The end goal is to create a cold audience that is most likely to lead to positive results.

That quote from Jon Loomer in his post yesterday I absolutely love and is so relevant because that is always the goal we are trying to reach right, to create Top of Funnel cold audience ads that get the best results to create a roll-on effect down through the rest of our funnel.

So here’s the lowdown on this new Lookalike Audience Tool

Built on the Uploaded Customer File premise the LTV (Lifetime Value) aspect allows us to prepare our customer file differently prior to uploading and give Facebook some data around where to put some weight into the resulting Lookalike audience based on which customers are rated from high to low in terms of their Lifetime Value.

The key difference between this and a standard Lookalike is pretty obvious. Imagine two Lookalike side by Side comparisons;

STANDARD CUSTOMER FILE
A valuable source of people who have bought from your business and a Lookalike audience of people who  are likely to be interested in your business because they’re similar to your best existing customers.

LTV CUSTOMER FILE
The same data source fro the Lookalike audience in the above way but with an additional field fora rateable dollar value attached to each client that allows Facebook to focus on the the customers who provide the most value to your business.

Is this for me?

This isn’t going to be for everyone yet because for this to work there has to be a way of generating a rankable (Yes I made that word up) lifetime value on your clients/customers. Additionally lifetime value might not come into play too much when all your customers have a similar lifetime purchase value so the effort might not match the return. Lastly, this is a manual process so there might be a bit of work to do to calculate your customer LTV’s and create an upload ready CSV.

In saying that though I think this could be a very powerful tool to supercharge a Lookalike audience into something that might yield higher quality results from a value based cold audience so experiment with it and see what results you get.

Some useful links

Jon Loomer’s Article: Facebook Ads: Create a Value-Based Lookalike Audience

Facebook Customer data preparation best practices

How do I create a value-based Lookalike Audience?

Previously we have been used to setting up events on our websites to essentially define what data we send to Facebook about certain actions that are fired on specific pages. Now with the latest Facebook Pixel update we are seeing more of an asynchronous form of tracking where Facebook is now sending additional information from your website to better understand actions people take on your website to optimize adverts for delivery and conversion.

Additionally, and something that is going to require some considerations for owners of websites is the fact that the Facebook Pixel is also firing a new default event called microdata which includes Schema.org and Open Graph data.

Firstly lets talk about Schema and Open Graph data. 

Schema.org is additional code that you can add to your webpage to define different kinds of items from products to events, recipes, location, opening hours and reviews. According to Search Engine Land “Without structured data, a webpage is essentially all information with no context; adding structured data gives it that context”

Open Graph is meta data i.e. og:title  that allows a website to define objects in a website so that the information sent to Facebook when a page is shared (or liked, etc.) can be defined i.e. URL, Image, Descriptions, Title, and Content Type. Without this information Facebook has no contextual reference on the data from your website.

Further Details on Tracking Actions In Facebook’s own words:

“Track actions such as registering, purchasing items, and so on. These actions can be used to retarget audiences, find new customers, track conversions coming from your Facebook ads, and ultimately allow Facebook to automatically optimize your ads to result in more of these actions. We recommend adding the pixel to your site for the following use cases:

“Track and Optimize Conversions – Enable Facebook to both report back on the number of actions Facebook ads drive as well as allow Facebook to learn which people that see your ads are more likely to result in doing a certain action on your site, such as a purchase, and optimize the delivery of ads to result in more of these actions.

“Website Custom Audiences – Record page visits and actions from all visitors on your site, not just visitors that click on Facebook ads. Use this data to create Website Custom Audiences for retargeting with your Facebook Ads.”

The final part of this update is The “SubscribedButtonClick Event”

This event fires on every click a user performs on your site, sending the button text as a parameter (buttonText) the text on the button together with other data (buttonFeatures) like type i.e. Post, value i.e. Submit and so on. There’s also formFeatures parameters which pulls comment, tags, author, inputType, post_ID and so on.

So what does this mean for you?

Previously Schema.org and Open Graph data this was more in the realm of SEO (Search Engine Optimzation) but now that this data is a default Facebook Pixel Event the need for ensuring that your website Schema.org and Open Graph data is well defined is now a critical component of any business advertising on Facebook regardless of whether they feel that SEO is a big part of their marketing.

Additionally Facebook is also looking at what I always called contextual relevancy and on page user actions including button click activity on clients websites, like “add to cart” or “purchase” clicks, and will also include information from clients website page’s metadata to better understand context associated with these actions. This will include contextual information from the likes of ecommerce product descriptions and meta descriptions to help determine quality and relevance. This Will Affect Your Quality Score and I Believe Quality Is An Important Algorithmic Factor For Facebook.