Commentary

Engagement: The New Frontier In Deliverability?

Recently there has been a great deal of buzz around the idea that "engagement" is a new metric used by ISPs and other mailbox providers to determine mailbox placement.

 However, engagement isn't new at all.  It has been a part of the filtering mix for quite a while.  The ISPs are simply adding clicks, opens and a few other measure of user engagement to the long list of other engagement metrics that have been in use for a while.  All these metrics try to do the same thing -- figure out which messages are truly wanted by subscribers.

 These engagement metrics include: 

  • This Is Not Spam:  This has been an important reputation metric for over three years.  It's the primary way that many ISPs get a feel for whether they are making a mistake in placing a message in a junk/bulk folder.  Most of the large ISPs that have built their own reputation systems are looking at this metric.

  • "This is Spam" and "This Is Not Spam Data" from Trusted Reporters:  Certain senders try to game reputation systems.  They mail to a large number of inactive accounts to bulk the denominator in the complaint rate metric since inactive accounts don't hit the report spam button.  Some senders will take this a step further and set up accounts at an ISP to register "this is not spam" votes for their messages. Mailbox providers have responded by reviewing only reporting from active, engaged accounts that don't look dormant or look like they have been set up for the purpose of gaming.  AOL has publically stated that they are including trusted reporter data in their filtering algorithms, but we know a number of other top-tier ISPs that are doing this also. 

    advertisement

    advertisement

     

  • Panel Data from Trusted Users:  Another technique to measure engagement is to create a large panel of trusted users to vote on messages that were sent to each user and placed in the inbox or junk mail folder.  The users are given a sample of their messages and asked to vote each as spam or not spam. This technique is most notably used by Microsoft.

     What should marketers do to maximize their deliverability now?  Here are three strategies I recommend:

     

  • Treat inactive subscribers differently: This is probably the biggest change that most marketers need to think about.  Mailing to a lot of inactive accounts may actually make your reputation look worse at some ISPs.  Segment out inactive users and run a win-back campaign. If you cannot win back these subscribers, you may simply want to stop mailing them altogether.

  • Segment out "this is spam metrics" (spam reports) for more active subscribers - These subscribers matter the most in terms of your overall reputation.

     

  • Do not try to game reputation systems: If you've been tempted to create a large number of accounts at webmail providers, signing them up for your newsletters and programmatically hitting the "this is not spam" button for all your messages if you go into those inboxes, I have two words of advice: Just don't.  You will be caught and your mail is going to have a very hard time getting delivered.  More importantly, is this really how you want to spend precious marketing resources?  The time it takes to set up and maintain this game would be far better spent on creating happier email experiences for your subscribers.

     I think the good news is that focusing on engagement metrics means focusing on the same strategies that drive all great marketers: to create compelling email conversations with their customer.  If marketers can craft a stream of messages that drives a lot of opens and clicks and drives active email users to rescue the message from the junk mail folder, they should have no deliverability issues.

  • 2 comments about "Engagement: The New Frontier In Deliverability?".
    Check to receive email when comments are posted.
    1. Marston Gould from Classmates, October 7, 2009 at 11:48 p.m.

      What we're all learning is that the ISPs would rather play a game of cat an mouse with senders rather than truly create a system that benefits everyone. In today's world, very few ISPs actually provide feedback loop information, so at best, you can only filter out members who mark your email as spam from some, but not all of your subscribers. It also doesn't help that none of the ISPs provide a method whereby legitimate senders can receive true data on opens or whether or not their send is being sent directly into a members trash can, spam folder, or some other folder in the inbox. Nor do any really help senders understand whether or not they should be sending a particular user emails that use more or less imagery. And don't get me started with the poor choice made by Microsoft to use Word as the rendering engine for Outlook. And what about the fact that practically anybody can set up a spam reporting business in their garage and hold legitimate senders hostage by blackmailing them to get off of lists.

      So what are emailers to do? Well we could be very conservative and take out all customers who we do have spam reports from. We could also take out customers who haven't responded to email in a long while. The problem with this approach is that this can often have a significant impact on revenue.

      We could go down the path of hiring a deliverability firm, but often times all they are providing back is processing of your log file data. Something you could probably do on your own if you had the technical bandwidth. Or you could hire someone like Goodmail - which again is essentially a tax on sending.

      Either way - email as a platform is heading to a tipping point. Real change is needed and I'm not talking about Google Wave.

      What I find interesting is that even though email supposedly has such high ROI, marketers really have less control of its placement in front of customers than they have with display ads or paid search.

      Why not create an environment within email that is more akin to search. Create a paid email and non-paid email display. Legit bulk emailers would gladly pay a fee to have their email positioned highly in particular customers inboxes at certain times of the day or pay for multiple high and/or persistent placement. This fee for placement would provide them a real opportunity to improve the ROI of their campaigns. Illegit bulk mailers wouldn't likely be willing to pay for placement because they wouldn't want to commit to a financially trackable system.

      If customers find a particular senders email engaging because of a true open and/or click - then that relevancy could be used to position mail higher as well. In fact, if it reaches some threshold, it could show up with non-bulk email until such time that it is less relevant and falls back to the paid display side of email. This would give senders a real goal to hit - if you make your email engaging, then you pay nothing for high placement. Make it irrelevant and it becomes virtually impossible to get high placement in the inbox.

      Solutions like this are a real fix - not the voodoo hoops senders are forced to go through today.

    2. Allen Maccannell from SenderOK, October 9, 2009 at 3:09 a.m.

      I agree with Marston's words about "cat and mouse". Classmates can be unsubscribed from and even Google has recently stated that it isn't fair to hit "This is Spam" on something you subscribed to (however long ago) and which you can take an extra few seconds to unsubscribe from.

      What's interesting is that our SenderOK plug-in does a really good job of keeping stuff you clearly subscribed to out of the spam box, especially if it is authenticated with DKIM or SPF. It isn't spam if you can unsubscribe from something (that you signed up for) in seconds, so qualified paying customers who also want anti-phishing icons to appear in user inboxes get a guarantee of not going to the spam box (for any of their addressees who happen to have the plug-in installed)...and that guarantee is iron tight because the software is on the user's machine and has the last word.

      Regarding the "gaming" discussion, that is another cat and mouse issue because one can assume that every good Email marketer has a personal Yahoo, Live and Gmail account they use for personal and other reasons that they *also* use to mail themselves to see how their mailings look from an end-user perspective and they, of course, will click "This is Not Spam" when appropriate. To punish this as "gaming" isn't fair.

      A real spammer is going to get so many complaints that he or she will not be able to make a difference by doing that.

      Every smart company I have been in has put its employees' private webmail addresses on the mailing list and asked the employees to put the sending Email address in their Contact address books + click "this is not spam" if they ever find the company email in the spam box.

    Next story loading loading..