The last post in this series ended with a line that tends to land harder than most things we write: sending to your dead weight hurts delivery to your best subscribers. It is worth taking that apart properly, because the mechanics behind it explain a pattern that catches out almost every list that has been running for more than a year or two, and the explanation is more counterintuitive than most marketing teams expect.
Most lists carry a segment that opens nothing, clicks nothing, and has not engaged with a single email in months, sometimes years. The instinct is to treat this segment as background noise. Nobody is actively unhappy with you, nobody is complaining, and the contact is just sitting there, neither helping nor hurting. That instinct is wrong, and understanding why it is wrong is the single most useful thing you can take from this post.
Inbox providers score the sender, not just the send
Gmail, Outlook and Yahoo do not evaluate each email you send in isolation. They evaluate you, the sender, as an ongoing relationship with their users, and they build that picture from the aggregate behaviour of everyone you mail. When a meaningful share of your list has stopped engaging, the provider reads that pattern and adjusts how it treats your domain, not just how it treats the specific contacts who went quiet.
This is the part people find hardest to accept until they see it happen in their own data. A dormant contact does not sit in a separate, contained bucket marked low risk. They are part of the same calculation as your most engaged subscriber, and when you send to everyone at once, the provider is looking at the whole cohort's response, not isolating the good behaviour from the bad. A list that is forty per cent dormant does not perform like a list that is genuinely sixty per cent of its stated size. It performs worse than that, because the inactive forty per cent actively drags the average down for everyone, including the sixty per cent who would otherwise be getting excellent placement on their own.
We touched on this from the complaint angle in the last post: people who have lost interest are the ones most likely to mark you as spam, simply because it is the fastest way to make something stop arriving that they no longer want. But the damage from dormancy runs wider than complaints alone. A contact who never opens, never clicks, and never complains is still telling the provider something through pure inaction, and providers read low engagement as a weak signal in its own right, separate from anything as deliberate as a spam report.
Why this is genuinely counterintuitive
Most marketing teams have been trained, reasonably, to think about list size as an asset. A bigger list means more reach, more opportunity, more potential revenue on the next send. Removing contacts feels like throwing away something you built, and the natural resistance to suppressing or deleting a large chunk of a list is understandable on its own terms.
The trouble is that deliverability does not work on the same logic as reach. A list of fifty thousand contacts where thirty thousand are dormant is not a fifty-thousand-person asset with a dormant tail attached. It is closer to a thirty-thousand-person liability sitting directly underneath a twenty-thousand-person asset, where the liability is actively suppressing what the asset could otherwise achieve on its own. Removing or properly suppressing the dormant segment does not shrink your reach in any way that matters commercially, because that segment was never actually reachable in the first place. What it does is stop that segment from dragging down placement for the people who were always going to open, click and buy.
This is why we are cautious about framing this purely as a hygiene exercise, the kind of thing that gets filed under good practice and quietly deprioritised when the to-do list gets long. It is closer to active damage limitation. Every send to a fully dormant segment is not a neutral, low-cost action. It is a small, repeated act of reputation erosion that compounds with every campaign, and the businesses who treat it as optional tend to be the ones who eventually find their engaged subscribers landing in spam through no fault of their own.
What actually counts as dormant
This is where teams often go too narrow or too broad, and either gets the wrong answer.
Too narrow usually looks like defining dormancy purely on opens, which on its own is an increasingly unreliable measure. Apple's Mail Privacy Protection, which we covered in an earlier post in this series, automatically loads tracking pixels for a large share of recipients regardless of whether a human ever looked at the email, which means an open-only definition will misclassify a meaningful number of genuinely dormant contacts as engaged. Too broad usually looks like an arbitrary cut-off, such as ninety days with no activity, applied uniformly across a list without any consideration of what is normal for that particular business or sending cadence.
A more useful approach looks at engagement across a meaningful window, combining opens with clicks, and treating clicks as the stronger signal of the two given the pixel-loading problem. It also has to account for sending frequency. A contact who has not engaged in ninety days on a list that sends three times a week is a very different situation from a contact who has not engaged in ninety days on a list that sends once a month, and treating both the same way produces a definition that does not match reality for either case.
A reasonable starting structure, which should be adjusted to the specific sending pattern and list history:
- Engaged: any open or click within the last 90 days
- At risk: no engagement in 90 to 180 days, candidate for a re-engagement attempt before suppression
- Dormant: no engagement in 180 days or more, candidate for suppression or removal depending on how the contact was acquired and what consent basis applies
The exact windows matter less than having a defined, deliberate threshold that the business actually applies consistently, rather than relying on a vague sense that the list probably needs a clean at some point.
The re-engagement step before you suppress
Suppressing or removing a contact outright the moment they cross into the at-risk window is usually too blunt, because some contacts in that window will respond to a genuine attempt to win them back, and it is worth making that attempt before writing them off entirely. A short, focused re-engagement sequence, distinct in tone from a standard promotional send, tends to work better than folding the same content into the regular campaign cadence and hoping dormant contacts notice it.
The sequence does not need to be elaborate. A small number of emails, spaced over a couple of weeks, that ask plainly whether the contact still wants to hear from you and make it easy to either confirm interest or leave cleanly, will usually surface the contacts worth keeping. Anyone who does not respond to a clearly framed re-engagement attempt has given you a reasonably confident answer, even though they never clicked unsubscribe.
What happens after that re-engagement window matters as much as the sequence itself. Contacts who do not respond should move to genuine suppression, not simply skip the next campaign and reappear in the one after. A sunset policy, which the next post in this series covers in more detail, is the structural mechanism that makes this consistent rather than something that only happens when someone remembers to run a manual clean-up.
How this plays out for SaaS and for e-commerce
The shape of the dormant segment problem differs by business type, though the underlying mechanism is identical.
For product-led SaaS businesses, dormancy often accumulates from trial sign-ups that never converted and free-tier users who stopped logging in long ago but were never removed from marketing communications. The list looks like it is growing steadily, but a meaningful share of that growth is contacts who were briefly interested at sign-up and have shown nothing since. Because onboarding and product emails frequently share infrastructure with marketing sends, the dormant marketing segment's effect on sender reputation can spill over into the transactional and lifecycle emails that genuinely matter for activation and retention, which is a connection most engineering-owned sending stacks are not set up to notice.
For e-commerce businesses, dormancy tends to arrive in large batches tied to specific acquisition moments, most obviously Black Friday and the Christmas period, when one-off discount-driven sign-ups inflate the list with contacts who had no real intention of becoming repeat subscribers. A list that looked perfectly healthy in November can be carrying a substantial dormant cohort by the following spring, and because UK retail volume concentrates so heavily around those same peak periods, the timing means the damage from the previous peak's list growth is often still working its way through reputation just as the next peak season approaches.
What to check in your own data
A few practical starting points, all checkable without needing anything beyond your ESP and Google Postmaster Tools:
- What proportion of your active list falls into the dormant window as defined above, and has that proportion been growing without anyone deciding that was acceptable
- Whether engagement rate trends, viewed at the cohort level rather than per campaign, show a gradual decline that maps to when a particular acquisition batch was added
- Whether your transactional and marketing sends share infrastructure, since a dormant marketing segment's reputation effect does not stay neatly contained to marketing sends alone
- Whether there is currently any defined re-engagement or suppression process, or whether dormancy has simply been left to accumulate
If the honest answer to that last point is that nothing formal exists, that is the single most common gap we find when reviewing a sending programme, and it is rarely the result of anyone making a deliberate decision not to deal with it. It is simply a task that never had an obvious trigger to prompt action, right up until the moment reputation has already degraded enough to notice.
Where this connects to the rest of your sending health
Dormant subscriber damage rarely shows up as an isolated problem. It tends to surface alongside the rising complaint rate we covered in the previous post, and it sits underneath the sender reputation fundamentals that govern how providers treat your domain over time. Looking at any one of these signals on its own gives a partial picture. Looking at them together, against your actual list history and acquisition patterns, is what turns a vague sense that something is off into a clear, prioritised list of what to fix first.
Ongoing Monitoring exists for exactly this reason. Reputation and engagement trends move slowly enough that a single point-in-time check will miss the gradual decline, and fast enough, once they start moving the wrong way, that monthly visibility is the difference between catching it early and finding out from a drop in revenue.
If you have not checked your own list yet
If you are not sure how much of your list currently falls into the dormant window, the place to start is our free health check, which takes about a minute and gives you an immediate read on your underlying sending setup.
If you already suspect dormancy is part of what is going on, or the health check raises something worth a proper look, book a free inbox check-up call. Fifteen minutes, no obligation, and we will walk through what your numbers are actually showing.
A Deliverability Review and Fix goes further still, looking at engagement cohorts, suppression logic and sending history together, alongside authentication and infrastructure, so the fix addresses what is actually causing the problem rather than treating the symptom in isolation.
Not sure if this applies to you?
Book a free check-up and we will walk through your sending situation. No obligation, no pitch.
