By Ken Magill with Allan Levy
Engagement. With the possible exception of relevance, it is arguably the most overused word in email marketing. Engagement became a hot topic in 2012 when deliverability professionals began touting it as a necessity for continuing to reach subscribers’ inboxes. Their argument went that the criminal spam issue had largely been solved with filtering technology. As a result, they said, email inbox providers were going to turn their attention to the next problem: commercial senders with large, largely inactive files.“
In the coming year, I think we will see more engagement-based spam filtering, so it’s going to be important for marketers to look beyond opens and clicks to determine who is actively engaged in their email program,” said Tom Sather, senior director of email research for Return Path, in 2012.But the engagement discussion led some email-deliverability professionals to questionable conclusions, arguably the most damaging of which was counseling marketers to remove email addresses from their files after a certain period of no open or click activity—typically 18 months. The result was a bunch of marketers needlessly butchering their email files, an often wrongheaded act that continues to this day.
In 2015, the email marketing world was rocked when ISP representatives at an Email Experience Council conference said they did not measure clicks, and that low levels of engagement alone would not get a sender blocked from reaching inboxes. Did this development render engagement unimportant? No. It still has a relational effect on deliverability. But ISPs aren’t nearly as strict when it comes to engagement as many deliverability experts led their clients to believe. And culling email addresses simply because they have shown no opens or clicks in a certain period is not the way to approach the issue.
Of course, spam complaints and unsubscribes should be removed immediately. Hard bounces should also be removed after efforts to resolve the issue have failed.But deliverable addresses that have been acquired on a permission basis should almost never be removed from a file solely for lack of opens and clicks. They simply need to be treated differently than the addresses of folks that have been opening, clicking and buying. First, the file must be segmented by address activity. A typical email house file has a small percentage of super-active addresses with high average order sizes and lifetime value.
Then there will be a segment that is active, but not as active as the super-active segment. Send more email to the super-active group, maybe seven emails a week as opposed to five and maybe two emails for the same promotion as opposed to one.Send the slightly less active group more email but not as much as the super-active buyers.Then there are the inactives. A typical house file will have a significant segment of addresses the marketer doesn’t even send email to anymore because of their inactivity. There is potential revenue in this group. They simply must be approached with care. But they should be approached with, say, one or two emails a month.It is important to use the same IP address when sending to inactives as is used to send to the actives and super actives. This way, the weaker email addresses ride on the deliverability benefit of the main IP’s presumably good reputation. However, it is often a good idea to break the inactive segment into multiple mailings to avoid tripping inbox providers’ spam filters.
The key to intelligently increasing engagement is testing. Test offers. Test creative. Test frequency. Test segmenting parameters. Test the number of links, and calls to action, product mixes—anything that might get more recipients opening and clicking. During this process, a smart marketer will continuously monitor for when addresses move into different segments, recreate the segments and treat them accordingly. The marketer who segments, tests, resegments and adjusts email frequency accordingly will naturally increase subscriber engagement and get the side benefit of better deliverability without needlessly butchering their file.
