What to do with inactive email subscribers?

Why inactive email subscribers may be too good to dismiss

Conventional marketing wisdom tells us removing inactive subscribers from your mailing list is good practice - but is it? Here, we turn that wisdom on its head. I'll show why inactive subscribers are still valuable to your business and explore how you can reactivate them. First, I'm going to tell a story of how I was recently shunned for all the wrong reasons and then give some ideas on a structured approach to make the most from inactive subscribers.

It's simple really, I was recently removed from Dell's mailing list despite spending around $40,000 a year with them online for the past few years.

In Dell's eyes however, I was inactive. I hadn't opened any of their emails for some time and so their marketing department had purged me from the list.

But the fact a customer hasn't opened your emails doesn't mean they're not engaged with your brand. In my case, I simply hadn't needed to open my email to buy any new computer equipment for a while.

Nudging customers towards your brand

Even unread, those emails were still a nudge towards the Dell brand. They were still performing a valuable marketing function whether it was in the form of an engaging subject line with the latest offers, or keeping the brand at the front of Dela's mind. And in my case, they were working.

Our research has shown that customers often buy a product or service through another channel within 24 hours of receiving an email. Here's an example:

That's why we provide clients with reports where we overlay the timing of emails sent with the timing of sales from other channels such as in-store, online, through call centers, pay per click and affiliates. Marrying together these data sets is crucial if you want to get a true impression of your recipients' buying patterns. And it's even more important when deciding whether to bump them off your subscriber list.

Email marketing can create a powerful stimulus which prompts purchases even if the emails themselves aren't being read.

Inactive subscribers are still engaged

Inactive subscribers are still valuable to your brand and can generate a significant amount of revenue.

It makes perfect sense to stop emailing someone you know would never buy your products again or who actively dislikes receiving your emails. But long-term inactivity isn't a good indicator of whether someone falls into that category. There are 5 reasons why your subscribers may be - or appear to be - inactive:

  1. They want your email, but haven't needed your product for a while.
  2. You're receiving false negatives - your email is optimized to be read with image blocking on, so some subscribers could be opening it without you knowing.
  3. The subscriber doesn't want your email, but doesn't care enough to unsubscribe.
  4. Email address churn - the subscriber no longer uses or rarely checks that email address.
  5. They don't see your email because it goes into the junk folder.

By far the largest group is the first one - we call these people the unemotionally subscribed. They will happily ignore your emails until they're ready to buy, because it's easier than unsubscribing and having to remember your URL or Google you later.

We've gathered plenty of evidence on this group and demonstrated that while they might not read an email, they're still a very important customer base. For example:

  • One of our clients generated $120,000 from subscribers who had not opened or clicked on the previous 25 to 40 emails.
  • Another saw 14% of revenue generated by subscribers who did not open or click a single email.

Common marketing advice would have been to delete those subscribers after a year's inactivity. But by retaining unemotionally subscribed addresses, the client brought in a significant amount of additional revenue.

When to remove email addresses

We have developed a simple strategy to help you decide if, and when, to remove an email address from your list.

I recommend these 4 steps to establish which addresses are truly inactive, and which fall into the unemotionally subscribed group.

  • Step 1 Start a reactivation campaign. Try to re-engage anyone who hasn't opened an email for more than 6, or even 12 months.
  • Step 2 Separate your lists. Anyone who still hasn't opened an email after the reactivation campaign should be placed on a separate list to your active recipients.
  • Step 3 Send the same email to each list and focus on activity. The active list will now show a truer representation of engagement and your results will not be dragged down by the dead email list. After every mailing (or month), move anyone who becomes active again to your active list, and anyone on the active list that now qualifies as inactive by your definition to the inactive list. You can now clearly identify how much revenue is generated by emailing the dead addresses versus how much it is costing you.
  • Step 4 Analyze over time before deleting anyone. Within 6 to 12 months you'll have a much better sense of how long you should continue to email an unresponsive email address before removing it from the list. We have generally found it to be the point at which almost every person who opens an email for the first time in a while goes on to unsubscribe.

So that's the way we see it. What's your view?

Share your thoughts

  • Kalclash commented on November 25, 2011

    The key is to segment your database by engagement. You can then hit your engagers far more regularly than the non-engagers, but ensure that the non-engagers are still receiving emails from your brand.

    As soon as the non-responder starts engaging again with your emails and/or website they will become an engager and move up to the top of your segmentation hierarchy. Simple.

    • Exactly! Simple.

      Would be interested to hear how you measure engagement at a field-level for subscriber.

      Some combination of recency and frequency of last open or click?

      Thanks for your comment, Dave

      • Kalclash commented on November 25, 2011

        Hi Dave,

        I have a hierarchy in place that looks at recent web engagers at the top of the pile, filtering down to recent email clickers, previous bookers, etc etc all the way down to never-engagers.

        We can track web engagement without having an email click in order for a dropped cookie, but these guys are obviously top of the top for us…

        • Kalclash commented on November 25, 2011

          Hi Dave,

          I have a hierarchy in place that looks at recent web engagers at the top of the pile, filtering down to recent email clickers, previous bookers, etc etc all the way down to never-engagers.

          We can’t track web engagement without having an email click in order for a dropped cookie, but these guys are obviously top of the top for us…

          • Thx for sharing Kalclash. So for email an engagement is a click.

            Sounds like this is quite sophisticated stuff since it’s cross channel – must need a “data warehouse” or analytics system that allows you to integrate personal info.

            Nice one.


  • If you’re following the advice in this post and need help with step 1, see this post http://blog.emailvision.com/eng/3-proven-win-back-subject-lines-and-why-they-work on win-back subject lines.

  • Thanks guys, for your comments – looks like we’re all in agreement and you like the process Dela has run through.

    The cynic in me sometimes wonder whether culling of inactive happens because it can boost the apparent value of the list when measured by oversimplistic engagement KPIs such as open/click rates, revenue per 1000 emails sent, etc

    Thanks Derek for adding the deliverability angle. That could make removing the long-term inactives valid after you have performed the analysis suggested. However if you are removing subscribers who “report as spam” through email platform closed-loop reporting this shouldn’t be such an issue.

    • Dave, just using feedback loops (FBL) may not be enough. With the move to intelligent inboxes and engagement based deliverability removal of inactives can make the difference, a large number of inactive customers can drag down engagement metrics. Also if data management has not always been 100% spam traps can have built up and removal of inactives will address this, whereas FBLs will not.

      • Derek Harding commented on November 22, 2011

        I was going to say much the same as Tim. If you aren’t removing FBL complainants you’re simply doing it wrong. However even with this there are scenarios where you won’t receive the complaint notification or where an address doesn’t get removed when it bounces. In such situations some inactives may start to harm your deliverability and so some removal may help.

  • Great article! We have been wondering what to do with the subscribers who have not opened our newsletters in a while. Brilliant idea to send out a reactivation campaign then separate out those who don’t respond. Thanks for a great solution to a problem we have been having.

  • Derek Harding commented on November 21, 2011

    This is great advice of course. I think what happens with the culling inactives advice, as has happened with this advice is that the reasons driving the decisions aren’t clearly explained.

    The only reason to remove inactives is if they’re putitng a drag on your deliverability. There are two main ways inactive users can reduce deliverability. 1. becoming spamtraps — monitor your trap hits with tools such as SNDS and blocklist monitors. 2. lowering your reputation due to lack of engagement — monitor your reputation with senderbase and senderscore.

    If deliverability becomes an issue your oldest inactives are the prime candidates for culling to fix the problem.

    Could this result in a false positive whereby a long-term high spender who is still receiving your email but who isn’t responding getting culled? Of course it might. However if you’re careful and use Dela’s advice the chances should be minimized.

    • Lots of agreement going on here, deliverability is the only reason to remove inactives.

      If engagement is an issue then for a typical list of 50% inactives, dropping the frequency to those inactives by 75% will increase the total open/click rate by around 60%.

  • Good post Dela. I totally agree, you have to look at all the relevant data before deciding who is inactive and who is not. We’ve been using a similar strategy to define engagement for several years and the results are certainly worth the effort. I think the important point is that the timescales that define engagement will invariably be different for different businesses, products and stages of the customer lifecycle. Using the techniques you have suggested here is a good starting point for any marketing department that wants to make more revenue from their email program.

  • I agree with this and have shown value in subscribers who would have been considered ‘inactive’ with no open/click in over 8 months. Don’t just drop inactives, as often suggested. I’m talking on this very topic next Tuesday at the DMA http://www.dma.org.uk/civicrm/event/info?reset=1&id=93