– by Rachel Corcoran
The new year started with a roar for Internet vigilantes when the California Court of Appeal for the Second Appellate District issued its ruling in Hypertouch, Inc. v. ValueClick, Inc., et al. (B218603, Cal. Ct. App., January 18, 2011). In a stunning decision, the Court of Appeal overturned a lower court’s award of summary judgment to defendants ValueClick and Primary Ads, Inc. holding that the CAN-SPAM Act did not preempt Hypertouch’s claims under the California anti-spam law (California Business and Professions Code section 17529, et seq.).
Just when it looked like ValueClick had a victory under its belt against Hypertouch’s allegations of spam under California law (as the result of the trial court’s finding that Hypertouch failed to show ValueClick had any knowledge or control over the alleged spam activity of affiliates and that the federal exemption for state spam statutes prohibiting “falsity or deception” in emails was only intended to permit state law claims based on all the elements of common law fraud – and nothing less) the Court of Appeal upset the machinery of email marketing by holding that the CAN-SPAM Act does not preempt state spam law claims provided plaintiff simply shows that the entity advertised in a commercial email ultimately sent an email which contained falsified or misrepresented header information or a subject line that a person knows would be likely to mislead a recipient about a material fact concerning the contents or subject matter of the message. (Cal. Bus. & Prof. Code § 17529.5(a).)
We can expect to see more spam lawsuits in the future brought by these self-proclaimed spam watchdogs intent on saving the world from unsolicited commercial email now that the foundation has been laid for emails that were not even intentionally or knowingly misleading.
Vigilantes only need to show that an advertiser was the party ultimately responsible for sending an email that contained either a falsified or misrepresented header (e.g., a “From” name that the sender can’t prove has any relevance to anyone in particular) or potentially deceptive subject line, such as the “free” stuff that was promised to recipients in the ValueClick case but was not, in fact, free.
Undoubtedly, the world would be a better place if everyone played by the rules and sent emails that did not contain any deceptive content. Reputable networks go to great lengths to make sure that their affiliates follow these rules and only use pre-approved, non-deceptive subject lines and “From” names in emails.
So, you may be asking yourself why you should care about the ValueClick ruling since these folks only pursue advertisers with deep pockets, right? While most of these professional spam litigants will seek maximum damages against advertisers, your commissions may be negatively impacted if the advertiser or network decides to invoke its indemnification or set-off provision under the terms of your publisher agreement. You could find them stealing your profits right out from under you!
Rachel Corcoran is an attorney with internet advertising experience.
Whether you choose to call it adaptive delivery, real-time throttling, rate-limiting, automatic back-off mode, or anything else, the truth of the matter is that delivery policy controls, an advanced and quite sophisticated set of established email delivery technologies, have been a core mechanism of carrier grade MTAs for a few years now. Truth be told, while there have been some qualitative distinctions surrounding delivery policy controls lately; there really haven’t been any “game-changing” technological advancements in the very recent months.
Delivery Policy Controls
To proactively address deliverability challenges in real-time, Delivery Policy Controls allow you to maximize email delivery, and comply with ISP anti-spam traffic shaping guidelines automatically. Senders have the unique ability to optimize delivery settings, on a per domain basis, or more granularly, at an individual mail stream level. Having more control over these settings permits senders to apply specific policies on a per server, per domain, per VirtualMTA, or per campaign basis. One of the essential delivery policy controls is rate-limiting. When an ISP senses erratic sending behavior from a particular IP, or a particular domain, ISPs will exhibit what I call a yield sign (or temporary block) to fully digest incoming traffic patterns, in part due to peak traffic congestion, or network connectivity issues. By assigning control of critical volume parameters, ISPs shape incoming traffic based on simultaneous connections, messages per connection, and active connection retry attempts for just about any given time period. These delivery pylons, when configured and optimized in your sending environment, work to monitor, alert, and most importantly preserve sending reputations.
Justifications
While all senders have their own justifications and policies conducive to their deliverability environments, what’s important to consider is that these delivery policy capabilities have been around and well documented for a few years now. Without a doubt, and despite best practices, a majority of senders still underestimate the potential of this feature to defend and restore sending reputations and dramatically improve inbox deliverability.
Speed Bumps and Traffic Jams
Traffic shaping is of particular interest to ISPs. Their high-cost, high-traffic networks are their major assets, and as such, are the focus of their attention. They often use traffic shaping as a method to optimize the use of their network. Sometimes optimization can be achieved by intelligently shaping traffic according to importance of email stream or client. At other times, it is necessary to discourage or block bad actors. When you imagine all the traffic that ISPs receive, think of a very busy interstate highway. When an accident occurs, the highway patrol has to shut down lanes and literally shape traffic patterns. A five lane highway is quickly whittled down to two lanes and traffic is backed up for miles. Sometimes the traffic slows to a crawl. In almost all cases, the delay causes frustrations in the drivers involved. Similarly, senders become frustrated when networks get clogged by bad actors. If a bad actor congests the ISP network, traffic jams occur, just as they do when an irresponsible driver fails to obey the speed limit and causes an accident. When you think about it, ISPs work the same way as the highway patrol. Not only do ISPs seek to efficiently deliver sends on your behalf, they also encourage and want to preserve your sending reputation. Ultimately, these sets of delivery policy controls, give responsible senders the advanced tools they need to manage traffic jams. Next time you consider escalation to your ISP, understand that the majority of the time, it’s probably temporary, so don’t panic.
Special thanks to Carlo Catajan, ISP Relations and Product Manager for Yahoo who inspired me to write this post.
Fred Tabsharani
Individuals thrive on the ability to choose for themselves. That’s why we have something like 20 brands of yogurt, over 30 flavors of jam and an aisle of potato chip selection at the grocer. Think about it. Aren’t you drawn to the brands that allow you to customize your experience? Consumers have plenty of choices about everything these days – and with those choices come power. What does this have to do with deliverability? Everything really. Whether you’re mailing to a B2B or B2C audience, likelihood is they’re used to having the power of choice…so if you’re not giving it to them, someone else will.
Let’s look at these from the viewpoint of the recipient:
My Choice
Most successful senders I’ve worked with that enjoy strong deliverability rates start by getting permission right. Ultimately to be this successful with email your recipients have to “want” your messages. In our 2009 Channel Preference Survey when 1,500 consumers were asked to rank on a 5 point scale the “acceptability” of receiving messages with varied levels of permission (1 = completely unacceptable and 5 = completely acceptable). The survey showed the average ranking of 1.7 (completely unacceptable) from “companies with whom I’ve never interacted” vs. a 4.1 (very acceptable) from “companies to which I’ve granted permission.” Ignore this simple fact and this level of acceptability shows up later in spam complaints.
What I want
Right behind choice and permission comes content. Email is about a value exchange. If you’re not keeping up your part of the bargain and providing the value you promised up front to entice the opt-in, then deliverability will suffer. Poor content is more likely to affect your deliverability not because it contains the wrong words (spam content filters are often less important than reputation), but because the content doesn’t meet the subscribers’ expectations. If subscribers don’t get “what they want” they’ll complain, they’ll unsubscribe, or worse they’ll ignore you. Yep, apathy may be the worst thing that will happen. That’s because many ISPs are now paying attention to engagement – opens, clicks, “this is not spam” actions (where the recipient pulls your email from the spam folder) etc. to determine if your mail is wanted.
Develop a preference center where subscribers can select specifically what they want, by publication, or content – then stick to those promises.
When I want it
In our 2011 “Social Break-up” survey of 1,500 consumers, 54% of people who unsubscribe from permission emails said the reason was emails coming too frequently. As I said above, unsubscribing is actually good in some ways since this action removes the subscriber from your list in a way that may be less impactful in the long-run than a complaint or apathy. As Dennis Dayman showed in his post from last fall, more companies are giving subscribers the option of determine how often their emails should come. Scotts MiracleGro even provides the ability to define your growing zone and the types of plants (flowers or vegetables for example) you like to grow. If the subscriber is in control of content and can choose the frequency, they are much more likely to respond, and much less likely to take the more drastic action of complaining.
Plus all other aspects of our lives are moving the direction of consumer choice. What isn’t on-demand these days? I’ve even begun to loathe the idea of ordering a Netflix CD delivered postal --- I mean why do so when I can have it delivered, right now, when I want it?
A simple on/off switch isn’t enough. Let your subscribers choose and you’ll get the best results.
Reposted from my ClickZ column.
------
Money and list sizes aren't everything. OK, maybe money is everything and here's hoping you get lots of it, but the size of one's email target list size doesn't indicate how successful you will be now or in the future. Last week, I had the pleasure again to speak at the famed Email Insider Summit; not on the topic of email best practices, but on the security of customer data and the brands and service providers who are entrusted with it.
I'm not here to give you another lesson about data governance and security. However, a very interesting, but all too well-known issue I heard from marketing managers at the event roundtables was "I'm blocked and when I looked into it, it turns out we are hitting spam traps or other problematic accounts/issues." We panelists simply suggested that if people are not interacting with your brand over a few months, then you either survey them to get them to reengage or flat-out drop them. (Sorry Dela Quist - my U.K. friend hates the dropping idea.) Now, I'm not talking about suggestions like decrease 50 percent of a list or confirm opt-in everything, but we suggested it would help cut down their list to individuals who were truly engaged with the brand and wanted to receive the message.
To our delight, many of the marketers had already had similar ideas to improve list inbox penetration, but to our horror it seems that their high-level executives freaked out when such programs were suggested or introduced. I began to ask myself "why" at the conference that afternoon.
Honestly, I don't have a full-on answer to this perplexing battle between the higher-ups and those in the trenches looking at the terrible delivery data in these situations. Is it possible that the folks in the trenches are not doing a good enough job of presenting the whys? The costs possibly associated with bad data? The higher-ups not taking enough interest in what is going on below them daily? I did tell those at the conference that this process is a hard one to sell to many executives, and to be honest, it's not the first time I've seen this problem or even been through it myself. It requires a lot of data and comparison matrixes, flow charts, many blacklistings, etc. to illustrate the need to cut down an email target list. This isn't typically a proactive thing for many companies. It is an unfortunate reactive situation.
To my point here, does anyone else have this issue with convincing executives that list size doesn't prove anything? Of course. My suggestion to everyone in this situation is this: sit down for a few weeks and capture the costs of reacting to junking at the ISPs, time down due to blacklistings, and any other negative situations that dirty lists put you in to illustrate where the company is losing time and money. This includes also looking at all your delivery statistics (failures and successes) with bounces, feedback loops, and unsubscribing. Take all this data and create an easy-to-understand and attributable chart that shows the downside of your current actions. If you can, figure out how many emails are not getting through and how much more "easy" money you could make if you only delivered to those addresses you know are not giving you a hard time.
I would even suggest bringing the executive team into some of your daily squabbles so they can see just how bad lists are in reality and how they are wasting employees' time and the company's money.
Let's make email lists successful by using them the right way. What is your company doing to address this problem?
-Dennis
Eloqua
Don't Just Send, Deliver!
A hot topic in web-based communications has for some time centered around the question of whether or not social media messaging will replace email as our primary means of communication over the interwebs. This conversation has grown significantly and in proportion to the rise of social networks like Twitter and Facebook.
As people deeply involved in email, the contributors at Deliverability.com are more than willing to point out the holes in the side of the argument that says social media will in fact kill email, as is evident from a recent post from the folks at Critsend right here on the blog. Similarly, at SendGrid, we believe that our mere existence, let alone our extremely fast-paced growth, is evidence enough - and we're telling the world when given the chance.
Luckily, we're not alone...
Just last week, two well known investors made their voices heard on the topic.
Fred Wilson published a very insightful blog post, pointing out the fact that email is actually a critical tool and contributes directly to the success of today's social media giants and budding upstarts alike. As he mentions:
"...more and more social applications are leveraging the power of email to drive repeat usage and retention."
In his follow-on post, Brad Feld, takes Fred's point a step further and discusses the importance of executing email programs efficiently and effectively within a web-based business. He makes clear that the dependency on email does not stop at social apps. Virtually all web applications send some sort of transactional email - account registration confirmations, password reminders, activity notifications, etc. He then goes on to discuss the headaches that implementing an in-house email solution can create and the corresponding logic around paying email experts to handle it for you. His final message to web developers:
"...if you are serious about what you are doing, focus on your app. Don’t waste precious development time on all the activities around the app."
So, as the number of Facebook messages, wall posts, comments, @-Replies, and Direct Messages grows, so too does email. Without email, how else would you know that Nancy tagged you in a photo from the dinner party last weekend? But more importantly, how would you know that your recent purchase from Amazon just shipped?
Not to worry, though, because as long as we continue to do our jobs right (and we will!), you'll continue to see those ever important notifications right there in your inbox :)
What are your thoughts on the topic? ... Is email simply riding the coattails of social media's success, temporarily avoiding its imminent death? ... Will social media be the one to go by the wayside, while email lives on in internet glory?
Let us know what you think!
Thanks for reading,
Tim Falls, Community Guy @SendGrid
One of the greatest fears for organizations that have a substantial email program is that of list attrition and stagnated growth in the subscriber base. List attrition happens when subscribers issue a spam complaint, bounces or “naturally” unsubscribes. A “natural” unsubscribe occurs when the subscriber takes the time to find the link, clicks on it and proceeds with whatever unsubscribe process the company has put into place.
Companies view the unsubscribe as a threat and a failure to the long-term viability of the program. Truth be told, some companies are just plain scared of subscribers leaving their list, because someone in the organization believes that bigger lists are better. Over the years, the common practice by organizations is to place the unsubscribe link at or towards the bottom of the email. Some make it clearly visible, while others find ways to hide it either in the legalese, via a smaller font or in the case with a few companies I have seen by lightly shading the link to blend into the background of the email.
As professionals, we all know that the act of a subscriber issuing a spam complaint against the company is far more damaging to the sender’s reputation than a “natural” unsubscribe. I have long thought that if a subscriber wants off your list, you should make it easy for them to get off. Keeping them on and unengaged in your program can lead to reputation issues. Issuing a spam complaint is easier for the subscriber because most if not all spam buttons are at the top of the email client.
Companies should make it easy for people to “naturally” unsubscribe which is why I recommend that organizations place their unsubscribe link both at the top and the bottom of their email. Before you as the reader haul off into some tirade that I am anti-marketing and what I have just said is email blasphemy, hear me out.
If someone wants off your list, they are going to do one of 3 things: hit the spam button, silently unsubscribe or click on a link. The first two can hurt an organization in the long run and have potential repercussions across the entire program. When someone leaves your program by clicking a link, they are leaving your program on mutual terms. They get off the list and they do it on your turf where you then have an opportunity to either save them or learn why. The “natural” unsubscribe may hurt your list size, but it does not damage your reputation.
Some people would argue that it is best to only place an unsub link at the top of an email to combat existing complaint issues and then, once clear of the issues, eliminate it and proceed as normal. I equate that line of thinking to a wreckless driver who thinks they can game the system of speeding tickets by driving the limit for the period of time so it doesn’t hurt their states “point” system. Then, once they are in clear, begin to drive like a maniac again.
Throughout my career, I have personally witnessed and have heard through colleagues that by placing an unsubscribe at the top and the bottom, both the percentage of spam complaints and “natural” unsubscribes have actually gone down. In fact, I have personally been associated with numbers ranging from 15%-45% reduction of spam complaints. However, I have also seen some issues associated with the placement at the top from mobile subscribers, specifically Blackberry users who accidentally click on the unsubscribe link in the pre-header when placed on the left hand side. Long and short is that you need to test placement as well as metrics around its placement.
Implementing an unsubscribe link at the top of email is not a hard thing to do, but it’s a hard sell to executives. A solid business case centered around the reduction of spam complaints and reputation sustainability leading to revenue preservation must be presented.
In the long run, a unsubscribe is an unsubscribe and the process by which you make the subscriber unsubscribe is one of the last impressions they have of your email program. Make that experience and impression a good one, even if your organization gets nothing in return.
The post below comes from guest contributor Barry Abel, VP of Field Operations at Message Systems.
Industry efforts to develop automated solutions to deliverability challenges have met with limited success over the years. That’s beginning to change as technologies become progressively more powerful and techniques more effective. But specialists in the field who cope with deliverability every day still have their hands full. They bring tremendous value to the table, but haven’t been able to apply their business intelligence to its fullest effect. That’s largely because most approaches to deliverability are reactive, and most available tools have been blunt instruments.
Where the new generation of deliverability technologies breaks with the past is in giving up on techniques that have proven unsuccessful in the past – or I should say, that’s the approach we’re taking at Message Systems. Let me explain.
The old approaches to ‘automated’ deliverability largely worked on a presumption of predictability. Yet, of course, email is anything but predictable. It’s a fast-moving, dynamic environment we operate in. Senders haven’t had good predictive tools, so they couldn’t know what would happen, and they couldn’t react to what was happening in real time.
As a result, they were left to manage deliverability based on what did happen – after the damage had been done. Senders were forced into a continual reactive mode, trying to repair things and prevent a recurrence using traffic shaping techniques such as configurable delivery parameter settings, throttling (send rate, number of connections, etc.), retry frequency and cadence adjustments.
There’s also been the reliance on static scripts that aimed to automatically adjust delivery parameters or possibly suspend mailings based on simplistic ‘if/that’ logic. But scripts are another blunt instrument – they lack the granularity and can’t cope with unplanned-for incidents, so they fail when new variables are encountered. Of course, unforeseen variables occur all the time, which is why script-based approaches have consistently proven inadequate.
Automated throttling programs are another approach that’s proven less than successful. They’re usually broadly applied across domains and sometimes by sender, but not by the individual send of individual mail streams to individual domains, and they can’t be adjusted on the fly based on how messages are actually being delivered. Even if you could adjust sending parameters by individual stream and domain, you’d still have manageability issues because there are simply too many variables across too many mail streams and receiving domains.
Indeed, monitoring ISP delivery codes, researching problems and fine-tuning settings are all inherently hit-or-miss, which is why the process has been so time-consuming. Traffic shaping is expensive, too, and it’s of no use aligning sending rules with ISP requirements on a per-message/per-sender basis.
The key success criteria for any technology solution should be making the management of deliverability easier so specialists can focus on the more important job at hand – correlating delivery results to marketing practices so as to enhance the company’s bottom line and its relationship with customers. The current crop of deliverability approaches has largely failed on that score.
The next generation of deliverability management solutions – we’ll be making an important announcement on this front in a follow-up post next week – recognizes the shortcomings of these approaches, and needs to provide the kind of granular, real-time corrective capabilities that empower deliverability specialists and saves them time.
Hello out there!
We've an opening for an email product manager to help us define and drive the vision and development of the IBM Unica Email product. If you are working in email for an ESP, have a product management background (read: email ninja/rockstar/jedi), then we really would love to hear from you.
You can find the listing here and can submit your CV through that site. If you have any questions feel free to drop me a line via twitter @lenshneyder.
Cheers!
-Len Shneyder
Product Marketing Mgr.
IBM
Inactive subscribers are a liability to anyone who sends email. They hurt deliverability, which in turn reduces your response rates, and before you know it, your email program isn’t making as money as it use to. Marketing managers usually understand that they key is to get rid of the deadweight to solve these problems, but most executives try to solve the revenue problem through a numbers game and sending to even more email addresses, many of which are inactive as well. So why should you care about removing inactive addresses?
ISPs define active vs. inactive addresses in a variety of ways through things like last log in date, how long they spend in their email client, and if the behaviors resemble that of a real person, such as reading, deleting and even marking email as spam or not spam, as well as other actions that we generally call engagement-level filtering. Marketers look at it slightly different based on the data at their disposal, such as opens, clicks, conversions, or website activity. ISPs care about active users because spammers have always been looking for ways to game the system, and one successful way has been to load their lists with inactive email addresses, because inactives will never hit the spam button, and complaints stay artificially low. ISP postmasters are a smart lot though and caught on, and as a result, ISPs calculate complaints based on active and trusted subscribers. This means that complaint rates are much higher for not only spammers, but any other sender as well.
So what does that mean for marketers? Most marketers have been calculating complaint rates incorrectly for years by doing it off of total volume, rather than email delivered to the inbox. If you’re mailing to a million addresses, and only half reach the inbox and 5,000 subscribers complain, then you have a 1% complaint rate, instead of 0.5% complaint rate based on total volume. Let’s say you only have 250,000 of those subscribers active according to the ISP. You now have a 2% complaint rate which is high enough to get you delivered to the spam folder, or maybe even blocked, at Hotmail, Yahoo, Gmail and AOL.
Let’s say you turn a blind eye to those inactive addresses. Given enough time, they’ll turn into unknown users and spam traps. Again, spammers ruin the email party through shady list practices that have high unknown users and spam traps. ISPs respond by blocking and sending email to the spam folder when unknown users start to exceed 5% since that is a common link with spammers. Additionally, ISPs recycle abandoned email addresses into spam traps because spammers buy and steal lists and this method is an easy way to find out who’s doing it. Unfortunately for marketers, this means that mailing addresses that haven’t been mailed in awhile could result in spam traps. Just one spam trap hit can cause deliverability issues, which is also supported by research we’ve done at Return Path.
There are other costs associated with delivering to inactives as well. If you have a large list, mailing to inactives can put strain on your mail system and cause a lag in your delivery. This can be a problem for time sensitive emails and during times, like Christmas, when the ISPs are under strain and trying to cope the increase in volume from all senders. If you’re using an ESP, you’re also unnecessarily paying for emails sent to inactives who will never read your email in the first place. By sun setting these addresses, you can deliver your mail on time and even save money.
Here are three simple things you need to do to sun set inactives:
Any executive will have a hard time arguing against this given the benefits of higher inbox placement rate, cost savings, and an increase in response rates and revenue.
Tom Sather
At the close of MediaPost's Email Insider Summit, a four day whirlwind of island events and ardent discussions on email, I sat at the famed Mucky Duck in downtown Captiva Island and reflected on the conference. Throughout panel discussions and roundtables on relevance, preference centers, testing, content marketing, and my own panel on email security, the humanity behind email shined through. Here are some highlights from a few of the people who made the event a success:
On Content Marketing
Joe Pulizzi demonstrated how brands can acquire and keep customers by creating and publishing their own content. A slide from his presentation stated, "73% of consumers prefer to get information from a company in the form of a collection of helpful articles over an advertisement or offer. 61% are more likely to buy." Content should be focused more on the customer and less on the brand.
On Customer Lifetime Value
Arthur Middleton Hughes described how to calculate Customer Lifetime Value over a three year period in email and include offline sales that occur as a result of email. Emails cause people to take action in the same way tv commercials or billboards might. Email departments should use this metric to justify budgets and also to consider the value of customers who unsubscribe, so they can fight to keep them. Another key to acquiring customers is getting the email team and the website team together to optimize new email signups.
On Preference Centers
The panel on real-time marketing discussed adapting to customer behavior and priority inbox technology to improve engagement and relevant segmentation. Preference centers allow consumers an alternative to unsubscribe as well as a way to define how they interact with your brand. Loren McDonald wrote an excellent follow up article on preference centers here.
On Security
Members of my own panel on data security, Dennis Dayman and Allen Nance, discussed the need for marketers to self-regulate and read OTA's "Security by Design" framework. Legislation has been focused around privacy and compliance, but there have not been nearly as many regulations passed specifically on marketing data and how it is secured. Email addresses are valuable and should be treated with the same sensitivity as PII and healthcare data. An annual security audit between a brand and an ESP is a good place to start. Getting the entire organization to embrace a data-sensitive culture is crucial.
On Being Human
Go off topic once in a while in your content and marketing stratgey. Phil Hollows from the final panel of the conference reminds brands that it's not natural for people to talk about the same thing every day, and you can get a lot of mileage by just talking to your audience in a real way. While immediate results and conversions are important, it's also about being memorable and engaging enough to differentiate from the competition and create a real, human connection.