

Incremental: Optimize delivery for conversions using models that predict whether a conversion is caused by an ad.

Background

Incremental attribution rolled out in 2025. When selected, you lose the ability to edit attribution settings. That makes sense when you understand how each works.

Ultimately, standard attribution doesn’t consider whether someone saw your ad or was influenced by it at all. Sure, that’s difficult to avoid if you get a conversion following a click-through, but it’s certainly plausible that a view-through conversion wasn’t incremental.

The point of incremental attribution isn’t to eliminate view-through conversions. You’ll actually see view-through results when using incremental attribution. The hope is that your results will be limited to those that were actually influenced by your ads.

How to Approach

In theory, incremental attribution should lead to a more accurate reflection of the impact of your ads. And, in theory, it should lead to higher quality results, even if that’s at the expense of volume. And these are tradeoffs most advertisers are willing to make.

Personally, I’ve seen little difference between my standard and incremental results. This may be because I approach attribution settings differently depending on the conversion goal. I prevent Meta from leaning into view-through conversions when they’re otherwise likely to be a problem, for example.

Does it make sense to use incremental attribution and should you prefer it in most cases? Yes. But does it make the difference you’d expect it to make? I just haven’t seen it. And because of that, I’m not all that motivated one way or the other.

I’d consider incremental attribution the best option by default if you deal in high budgets and already have no problem getting high volumes of results. Otherwise, I wouldn’t recommend intentionally limiting results if you already struggle to get out of the learning phase (exiting the learning phase is a general rule of thumb, not a requirement).

5. Conversion Count

Conversion Count is set to “All Conversions” by default, but you do have another option…

Definitions

All Conversions: Meta will count every conversion that happens after someone clicks or views your ad and takes an action within your attribution window.

First Conversion: Meta will only count the initial conversion that happens after someone clicks or views your ad and takes an action within your attribution window.

Background

Meta introduced first conversion in 2024, but only as a reporting option at the time. Meta eventually launched the ability to customize how your ad set is optimized by conversion count in late 2025.

The ability to separate “first” from “all conversions” is valuable because it highlights a common point of confusion for advertisers. Your default results can include multiple conversions by the same person.

For example, let’s assume your performance goal is to maximize the number of conversions where the conversion event is a purchase and you’re using the 7-day click-through attribution setting. Someone clicks on your ad and immediately goes to your website and makes a purchase. They then decide to make another purchase five days later (the product doesn’t matter). Both purchases would be reported when using “all conversions.”

But when using first conversion, Meta would only report the first purchase. This would impact both the raw number of results and all value-related data — the value would reflect the value of the first product, not the second.

Refining your results to only the first conversion can also help clear up reporting conflicts. We may assume that Ads Manager is inflated because it doesn’t line up with back-end or third-party reporting. But Ads Manager results can include multiple conversions by the same person.

How to Approach

While I’m a big fan of first conversion for adding context to results, I’m not sure I understand the argument behind customizing conversion count to impact optimization in the ad set. It feels counterintuitive.

By changing the conversion count in the ad set, Meta will only report on the first conversion and make decisions based on that. Meta will not value a customer who makes two purchases more than a customer who makes one. When using first conversion, Meta sees both customers as equals.

But don’t we prefer the scenario where a customer makes multiple conversions? Don’t we want Meta to learn from that? And don’t we want to see that in our results?

The use cases I can think of aren’t particularly strong. Maybe if you’re optimizing for leads and you don’t want to reward a situation where someone completes your form twice in a few days. Or maybe you’re optimizing for an engagement-related website event and you don’t want Meta to lean into inflated results of people performing actions over and over.

I guess it wouldn’t hurt in either case to limit conversion count to first in the ad set. But it’s a limited scope.

General Approach to Attribution

Hopefully this post provides a basic understanding of all of the attribution-related elements so that you can make informed decisions.

The main thing I’d say is that you should normally use the default settings unless there’s a specific problem to be solved. This applies to attribution just as it does to targeting, placements, creative enhancements, and just about anything else.

The default settings are going to be better for the majority of situations, which is why they’re the defaults in the first place. Know what you’re doing and why when making customizations.

Your Turn

Do you have any questions about attribution that aren’t answered here?

Let me know in the comments below!

The post How Meta Ads Attribution Works in 2026 appeared first on Jon Loomer Digital.

Meta made several big updates to attribution recently. The primary change is that they finally updated click-through attribution to require a link click. This could impact some advertisers more than others.

While requiring a link click to qualify for click-through attribution may seem obvious, it wasn’t previously the case. Until now, click-through attribution reflected conversions that happened after any click on your ad.

But social clicks including likes, comments, shares, and saves will no longer be included in click-through attribution. Those will be moving to a new engage-through attribution.

The Lost Conversions

But there’s an important caveat. Click-through attribution uses a default 7-day window while engage-through is only 1-day. That means that any conversions from social clicks that happened beyond a day will no longer be reported.

This is a good thing since that type of conversion isn’t all that different from a view-through conversion. But, unlike view-through conversions, these conversions were previously getting the benefit of a 7-day window.

While I doubt this will noticeably impact results for most advertisers, there is an exception. If you lean heavily into remarketing, you could see your reported conversions drop.

One of the reasons advertisers swear by remarketing strategies is due to the misleading results. While they’ll continue get the benefit of view-through conversions, losing these numbers could start altering perceptions.

The post Conversions Lost Due to Attribution Changes appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Weekly observations from inside Meta ads

A handful of attribution updates, a short Ads Manager outage, rumors about monthly invoicing, and a few observations about how Meta is leaning further into creative diversification. None of these require a full deep dive, but each one says something about where things are heading.

1. Will Remarketing Take the Biggest Hit?
2. How to Approach Engage-Through
3. Breakdown by Attribution is Back
4. Tuesday Outage
5. Related Media Observations
6. What is Monthly Invoicing?

Let’s get to it…

1. Will Remarketing Take the Biggest Hit?

Ever since Meta announced changes to attribution, I’ve been thinking about how this could impact results. There’s one strategy I keep coming back to as the most likely to see a substantive drop in results: Remarketing.

First, let’s summarize the changes…

Click-through attribution now requires a click on a link.

The previous definition of click-through attribution required any click on your ad prior to converting. The change eliminated situations where the only clicks were social (like/reaction, comment, share, save) or other non-link clicks.

Engage-through attribution replaces engaged-view.

The clicks that moved out of click-through attribution moved into the new engage-through attribution, which also includes engaged-view actions (someone watched at least 5 seconds of a video prior to converting).

But this means that some previously reported conversions will disappear. While all types of clicks that were previously included under click-through attribution are now covered in engage-through, there’s a major caveat: The window has changed from 7 days to 1 day. This means that conversions that happened within 2-7 days following an engage-through action will no longer be counted.

What percentage of total conversions will be lost? It’s impossible to say. I assume the impact will be minimal, if not inconsequential, for most advertisers. But there’s a reason I think advertisers stuck on remarketing strategies will see the biggest change.

Let’s consider the most likely scenario where these kinds of conversions would happen…

1. Person sees your ad and likes, comments, shares, or saves it
2. They do not click the link, CTA, or thumbnail image to go to your website
3. They get a marketing message from you on another channel
4. 2-7 days after the initial ad was clicked, they convert

The scenario above is very similar to what we would normally consider a view-through conversion. In both cases, they didn’t click through to your website, but they ended up going back to your website later — usually driven by another channel. But the biggest difference here is that the social clicks were previously given 7 days to convert, not 1.

I often talk about the misleading results of remarketing, particularly related to Cost Per Result and ROAS. Those results tend to look far different when you remove the view-through conversions.

But we may not have considered the potentially bigger source of fluff results. These are essentially view-through conversions that were given a 7-day limit.

Of course, some level of remarketing happens naturally, so it’s entirely possible that all advertisers will see some impact from this. But if your entire strategy is built around remarketing because of the results, they’re likely to look less impressive soon. And the view-through conversions might not save you.

2. How to Approach Engage-Through

Because of this change, you’re likely to see far more results from engage-through attribution than you ever did from engaged-view. That assumes, of course, that you turn it on in the ad set.

This change will officially go into effect in March, and I don’t have it yet. But the default attribution setting will be 7-day click, 1-day engage-through (currently engaged-view), and 1-day view.

But there are reasons to consider changing from the default, and those changes certainly involve the new engage-through attribution setting.

If your goal event is a purchase, stick with the default 7-day click, 1-day engage-through, and 1-day view in most cases. View-through and engage-through conversions make sense when it requires a financial commitment. Someone may need to think about it longer or discuss with a partner. Then they’ll either Google the product or go directly to the website to convert.

But if you’re giving something away for free, this behavior makes much less sense. If it’s a lead magnet of any kind where you offer something of value in exchange for an email address, the behavior should be straightforward. If the ad inspires them to act, they’ll click the ad, go to the website, and complete the form. There are fewer reasons for deliberation.

So in those cases, I recommend using 1-day click only. You could conceivably use 7-day click, especially for a lead that may be to discuss a financial commitment. But I’d want Meta to focus on the clicks — and for my results to reflect that type of conversion.

Remarketing, of course, is where both engage-through and view-through could make up a very large percentage of your results. If you are showing ads to someone who also gets emails from you, a high number of view-through conversions would be natural. They may not have even seen the ad when it was displayed, and then they acted on an email that day.

A similar behavior can be true of the new engage-through conversion. You reach someone who receives emails from you, but they definitely see the ad. They may not even put much thought into what you’re promoting, but they like it to show support. But it didn’t inspire an action. Then they act on an email later.

As mentioned earlier, you should see fewer of these than before because of the 1-day window, but they’ll now stand out since they’re separated from clicks. And because of that, this type of fluff conversion will be easier to spot.

You want to eliminate those fluff conversions when Meta can lean heavily into them to get you results. If you at least limit conversions to click-through where they require a click on a link, the results should be more realistic.

Of course, I still think you should stop remarketing entirely, but you do you.

3. Breakdown by Attribution is Back

In a relevant development, the Breakdown by Attribution feature is back (at least for me). I had it long enough to write about it more than a month ago, but then it disappeared. It strangely reappeared while I was writing about Meta’s changes to attribution.

The most useful element of this breakdown given the latest changes will be to see how many of our reported conversions come from engage-through (when the change officially rolls out). This breakdown allows you to generate rows to see how conversions break out by each attribution type.

That assumes, of course, that you use the default attribution setting. Unlike the Compare Attribution Settings feature, which can uncover conversions that weren’t reported by default, this breakdown will only segment results based on your attribution settings.

Come back to this feature once the attribution changes are complete. I’ll be especially interested in breaking down remarketing results.

4. Tuesday Outage

I hope you weren’t wanting to create any new ads on Tuesday because it was an absolute blood bath. I obsessively check Ads Manager throughout the day, and I noticed everything was down around mid-afternoon. When that happens, I know my next step: Check metastatus.com.

Yep, there was a known outage. “High disruptions” impacting Ads Manager.

These high disruptions lasted a couple of hours, and then there were a couple of moderate blips after. But it wasn’t pretty.

It’s a good time to remind you that if something like this happens to you, use Metastatus to see if there’s a widespread problem. You can save a lot of headaches if you’re at least able to verify that it’s nothing on your end.

5. Related Media Observations

When you create an ad, Meta will recommend adding Related Media. This is creative that Meta detects you’ve used in other ads to promote the same thing.

When selected, Meta will include that media as options for your ad. In other words, Meta may replace the primary media you provided with Related Media.

Something I’ve noticed is how well this works with my approach to “stack” creative diversity in phases. I won’t create a bunch of ads at once. I’ll create a couple at a time, based on a unique theme. I’ll learn from those ads and then create new ads later based on a uniquely different visual and theme.

The thing is, you won’t get Related Media recommended if you create all of your ads at once. But more and more will be recommended the longer you go. In the example above, these are creative I used during the past few months.

My point is that eventually I can create one monster ad that makes use of all creative I’ve used to promote the same product. Meta can then mix and match to find the creative that works best for the right people and in the right placement. I might even be able to turn the old ads off (or I wouldn’t be surprised if Meta stopped showing the old ones that become overlap).

As I’ve thought about this more, I have a couple of primary observations…

It’s well past time that we stop trying to find “winning” creative.

I had already been saying this for a while now. The point of creative diversification is to generate lots of options — creative and text — for Meta to help find the right combination for the right person. Once you include Related Media, this becomes all the more obvious.

When discussing the Breakdown Effect, Meta stresses how important it is not to micromanage results. Instead of focusing on the performance of individual ads, we should evaluate results in aggregate. We can drill down even lower. If you use multiple media for one ad, worry less about the performance of each creative and more about the ad itself — if we even care about that.

Meta really leans into Related Media.

I hadn’t used the Breakdown by Related Media in a while, but I guess I also hadn’t used Related Media all that much until recently. And, wow, what it reveals is pretty interesting.

When you use this breakdown, Meta will generate separate rows (for a single campaign, ad set or ad) for your original media and Related Media. Take a look at this breakdown for an ad where I experimented with Related Media recently…

Meta spent six times more on Related Media than on the original media. If your instant reaction is “Oh, no!” you’re looking at this all wrong.

Do you know why Meta spent six times more on Related Media? Because the results were better. And that’s good validation for providing creative variation.

6. What is Monthly Invoicing?

I’ve seen some rumors recently that all Meta advertisers will be forced to use monthly invoicing instead of paying for ads with credit cards. I’ve been digging into the source and accuracy of the rumor, and here’s what I’ve been able to figure out so far…

Monthly invoicing is a billing option for eligible businesses that replaces Meta’s standard threshold-based charging with a single monthly invoice. You’re given a credit limit, spend accumulates throughout the month, and you have 30 days to pay once the invoice is issued. If you hit your limit before the month ends, ads pause until payment is made.

Unlike the standard payment system, you connect your bank account to pay these invoices, rather than use a credit card. And that, I believe, is where the panic begins.

First, it doesn’t appear that all advertisers are being forced into monthly invoices. It’s an option that Meta lists as a benefit. Though Meta’s documentation does indicate that “some” advertisers are being forced into monthly invoicing.

What’s not clear is what the circumstances are for an ad account to be forced into this. Is it due to high ad spend? Or is it the opposite, because of an account that lacks history?

One concern I’ve heard from advertisers is that forcing them into monthly invoicing would be problematic for situations where credit cards are used for points and other benefits. That element would be lost.

I’ve also seen the rumor that this is Meta’s way of taking more money on financing and interest, but that doesn’t seem to be true either. Meta says that “Your credit limit is not a form of credit or financing.”

I haven’t heard anything directly from Meta on this, but what appears in their official documentation conflicts with many of the rumors I’ve seen. Are some advertisers being forced into monthly invoicing? Yes. The details behind that are murky.

More to Come

I’ll be sharing observations like this every week, as long as there’s something interesting or confusing to write about.

The post Attribution Changes Could Hit Remarketing Hardest (Plus 5 Updates) appeared first on Jon Loomer Digital.

Conversion results in Ads Manager attributed to a click have historically included more layers than most advertisers understood. It led to confusion about what results meant. Thankfully, Meta is making an important change.

Click attribution will finally require a click on a link — no other clicks will be included. In addition to announcing this change today, “engaged-view” will be updated in name and substance to “engage-through.”

Here’s what you need to know…

The Change to Click-Through Attribution

Click-through attribution is how Meta gives credit for a conversion when someone converts after clicking an ad. While that may seem straight-forward, it’s not — and certainly wasn’t.

In fact, I misunderstood the underlying meaning for a decade. It wasn’t until I ran a test more than a year ago that I was able to confirm the truth behind click-through attribution: Meta counted a conversion when it happened following any click on your ad.

It didn’t need to be a click on a link. It could have been a like, share, comment, or other click. If someone clicked your ad and then converted within seven days, Meta counted it under click-through attribution.

I found this to be unnecessarily confusing, if not misleading. It was easy to assume that a link click would have been required. But part of the reason that this was easy to miss was that Meta’s old definition was vague.

Eventually, Meta updated the definition of click attribution to accurately reflect any click. But now, Meta is changing the definition of click attribution again.

From Meta:

“Going forward, we are changing the definition of click-through attribution for website and in-store conversions to exclusively include link clicks.”

This is good news. It should help align conversion results with expectations based on traffic driven from ads.

The New Engage-Through Attribution

Meta isn’t done making changes to attribution.

Conversions that resulted from reactions, comments, shares, saves, or other non-link click actions will be included in engaged-view attribution. To account for this change, Meta is renaming it “engage-through attribution.”

If you’re not familiar with the formerly named “engaged-view” attribution, it is part of the default attribution setting (7-day click, 1-day engaged-view, and 1-day view).

An engaged-view is counted when someone views your video for at least five seconds (formerly 10) and converts within a day. If the video is less than 5 seconds long, at least 97% of it needs to be watched.

So the new engage-through attribution will include the former engaged-view conversions, but also all conversions that resulted from clicks that weren’t link clicks. And unlike engaged-view, which applied only to videos, the new engage-through will apply to all formats.

What Are Link Clicks?

Some further clarification is required here about link clicks because it’s not quite as obvious as you may think.

Meta’s definition of link clicks includes clicks on links to websites. But it also includes several other clicks.

• Websites
• App stores or app deep links
• Click to call
• Click to message
• Maps/directions
• Shops
• Lead forms
• Marketplace
• Link clicks to Instagram profiles for ads promoting Instagram profile views
• Playable experiences
• Videos that launch the watch & browse experience
• Videos hosted by another website (including videos embedded in Feed ads but hosted on a video platform such as YouTube or Vimeo)

While this may seem like a long list of potential loopholes, these “other” link clicks should be exceedingly rare when using the Website conversion location and the performance goals to maximize the number or value of conversions. And that’s when limiting link clicks to those to an external website will be most important.

Still, it would have been better if Meta had gone all the way on this. Why not change the definition of click-through attribution to require a click to an outbound link?

Summary of Changes

Let’s simplify what happened…

1. Click-through attribution now requires a click on a link prior to a conversion.

2. Engage-through attribution replaces engaged-view.

3. Engage-through includes engaged-view and all other clicks that aren’t clicks on links.

Attribution is messy, so this remains somewhat complicated. But it’s a positive change.

Impact of Changes

In order to project the impact of changes, we’d first need to be able to isolate the conversions that were previously reported as click-through that would no longer qualify. Unfortunately, there’s no way to reliably do that.

But it may help to summarize what this behavior looks like. Someone would have clicked on your ad, but without clicking the CTA button or thumbnail if a static image link ad. The most likely types of clicks would have been a like, reaction, comment, share, save, or media expansion.

Of course, just clicking those things without clicking the link wouldn’t have been enough. They also would have needed to go to your website (via another method or later on) and convert within seven days. This behavior is similar to that of a view-through conversion, which is when it would be most common. And especially if you reach someone via a remarketing audience (intentionally or algorithmically) and they are driven to the conversion from another channel.

Will it happen? Yes. But how often is tough to say, and it is likely to happen more often when restricting your audience to remarketing. Especially since the default click-through attribution window is 7 days.

Something to remember is that some of these conversions will move to engage-through, but not all. If someone performed a click that now qualifies as engage-through and converts within 2-7 days, it will no longer be counted. Engage-through, like view-through, only has a 1-day window. You may lose some of those conversions.

My guess is that any impact on total results will be subtle, so most won’t notice it. You may be able to spot changes if you use the Compare Attribution Settings feature or Breakdown by Attribution to see how click-through and engage-through numbers evolve. But it may not be more than a loose correlation to these changes.

But it’s those who focus on remarketing strategies who are most likely to notice a difference. This is when view-through conversions often prop up results. This type of conversion that is moving to engage-through is very similar. And since you’ll lose the engage-through conversions that happened beyond the first day — which were previously reported under click-through — a drop is possible.

The other group of advertisers who may see an impact will be those who regularly turned off engaged-view, and now engage-through. They won’t have the safety net of some conversions moving from click-through to engage-through. They’ll just disappear entirely.

Your Turn

What do you think of this change?

Let me know in the comments below!

The post Click-Through Attribution Now Requires a Link Click appeared first on Jon Loomer Digital.

Ignore most of the nonsense about Andromeda. Stick to the facts…

What is Meta Andromeda?

Andromeda is responsible for ad retrieval, which is the first of many steps to determine who sees your ads. It was a necessary upgrade to hardware and software that couldn’t handle the current scale.

Ads are different now due to multiple text options, creative enhancements, related media, and so much more. Advertisers are generating thousands of combinations of copy and creative for a single ad.

Because of that, the ad retrieval step can handle more variations to help find the right audience. It’s why Meta recommends creative diversification.

More options give you more opportunities to reach your audience. Targeting is controlled with your ad variations.

But so many advertisers make Andromeda into more than it is. And they completely misunderstand creative diversification and what it means.

Free Mini-Course

I created a free mini-course to clear it up. It consists of seven lessons to walk you through what Andromeda is and how to approach creative diversification. Subscribe here.

The post Avoid Meta Andromeda Misinformation appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Wednesday observations from inside Meta ads

A mix of subtle language changes, a meaningful update to the Manus AI integration, a display bug that caused panic, and a few observations about how I’m evolving my own creative process. None of these require a full deep dive, but each one reflects something about where Meta is heading.

1. Meta changed wording about how they use suggestions
2. Manus AI Integration with Ads Manager
3. “Conversion event blocked” message in Ads Manager
4. Evolution of my process to stack creative diversity
5. Move to “Ad” from “Sponsored”
6. Advantage+ Creative Video Generation example

Let’s get to it…

1. Meta Changed Audience Suggestions Language

This is one of those changes that was easy to miss. It was so subtle that you may have noticed but didn’t think much about it. But language changes like this one aren’t accidental.

When Meta initially rolled out Advantage+ Audience in 2023, the concept of audience suggestions was new. Meta had already introduced audience expansion in prior years, but “suggestions” attempted to reposition what targeting inputs meant.

Pay close attention to how Meta defined how your suggestions were handled in those initial versions…

Meta said:

If you share an audience suggestion, we’ll prioritize audiences matching this profile before searching more widely.

A variation of this was also stated in Meta’s documentation about Advantage+ Audience…

In that explanation, Meta said that data like your prior conversions, pixel data, and prior interactions with your ads would be used to find your audience. But if you provided a suggestion, Meta “prioritizes audiences matching your suggestions, before searching more broadly.”

Well, that mention of “prioritizing” your suggestions seems to have disappeared. First, from Advantage+ Audience in the ad set…

Under “Suggest an audience,” Meta says:

We’ll also reach people beyond any custom audience, age, gender and detailed targeting settings you apply when it’s likely to improve performance.

No mention of prioritizing suggestions. Also notice how Meta de-emphasizes suggestions now (no more button).

The mention of prioritizing your suggestions also disappeared from Meta’s documentation on Advantage+ Audience.

This is a very loose commitment to your suggestions…

We’ll show ads to people matching your suggestion, and other audiences when it’s likely to improve performance.

That could be one impression. It’s not particularly difficult to commit to showing ads to people matching your suggestion, even if it’s by accident. And it’s also impossible to prove in most cases.

You may think this language change is too subtle to be meaningful. My response is that Meta doesn’t make language changes unless there’s a reason.

Combined with the removal of the Audience Suggestions button, this language change is consistent with my belief that Meta could (and possibly should) eliminate suggestions entirely at some point. It’s clear that Meta doesn’t want you to think suggestions are all that important now.

2. Manus AI Integration with Ads Manager

Last week, advertisers started seeing mentions of Manus AI, Meta’s acquisition in late 2025, in Ads Manager. Based on my initial findings, this was little more than promotional links to Meta’s new AI agent. But whether I was just missing it or it happened since then, I am now seeing where the integration can happen.

First, a reminder. You can click the link to Manus AI in the Tools menu of Business Suite.

You’ll be redirected to the Manus AI interface, which looks similar to ChatGPT and Claude.

You may need to first create an account. While it initially appeared you’d need to sign up for a paid Manus subscription, that doesn’t seem to be necessary now. You’ll have limited credits if you use Manus 1.6 Lite, but you will be able to use it.

Then you’ll see a button to “Connect your tools to Manus.”

Right at the top, you should see a new connector for Meta Ads Manager.

You’ll see this…

It reads that you can “Connect Meta Ads to Manus to analyze performance, get optimization recommendations in natural language, and automate reporting.” Click the button to “connect.”

Once connected, you’ll see this…

You can click one of the example prompts or “Try it out.” Then it will prefill a prompt to show you what is possible with this connection.

After selecting ad accounts, Manus generates a “Meta Ads Manager Connector Capabilities Brief.” Here are the primary capabilities…

At the bottom is this message: “Great Meta Ads breakdown! Turn this into a live dashboard to monitor campaigns?” And a button to create a website.

That’s where this falls over my pay grade and I start to get nervous. I’ve built some dashboards during the past month using Claude Code, and this is similar. If you have expertise in that area, you may be more eager than I was to keep digging.

But you’ll be able to create custom dashboards of your advertising. Personally, I’d rather see Meta integrate something like this into Ads Manager than force us to build stuff ourselves (and run into extra costs). Still, there is promise here.

3. “Conversion Event Blocked” Message in Ads Manager

If you saw this “Conversion Event Blocked” message in Ads Manager recently like I did, you probably had a temporary panic attack. Or maybe that was just me.

It reads:

One or more conversion events are blocked because they suggest the use of information not allowed under Meta’s terms.

It was probably good that this was reported by someone in my community first. So when I saw it, I was immediately skeptical.

If you ever see a message like this, be sure to click the “View Details” button. That will take you to your blocked events in Events Manager.

It’s not uncommon to have blocked events. The question is whether Meta blocked those events or you did. In my case, I have several events I’ve blocked for various reasons.

I have 55 blocked events in all.

But I blocked all of those events myself. When I filter to view the events blocked by Meta, nothing appears.

So once I saw that Meta hadn’t blocked any of my events, the odds that this was a display bug increased significantly. I headed back to Ads Manager and took a closer look at the events dropdown menu when using the “Maximize number of conversions” performance goal.

There was no indication that an event wasn’t selectable because it had been blocked. And since it seemed many others were having the same issue, it was obviously a bug. And it cleared up within the day.

My theory is that this message was appearing for all ad accounts when an event had been blocked, regardless of whether it was blocked by the advertiser or by Meta. Or maybe it appeared for everyone, I don’t know.

I realize it’s easy to freak out when you see stuff like this, but always remember the likely possibility of bugs. In most cases, it’s nothing.

4. Evolution of My Process to Stack Creative Diversity

When Andromeda became all the rage last year and Meta started recommending creative diversity, I reacted the way many advertisers did: I overreacted.

I created ad sets with 20, 30, and even 50 ads, regardless of the budget. I wanted to push this new Andromeda to the limit. And maybe, we thought, more ads would mean better results.

But most of us quickly realized a couple of things. First, more ads does not mean better results. In fact, you’ll likely cut corners in an effort to create more ads, so you’ll create worse ads. Most importantly, creating a whole bunch of ads is exhausting, especially when you don’t get the payoff.

As a result, my approach to creative diversity has steadily evolved since those early days. I started focusing more on creating ads “in the spirit of diversity.” In other words, a single ad might leverage multiple formats, Advantage+ Creative enhancements, and the maximum number of text options.

How I do things now is much more deliberate. Here’s how it goes…

1. Create two ads around a single theme.

These two ads share a visual theme and text approach. The difference is that one ad is a static image and one is a video.

Here’s an example promoting my Modern Targeting mini-course. Both have similar visual branding, but the presentation is different.

I like to start new ads with a creative test, just so I know what will happen if the ads get the same (or similar) distribution.

2. Create two ads around a completely different theme.

After the test is complete, I might let the ads run naturally for a while. Once things have stabilized, I create new ads if I feel it’s necessary. No matter how well the first two ads are doing, I’ll tend to at least create one more batch to add some variation.

So this second set of ads will look completely different from the first set. The text that the two new ads share will also be completely unique from the first.

My first two ads utilized a dark gray theme, so I went with something more colorful this time. The first video was a slide show of the lessons, where this was more of an animation of the ad creative.

Once again, I used a creative test in the existing ad set to introduce these two ads. I know that this isn’t for everyone, but it at least guarantees (or should guarantee) impressions during the test. There’s nothing more annoying than spending a bunch of time on new ads, only to see them get zero visibility.

3. Utilize opportunities for further diversification.

For all of these ads, I embrace the options that will increase creative diversity. That means accepting things like Advantage+ Creative enhancements, AI-generated images or backgrounds, Format Display Options, and AI-generated text when possible. I don’t do so blindly, but I accept what I can.

I also customize my creative by placement to be sure there are multiple options of each creative that will appear differently depending on the placement.

4. Add more as needed.

Four ads should be enough for an initial push. One of the main things I’m trying to avoid is creative fatigue. But I also want to generate multiple variations so that Meta can show the right combination to the right group of people.

I don’t micromanage my results by ad. I evaluate the ad set in aggregate, knowing that what is being shown today may not be what Meta chooses to show tomorrow or next week. What works may also shift.

But once it seems like the ads are getting stale or results aren’t consistently solid anymore, I’ll revisit things and publish two new ads to the ad set.

How you approach this is up to you. I prefer this slower approach since I can also learn from the results early and apply these lessons to later ads. It also allows me to think through what I want to do next, rather than feeling rushed into creative directions.

5. Move to “Ad” from “Sponsored”

I can’t say for sure when this happened, but I believe it was during the past couple of months. Meta seems to be transitioning from the “Sponsored” label to “Ad.”

The ads I see on Facebook still read “Sponsored.”

Ads on Instagram are getting the “Ad” label.

For some reason, I’m not currently seeing any ads on Threads. But that was the first place I spotted this.

Meanwhile, Meta has transitioned to using the “Ad” label for nearly all placements in the advanced preview when creating an ad.

I know. Big deal, right? But my theory is that “Ad” is much tougher to spot than “Sponsored” because it’s only two letters. Maybe it will make an incremental difference.

6. Advantage+ Creative Video Generation Example

A few weeks ago, I shared that I had access to the new Advantage+ Creative video generation feature. I planned to spend a lot of time on it so that I could provide some thoughts and examples.

Unfortunately, I no longer have that feature. But I assume I saw it in the middle of a test, and I’ll likely see it again. While I had hoped to share some polished examples, the only example I have isn’t all that great. But it at least gives you an idea of what it can do.

Meta generates video concepts that are built around your ad copy and creative. If there isn’t much creative to go off of, Meta will add in scenes that use what looks like AI-generated stock imagery. Once you approve it, Meta will turn those scenes into a moving video.

In my case, my ad lacked interesting visuals. But you can also add images that Meta can use to generate concepts and videos based on your product or business. So I added several portrait photos of myself, and this is what Meta generated…

It’s not amazing. I wish I could have had more time with it to show you something better. But no matter what you think of this, we’ve come pretty darn far in a year or two. This stuff is only going to get better.

More to Come

I’ll be sharing observations like this every Wednesday, as long as Meta keeps giving me things to write about.

The post Meta No Longer Prioritizes Suggestions (Plus Manus AI and 4 Updates) appeared first on Jon Loomer Digital.

Last week, I walked through all of the various scenarios where your audience inputs are seen as suggestions. Today, let’s discuss when you might restrict your audience and the recommended ways to do that.

I recommend a very hands-off approach to targeting these days. I rarely restrict my audience. When possible, I use other solutions.

The reason to restrict your audience will typically fall into one of two categories:

1. Legal requirements: You are not legally able to sell to certain people.

2. Performance-related: You need to solve a problem of low-quality, optimized results which are tied to the selected performance goal.

Let’s run through all of the various targeting inputs and the considerations related to when and how you might restrict your audience…

Age Range

The default age settings are used as suggestions only. But there are times when it may seem reasonable, if not best practice, to restrict targeting by age.

Performance-Related Restrictions

The danger of age restrictions is that an advertiser will consider their ideal customer and prevent Meta from showing ads to anyone outside of that customer persona. This actually looks at age restrictions backwards, assuming Meta will waste your money on a group of people who won’t convert.

Instead, the reason age-related restrictions might be required for performance reasons is that Meta can get cheap and low-quality optimized actions from a specific age group. Since the algorithm is literal, the only focus is getting you as many optimized events as possible, as defined by your performance goal.

An example I often use that reflects this problem is related to leads. I found I was getting cheap leads, but I was suspicious about the quality of them. So I performed a breakdown by age and discovered that I was spending a very high percentage of my budget on people 65 and up. Meta found that I could get cheap leads that way, so a weakness was exploited to get me more optimized actions.

So it’s important to remember that the need for age-related restrictions will be in response to a discovery that Meta is wasting your budget on an age group that is resulting in low-quality, optimized results. Meta will rarely waste significant budget (for a prolonged time, at least) on an age group that doesn’t perform.

Address with Value Rules

In the past, I’d recommend restricting by age to address this situation. But a full restriction is rarely necessary now. You can use value rules instead.

Let’s revisit my issue with low-quality leads. The problem was that if I restricted by age and eliminated people over 65, Meta would move that budget to people 55 to 64. And while I’m fine reaching either group, I also don’t want to spend a majority of my budget there.

I used value rules to bid 50% less on people over 65 and 20% less on people aged 55 to 64.

The result was that I could limit my spend on people aged 55 and up without eliminating them entirely.

Legal-Related Restrictions

Another reason you might want to restrict by age is that you sell an age-restricted product. The obvious example would be alcohol, but you also may offer something that is only available to seniors.

Either situation can be addressed within the ad set, assuming you’re not dealing with a special ad category.

You also might be able to apply an account-wide restriction. Within Advertising Settings, click on Account Controls.

Then select Audience Controls.

One of the options is “My business advertises age-restricted goods or services.” Turn it on, and you can set a minimum age to prevent your ads from being shown to younger people.

Of course, this account-wide restriction is only for setting a minimum between 18 and 25. You can’t set an account-wide maximum or set a minimum higher than 25.

Bottom Line

1. Do not restrict by age to reflect what you believe is your ideal customer.

2. If Meta is wasting your money on an age group because they lead to a high volume of low-quality optimized actions, use value rules instead of restricting by age.

3. If you need to restrict by age due to legal reasons (you sell age-restricted goods), apply this restriction in the ad set or use Audience Controls in your Advertising Settings to restrict by age account-wide.

Gender

In many ways, gender is lumped with age range when discussing demographic restrictions. But how you approach gender restrictions is actually a bit different.

Performance-Related Restrictions

As with age range, the danger is that an advertiser will envision their ideal customer and then restrict by gender. But gender-related restrictions are rarely necessary.

The only time you should need to restrict by gender is because not doing so leads to a high volume of low-quality optimized actions. If your primary customer is women, Meta’s unlikely to waste a significant portion of your budget on men if you’re optimizing for purchases and men don’t buy. The bigger problem, sticking to this example, would be if men perform a high volume of cheap, low-quality optimized actions.

The most likely scenario when this might come up would be when optimizing for an engagement-level event (link clicks, landing page views, post engagement, video views, etc.), which eliminates the level of commitment required to perform the action that you want. If you want engagement, Meta won’t care if men aren’t likely to buy from you. If they’re likely to engage, men will be shown your ads.

Now I wouldn’t typically advise optimizing for engagement-level actions because the quality issue is impossible to eliminate. But restricting by gender, particularly if your customers are only one gender or another, may be a reasonable consideration in those cases.

Address with Value Rules

While it might be reasonable to restrict by gender in certain situations, I’d consider it a last resort. You can keep Advantage+ on and use value rules instead. If Meta’s spending too much of your budget on men, resulting in empty optimized actions, use value rules to lower the bid.

By lowering the bid by 90% for men, you’re likely to eliminate men from your ad spend entirely without needing to restrict your audience.

Bottom Line

1. Do not restrict by gender to reflect what you believe is your ideal customer.

2. If Meta is wasting your money on a gender because they lead to a high volume of low-quality optimized actions, use value rules instead of restricting by gender.

Detailed Targeting

The challenge here is that you can rarely restrict by detailed targeting (interests and behaviors) options now anyway. If you choose one of 11 performance goals, any detailed targeting inputs will be used as suggestions. This can’t be turned off.

So any exceptions to this, where restricting an audience by detailed targeting is possible, would need to fall within the performance goals not impacted.

Performance-Related Restrictions

If you are going to restrict by detailed targeting, it will need to be while using one of these performance goals:

• Daily unique reach
• Impressions
• 2-second continuous video plays
• ThruPlay views
• Ad recall lift
• Interactions
• Event responses
• Page likes
• Reach

These aren’t the most commonly used performance goals, and they all represent light-touch, surface-level actions. Which, if we’re to be honest, are also the times when you’d want to have some control over your targeting.

If you optimize for daily unique reach, impressions, or reach, Meta only cares about getting your ads shown. The rest require some sort of engagement. But the level of engagement is so basic, it’s difficult to avoid extremely low-quality results.

Restricting by detailed targeting could help in these cases. If you’re going to focus on getting thru-play views or interactions, it would be helpful if you can at least put some guardrails on who performs those actions.

That guarantees nothing, of course. It’s nearly impossible to avoid low-quality results from any of these performance goals. But if the option is to let Meta find these results wherever they can be found or restrict the pool to people by detailed targeting, I’d prefer the restrictions.

Actually, I’d prefer doing neither and optimize for a conversion of some kind. But if you forced me to use one of these performance goals, restrictions would be required.

Bottom Line

1. You cannot restrict by detailed targeting when using 11 of the most common performance goals.

2. If you’re using one of the remaining nine performance goals (rarely recommended), adding some guardrails with detailed targeting restrictions will help, though it won’t fully solve the likely problems you’ll have related to quality.

Lookalike Audiences

The approach to lookalike audiences, and the challenges associated with them, are similar to those of detailed targeting. If you choose one of nine performance goals, any lookalike audience you provide will be used as a suggestion. This can’t be turned off.

So any exceptions to this, where restricting an audience by lookalike audience is possible, would need to fall within the performance goals not impacted.

Performance-Related Restrictions

If you are going to restrict by lookalike audience, it will need to be when using one of these performance goals:

• Daily unique reach
• Impressions
• 2-second continuous video plays
• ThruPlay views
• Ad recall lift
• Interactions
• Event responses
• Page likes
• Reach
• Instagram profile and Facebook page visits
• Reminders set

Once again, these performance goals represent light-touch, surface-level actions. If you optimize for any of them, it’s difficult to avoid extremely low-quality results.

Restricting by lookalike audience could help. You’re not going to completely solve the problem of low-quality results with these types of optimization, but lookalike audiences and detailed targeting at least provide some basic guardrails.

Bottom Line

1. You cannot restrict by lookalike audience when using nine of the most common performance goals.

2. If you’re using one of the remaining 11 performance goals (rarely recommended), adding some guardrails with lookalike audience restrictions will help, though it won’t fully solve the likely problems you’ll have related to quality.

Custom Audiences (Remarketing)

If you want to target a specific group of people who are connected to your business (visited your website, on your email list, used your app, an existing customer, or engaged with your social content), you can restrict by custom audience.

Of course, the question is whether you should.

Remarketing Happens Automatically

A common mistake is assuming that Meta needs you to target custom audiences to focus on people most likely to convert. But assuming you’re passing events and creating custom audiences, Meta already knows about this activity and prioritizes them automatically.

You can prove this to yourself by thoroughly defining your audience segments and running Sales campaigns. Then you can use the breakdown by audience segments to see how much of your budget was spent on your engaged audience, existing customers, and new audience.

I do not restrict my targeting, yet I regularly see that anywhere from 10 to 25% of my budget is spent on remarketing. This happens naturally because Meta knows that people who visit my website, are on my email list, and have bought from me before are those who are likely to convert.

So you don’t need to restrict targeting to all of your website visitors, email list, or social media followers. That stuff happens already.

Other Remarketing Issues

There are other issues related to remarketing that are easy to miss.

First, the results are almost always misleading. Conversions from remarketing are almost always heavily inflated by 1-day view. In other words, your ad appeared for someone who made an action within a day — an action that likely would have happened anyway.

Second, remarketing results simply aren’t incremental. If someone clicked your ad and converted, you can make the argument that the ad contributed to that conversion. But if they regularly visit your website or are closely connected to you already, how much that ad contributed is debatable.

And finally, the great (and misleading) performance you receive from remarketing isn’t scalable. You’ll eventually run into a wall because it’s so easy to exhaust these small audiences.

Rare Exceptions

While I’m generally anti-remarketing, I also acknowledge the rare exception when it can make sense.

The type of scenario that could benefit from remarketing is where you are selling a high-priced product and want to help your other marketing efforts with ads. For example, you know that you’ll be sending emails and having your sales team reach out to a recent lead. You can run ads at the same time to keep your product top of mind.

Of course, doing so would be while understanding that any results you get in Ads Manager are a combined effort. Your ads were not fully responsible for the conversions — and they may not have done much of anything. But they can help in a situation when your emails and sales team could use a bit of reinforcement.

This makes most sense for high-priced upsells because any benefit is worthwhile. You won’t get that same benefit from trying to convince people to buy a low-priced product, and those ads will quickly get expensive.

Bottom Line

1. General remarketing is mostly unnecessary now because Meta prioritizes those closest to you automatically. This can be proven with Sales campaigns and audience segments.

2. Remarketing results are misleading, inflated by view-through conversions, not incremental, and not scalable.

3. A rare exception for remarketing can be to help with high-ticket upsells. In this case, you know that these people are also getting emails and calls from you, and the hope is that the ads can contribute to the overall messaging.

Summary of Audience Restrictions

I recommend prioritizing a hands-off approach to targeting these days. While there are some exceptions when you should take control, they are exceedingly rare.

Here’s a summary of the general approach that you should take to audience restrictions…

Age Range: Rarely necessary. Consider restricting when selling age-restricted goods. This can also be controlled at the account-level, though only up to a 25 year-old minimum. If age restrictions are necessary to solve a performance problem related to cheap and low-quality optimized results from an age group, use value rules instead.

Gender: Rarely necessary. Do not restrict by gender because you believe it represents your target customer. If gender restrictions are necessary to solve a performance problem related to cheap and low-quality optimized results from a gender, use value rules instead.

Detailed Targeting: Rarely possible since detailed targeting is only seen as a suggestion when used with 11 of the most common performance goals. But if using any of the other nine performance goals (not recommended), restrictions by detailed targeting would be a reasonable guardrail.

Lookalike Audiences: Rarely possible since lookalike audiences are only seen as suggestions when used with nine of the most common performance goals. But if using any of the other 11 performance goals (not recommended), restrictions by lookalike audiences would be a reasonable guardrail.

Custom Audiences: General remarketing is rarely necessary because it happens naturally. This can be proven with Sales campaigns and audience segments. Remarketing leads to misleading results and is neither incremental nor scalable. The rare exception when remarketing might make sense is when used in addition to other marketing efforts to encourage high ticket upsells.

Your Turn

Anything you’d add to this summary of how to approach audience restrictions?

Let me know in the comments below!

The post When to Restrict Your Audience in Meta Advertising appeared first on Jon Loomer Digital.

Nothing about targeting is the same as it once was. If you’re still trying to control targeting like it’s 2018 — or even 2021 — the changes have passed you by.

Consider just a few of the way things have changed…

Targeting Changes

When you provide detailed targeting inputs, they’re only seen as suggestions when used with 11 performance goals.

This expansion of how detailed targeting inputs are handled happened recently.

When you provide lookalike audiences, they’re only suggestions with 9 performance goals.

Meta made this update late last year. In each case, you can’t turn suggestions off. We don’t know how much your suggestions matter, if at all.

Age and gender inputs are seen as suggestions by default, but you can restrict both. Sure, you can, but you probably don’t need to. Even if there’s a problem to be solved, you should use value rules instead.

Some advertisers provide custom audiences as suggestions, thinking they’re remarketing. They’re not. But even if you restrict by those audiences, it’s rarely necessary. Meta prioritizes remarketing audiences automatically.

It doesn’t stop here. You’re probably getting location targeting and exclusions wrong, too.

What to Do

So, what should you do to catch up?

I built a free mini-course to help you master modern targeting. It consists of eight lessons to clarify best practices on all of these things. Subscribe to the Modern Targeting Mini-Course here.

The post Nothing About Targeting is the Same appeared first on Jon Loomer Digital.

Meta Advertiser Field Notes
Wednesday observations from inside Meta ads

A mix of clarification, deprecations, reporting quirks, and a few “what exactly is this?” moments this week. None of these updates require a full deep dive on their own, but each one impacts how we think about targeting, attribution, or the direction Meta is heading.

1. Manus AI Ads Manager integration
2. Exclusions aren’t suggestions
3. Modern Targeting mini-course
4. Final ASC deprecation schedule
5. Recurring Notifications ending
6. Conversions from other products
7. Meta Pixel Helper gets an update
8. Shops 3rd party reporting integration

Let’s get to it…

1. Manus AI Ads Manager Integration

If you’ve been on your Account Overview page lately, you may have noticed something new: “Automate your workflows with Manus AI.”

You may also see a new option for Manus AI in the Tools menu.

Meta acquired Manus AI at the end of last year. The focus of Manus AI is to develop “general-purpose AI agents designed to help users tackle research, automation, and complex tasks.”

Cool! Is Meta offering a direct integration with Manus AI to automate advertising workflows? Sounds promising!

If you click “Try it now,” you may be disappointed. This isn’t a direct integration, at least yet. Instead, you’ll be redirected to the Manus AI LLM interface.

But I’ve seen some examples of this page with example prompts for things like “Instagram Ads Generator” and “Product Image Enhancement.”

Here’s the prompt for the Instagram Ads Generator, based on a random pottery course scenario.

When I submit the prompt, I’m asked to log in. And once I log in, I quickly realize Manus AI is a paid product with a 7-day trial.

This feels like something Meta slapped together. There’s no integration at this point, from what I can tell. The Manus AI mention is little more than an ad promoting a Meta-owned product.

I read other reports (here and here) of an actual integration. So either I don’t have it yet, or the other reports out there are based on something more that isn’t available yet.

So, maybe this will eventually give advertisers the ability to seamlessly integrate AI agents to automate their advertising. And that may even be possible now, but it’s far from seamless. It’s confusing and sloppy.

Keep an eye on this.

2. Exclusions Aren’t Suggestions

I’ve realized lately that Meta’s created a mess related to targeting. Because terminology and functionality change so often, very few advertisers understand how their targeting inputs work. And when things frequently go the opposite way they expect, advertisers lose trust in the entire targeting process.

It’s why I wrote this week’s blog post regarding audience suggestions. More often than not, the targeting inputs advertisers provide are audience suggestions now. And when they don’t know that, that misunderstanding spills down to distrust in functions like exclusions.

I’ve heard multiple claims over the past several months that Meta ignores custom audience exclusions. In some cases, the theory is that Meta treats exclusions differently depending on whether you’re using Advantage+ or not. My interpretation of that claim is that your exclusion would be seen as a suggestion.

But it’s not. Exclusions (and Locations, for that matter) are part of Audience Controls. This section is separate from suggestions, and Meta will respect these inputs.

It seems that Meta is well aware of the conspiracy theories, though, because the description of how controls are handled in Ads Manager is different than I remember.

Meta says in the description of Controls that they “won’t reach people beyond these settings, even with Advantage+ on” (emphasis is mine). And the tooltip for excluded custom audiences reads that “any custom audiences added here will be excluded from the audience for this ad set.”

Even with that clarification, there will be advertisers who swear this is false. And the reason for this is the perfect combination to make conspiracy theories thrive.

First, you are much more likely to notice when you reach someone you think you shouldn’t than if you don’t reach someone you think you should. And second, custom audiences are generally incomplete and imperfect.

Let’s assume you are trying to exclude the members of your community. Meta will not be able to match up 100% of the users in a website custom audience to users who can be excluded. This is largely due to privacy and browser settings, but also due to a maximum time window of 180 days.

Okay, fine. So you add a data file custom audience of your members, too. Once again, Meta will not be able to match up 100% of that customer list to users. Your members may have provided different email addresses and phone numbers than Meta is able to attribute to them.

What I’m describing is precisely my experience. I gave up long ago on being able to exclude all of my members from seeing membership-related ads with 100% completeness. It’s frustrating. I try to make it as complete as possible. But it is what it is.

That doesn’t mean, of course, that the exclusions are suggestions and Meta ignores them. It just means that, no matter how hard you try, a person here or there is going to slip through.

3. Free Modern Targeting Mini-Course

This year, I’ve focused on doing something new. Every month or so, I will launch a new free mini-course on a popular topic. These mini-courses consolidate my recommendations into a small number of easily consumed lessons.

I started with Andromeda and Creative Diversification. I surveyed those who subscribed to that course, and they overwhelmingly wanted me to design my next course around modern targeting.

And that’s what I did. I created eight lessons that break down each category of targeting (age, gender, location, exclusions, detailed targeting, lookalike audiences, and custom audiences) as follows:

• How things were before
• What changed
• How to approach it now

So it’s both a history lesson and an instruction manual on how to approach targeting now.

Did I mention it’s free? Subscribe here.

4. Final Days Coming for Advantage+ Shopping Campaigns

When Advantage+ Sales Campaigns replaced Advantage+ Shopping almost exactly a year ago, the clock officially started on ASC’s numbered days. We’re now approaching the end.

Effective last fall, advertisers using third-party apps were no longer able to create Advantage+ Shopping Campaigns using Marketing API V24. And now, ASC campaigns can no longer be created or updated using Marketing API V25.

But… that doesn’t mean that Advantage+ Shopping Campaigns are dead — yet. Advantage+ Shopping Campaigns created using API V26 will finally pause in September of 2026.

Presumably, the old ASC campaigns created in Ads Manager were stopped long ago, but any such campaigns created externally may still run for a bit. Talk about refusing to let something go.

It’s unlikely that there’s anything magical found in your old setup. Just move on with the new Advantage+ Sales Campaigns.

5. No More Recurring Notifications

Meta’s Marketing Messages (also known as Recurring Notifications) ended effective February 10th. This feature, which allowed marketers to send marketing content to opted-in users via Messenger, was removed from the Messenger API.

Recurring Notifications previously allowed marketers to send one message per subscriber every 24 hours. Once updated to the new Marketing Messages API for Messenger, marketers will be able to send one message every 48 hours.

If you’ve been around for a while, you know that Meta has gone back and forth throughout the years on this type of messaging access for marketers. While it’s desired by brands, there’s understandably potential for abuse. And that balance adjustment is something that Meta appears to be making here once again.

6. Conversions From Other Products

Meta ads attribution is complex. Because of that, it’s easy to misinterpret or misunderstand conversion results.

One attribution quirk that most advertisers don’t understand is that the conversions reported in Ads Manager don’t necessarily reflect the product you’re promoting. Someone might click your ad, go to your website, and then decide to purchase something else. This often leads to confusion.

It came up twice on this week’s strategy session, something I conduct every week for my Power Hitters Club – Elite members. Both examples highlighted common advertiser scenarios, so it inspired me to discuss those scenarios here.

Ad for low-performing product getting the most budget.

This was a fascinating situation. An advertiser running catalog ads was confused that Meta was spending 80% of their budget on an ad promoting a low-performance product. Ads aside, it’s not a popular product for the e-commerce brand.

But as the advertiser shared more information, it started making more sense. The ad was what they described as “click-baity,” driving lots of clicks. While Meta wouldn’t necessarily care about the CTR when optimizing for conversions, this ad was resulting in reported purchases.

The caveat was that people weren’t buying the product in the ad. People clicked the ad (because it generated a high number of clicks) and hung around the site to buy something else.

Meta doesn’t care what product was purchased, only that any product was purchased after engaging with the ad. And that ad was resulting in more purchases than others.

The advertiser’s instinct was to turn off that ad. But while it may not be a popular product overall, it’s doing a better job than other ads to drive traffic that eventually leads to conversions. The better approach may be to improve ads for the more popular products.

Reported attributed conversions are higher than actual conversions.

This is a common scenario. An advertiser gets 20 reported attributed conversions in Ads Manager, but they see only 15 sales of the product they were promoting. Meta isn’t making up these conversions. Assuming there isn’t an issue with your events, the purchases actually happened.

But once again, Meta isn’t reporting the number of purchases of a specific product. Ads Manager is generically reporting purchases that resulted from people engaging with your ad. It doesn’t matter what products were purchased.

These are good reminders as you try to make sense of your conversion reporting.

7. Meta Pixel Helper Gets an Update

The Pixel Helper Chrome extension has been an advertiser staple since the beginning. It’s recently received a long-deserved facelift. And based on the developer notes, this update happened on February 12th.

The first thing you’ll notice is a new icon. The old icon has been replaced with a more colorful version.

And now when you click on it, it opens a new sidebar on the right side of the browser. You can see all of the pixels detected on that page at once.

See the details of detected events on that page, including warnings.

And get details about what could be going wrong.

Otherwise, you may not see anything new, but event details are given a cleaner design.

8. Shops Third-Party Reporting Integration

When creating an ad, you may see a “Third-Party Reporting Tools” section at the very bottom.

It reads:

Meta purchases may not be included in your Google reporting. Connect your account to measure actions on ads that send people to your website or shop.

So the purpose of this integration is to make your third-party reporting (like GA4) more complete, not Ads Manager. If you click the link to learn more, you’re directed to the Help Center article “About connecting a third-party reporting tool to measure Shops ads performance.”

Here’s the detailed description:

Third-party reporting tools may not have direct visibility into events that take place in your shop on Facebook and Instagram. That means mid-funnel events like View Item, Add to cart, Page View Storefront, and Page View Collections may not be included in their reporting.

To address this gap in reporting, you can connect third-party reporting tool Google Analytics in Commerce Manager, Ads Manager or the Facebook & Instagram by Meta app for Shopify, Facebook & Instagram app for Magento, or Facebook & Instagram app for Salesforce Commerce Cloud.

When you click to connect, you’ll need to select your third-party reporting tool (only GA4 for now) and your unique measurement ID.

So, to summarize:

1. This is only for advertisers who drive traffic to Shops.

2. The integration is to give GA4 (and potentially other third-party reporting tools) more visibility into the funnel from Shop to website.

3. It won’t impact your attributed results in Ads Manager.

This reminds me a bit of the GA4 integration in Events Manager that we were all talking about last year (and I still have no idea what it does).

More to Come

Meta isn’t slowing down anytime soon. So I’ll keep watching, testing, and reporting back each Wednesday.

The post Manus AI Ads Manager Integration (Plus 7 Updates) appeared first on Jon Loomer Digital.

Targeting is the primary source of confusion among advertisers today. It’s so bad that most advertisers don’t even realize how confused they are. They completely misunderstand their level of control and required involvement in the process.

It’s not the advertisers’ fault. Meta has made so many changes during the past five years to targeting terminology and functionality that any confusion is understandable.

Ultimately, the bulk of the confusion centers around the idea of audience suggestions. Even the way Meta defines audience suggestions has changed.

Originally, Meta said that your audience suggestions would be prioritized before searching more broadly.

But Meta makes no such guarantees now, only that your ads will be shown to people outside of your suggestions “when it’s likely to improve performance.”

In my experience, suggestions are not prioritized. They may even be ignored. We can prove this when using breakdowns by age, gender, and even audience segments (when using suggestions that match our defined audience segments).

And since suggestions may not even be prioritized, it’s more important than ever to understand when our inputs are used as suggestions in the first place. Confusion isn’t limited to the inputs that Meta uses as suggestions. There are also baseless conspiracies related to location targeting and exclusions.

Let’s clear up all of this nonsense. Here’s a summary of the level of control that’s possible, focusing primarily on turning suggestions on and off, for each targeting type…

Detailed Targeting

In the default Audience section of the ad set, advertisers can provide detailed targeting inputs (interests and behaviors). When provided here, detailed targeting is only an audience suggestion.

From there, it gets complicated. You can click the “Further limit the reach of your ads” button at the bottom and switch your setup.

When you do, you might be able to turn off suggestions for detailed targeting. In reality, you probably won’t.

There’s a bit of history here. Ever since the original Detailed Targeting Expansion feature was released in 2021, Meta has repeatedly further limited advertiser control. The original feature was optional. Eventually, it was forced for three performance goals, including number and value of conversions. Then it expanded to link clicks and landing page views.

Today, advertisers are unable to restrict an audience to detailed targeting inputs when using any of the following 11 performance goals:

• Maximize number of conversions
• Maximize value of conversions
• Maximize number of landing page views
• Maximize number of link clicks
• Maximize number of app events
• Maximize number of conversations
• Maximize number of Instagram profile and Facebook page visits
• Maximize number of calls
• Maximize number of reminders set
• Maximize number of leads
• Maximize number of conversion leads

When any of these are selected, detailed targeting is only seen as an audience suggestion. This list consists of a large portion of the most commonly used performance goals, so your options for control are severely limited. But if you use one of the few performance goals not listed above, you’ll be able to turn suggestions off.

But again, this is the rare exception, not the rule.

Lookalike Audiences

When you provide a lookalike audience in the main Audience section of an ad set, it will only be seen as a suggestion.

Your options for restricting targeting to a lookalike audience, like with detailed targeting, are severely limited. You can choose to “further limit the reach of your ads.”

When you do, you might be able to remove suggestions. But, like with detailed targeting, your ability to do so is limited by the performance goal.

For the longest time, Meta only prevented advertisers from restricting targeting to a lookalike audience when using performance goals to maximize the number or value of conversions. But in late 2025, that changed.

You cannot turn off suggestions from lookalike audiences when using the following nine performance goals:

• Maximize number of conversions
• Maximize value of conversions
• Maximize number of landing page views
• Maximize number of link clicks
• Maximize number of app events
• Maximize number of conversations
• Maximize number of calls
• Maximize number of leads
• Maximize number of conversion leads

But if you use any of the less common performance goals not listed, you can still remove suggestions.

More often than not, however, the inputs you use for detailed targeting and lookalike audiences will be suggestions only.

Custom Audiences

If you provide custom audiences in the initial Audiences section, they’ll only be seen as a suggestion. Meta can and will completely ignore these inputs to get you more results.

It is incredibly common for advertisers to completely miss this. Based on my experience, many advertisers who think they’re remarketing aren’t actually remarketing at all. The audience was only a suggestion.

It’s still possible to restrict targeting to a custom audience, of course. But first you’ll need to click “Further limit the reach of your ads.”

Once you switch the setup, you’ll need to uncheck the box to use your selected custom audiences as a suggestion.

There’s a separate conversation about whether remarketing is necessary in most cases now. But you can remove suggestions when targeting custom audiences if you want.

Age and Gender

Within Audience Controls, advertisers have the option of providing a strict age minimum up to 25 that Meta will respect.

Otherwise, when customizing age and gender selections in the main Audience section, you’ll be doing this as a suggestion.

But Meta can and will ignore those suggestions to get you more results. Unlike suggestions for detailed targeting and lookalike audiences, you can use breakdowns to prove how much Meta goes beyond demographic suggestions. And I’ve found these suggestions make very little impact, if any at all.

You’re not required to use suggestions with age and gender settings. You can click the link to “Further limit the reach of your ads” to apply more control.

When you switch setups, you’ll need to uncheck the box to use your age or gender inputs as suggestions.

This should be avoided when possible. Even if there’s a specific problem to be solved related to information that you have and Meta doesn’t, you should use value rules to influence delivery instead of restricting by demographics.

Locations

Locations are part of Audience Controls, which are separate from Audience Suggestions. They will never be seen as suggestions.

It’s apparent that Meta knows about the conspiracy theories related to location targeting, though. I’ve often heard the claim that locations aren’t respected when using certain setups. But Meta states explicitly that “We won’t reach people beyond these settings, even with Advantage+ on.”

Location targeting isn’t perfect, of course, and the only option is to reach people “living in or recently in” your selected location. You can’t restrict your targeting only to locals or tourists, and people can get around these controls with VPNs.

But the locations you define are not suggestions, and Meta won’t choose to simply ignore these controls to get you more results. Your location inputs are respected.

Exclusions

You can exclude either custom audiences or lookalike audiences. In the fall of 2024, Meta removed the ability to exclude detailed targeting.

Like Locations, exclusions are part of Audience Controls. Exclusions will never be seen as suggestions.

As is the case with location targeting, I’ve seen the claim that Meta will ignore your exclusions when using certain setups. This just isn’t true.

That doesn’t mean that exclusions are perfect. You will undoubtedly reach people you were hoping to exclude. But that’s not because Meta ignored this control, and it’s not because exclusions are suggestions.

Your ability to exclude a given custom audience is only as good as the completeness of that audience. User error assembling these audiences aside, it’s virtually impossible to guarantee you won’t reach any of the people you’re trying to avoid.

Accept that this is normal. Make it a goal to provide custom audiences that are as complete as possible to limit the number of people who slip through. But it’s going to happen.

Summary of Control

Your inputs used for locations and exclusions are always respected. They are audience controls and never seen as suggestions. That doesn’t mean there aren’t holes in each, of course, but Meta won’t randomly show your ads to people outside of these controls.

When you provide age and gender settings within the main Audience section, these are suggestions. Meta can and will show your ads to people outside of those demographics to help get you more results. While you can restrict by age and gender, you should usually use value rules when there’s a problem to be solved.

The custom audiences provided in the main Audience section are suggestions only. Meta will show your ads to people outside of these inputs, so you aren’t technically remarketing when taking this approach. You can restrict targeting to custom audiences by removing suggestions, though Meta will prioritize remarketing audiences automatically.

Detailed targeting and lookalike audience inputs are suggestions when provided in the main Audience section. In most cases, you will not be able to remove suggestions and restrict targeting to either of these audience inputs. You cannot remove suggestions from detailed targeting inputs when using 11 performance goals or from lookalike audiences when using nine.

General Approach

It’s important to understand that even if you want to control targeting, it’s rarely possible. And when it’s possible, it’s rarely recommended.

Age and gender should rarely be restricted. Meta will go beyond your suggestions to get results. If there’s a problem to be solved (you have information that Meta doesn’t), use value rules.

Detailed targeting and lookalike audience inputs are almost always suggestions. You can’t avoid it. We have no way to reliably prove the impact of these suggestions or whether Meta follows them all that closely. Spend very little time worrying about detailed targeting and lookalike audiences as a result, and never create multiple ad sets based on these inputs.

Your custom audience suggestions are not tight constraints. You are not remarketing, and a separate ad set with custom audience suggestions is unnecessary and counterproductive. Restricting by custom audiences is rarely necessary since Meta will prioritize these audiences anyway. You can prove this with sales campaigns and audience suggestions.

The bottom line: The less you obsess over targeting inputs now, the better. Suggestions aren’t restrictions. Restrictions are not always possible or even helpful.

Let go of the targeting control you do not have. Question whether you need the control that’s possible. And dedicate your time and energy to the areas that make the most impact: Your ad copy and creative.

Your Turn

Anything you’d add related to targeting suggestions and control?

Let me know in the comments below!

The post When Your Targeting Inputs Are Only Suggestions appeared first on Jon Loomer Digital.