Here's the advice you'll find everywhere: "Win back your inactive subscribers with a re-engagement campaign!"
And here's what nobody mentions. A re-engagement campaign is a deliverability decision before it's a copywriting decision — because you are, deliberately, about to send email to the exact group of people least likely to open it. That's the worst possible engagement signal you can generate, and if you do it carelessly you can damage placement for the subscribers who do still read you.
Which doesn't mean don't do it. Dormant subscribers are already hurting you just by sitting there. It means the order of operations matters enormously, and most guides skip straight to the copy.
The short version
- Segment properly first — by clicks and replies, not opens
- Send from a healthy position — never during an active deliverability problem
- Three emails over about ten days — direct, personal, one link
- Then suppress the non-responders. Not "hold." Suppress.
Step 4 is the one people can't bring themselves to do, and it's the one that actually works.
Why dormant subscribers cost you
Gmail, Microsoft, and Yahoo judge your sender reputation on engagement signals. If most of your list never opens or clicks, the filters draw the obvious conclusion: this mail isn't wanted.
Then the damage generalises. You start landing in spam for everyone — including the people who genuinely want to hear from you. The dead weight doesn't just fail to generate revenue; it actively degrades delivery for the subscribers who would have.
So you can't mail unengaged people forever. You try to wake them, and if they don't wake, you let them go.
Define "inactive" correctly — this is where people go wrong
A strong default is 90 days of inactivity. If you send daily, shorten it to 30 or 60.
But the definition of "inactive" matters more than the number, and here's the trap almost everyone falls into.
Never sunset on opens alone.
Email software has always tracked opens by embedding an invisible image and recording when it loads. Apple's Mail Privacy Protection broke that — Apple's servers now pre-load the image the moment mail arrives, whether or not a human ever looks at it. A meaningful share of your "opens" are machines.
Sunset on opens and you get the worst of both worlds: you keep Apple users who haven't genuinely read you in two years, while deleting real humans who read every issue in a desktop client that doesn't load images.
Treat opens as a rough directional signal. Treat clicks and replies as proof.
The segmentation that actually works:
| Segment | Definition |
|---|---|
| Hot | Engaged in the last 30 days |
| Warm | Engaged 31–90 days |
| Cool | Engaged 91–180 days |
| Cold | 180+ days — your exclusion list |
Your default campaign audience should be Hot plus Warm. A healthy list has 40% or more sitting in those two.
Before you send: check you're in a position to
This is the part that makes re-engagement risky, and it's why sequencing matters.
Don't run a re-engagement campaign while you have an active deliverability problem. If you're currently landing in spam, or you've just been blocklisted, or your complaint rate is elevated — deliberately mailing your least engaged segment is pouring petrol on it.
Fix placement first. Run the seed test, confirm you're reaching inboxes, and check your complaint rate is under 0.1%. Then run the win-back.
The right order is: repair, stabilise, then re-engage. Not the reverse.
The three-email flow
Direct, personal, minimal design, one link. These should look like a note from a person, not a campaign — partly because that converts better, and partly because plain personal email reads better to the filters you're currently on thin ice with.
Email 1 — the check-in (day 1). A low-friction ask for a micro-commitment.
Subject: still want these?
It's been a while since you opened an email from me, and I'd rather not clutter your inbox if this isn't useful anymore.
If you still want in, just click here: {one link — "Yes, keep me subscribed"}
If not, do nothing at all.
Tag anyone who clicks as engaged.
Email 2 — the question, or the best-of (day 4). Two approaches work here. Either bribe them back with your single best free resource, or ask a genuine question:
Subject: last check-in (1 question)
Before I remove you — one question, because it genuinely helps me:
What's the #1 thing you're struggling with when it comes to {topic}?
Just hit reply and tell me. I read every one.
(And if you want to stay on the list, replying does that automatically.)
That parenthetical is doing quiet work — it gives a reply two reasons to happen.
Email 3 — the breakup (day 9). Loss aversion, honestly deployed.
Subject: removing you tomorrow
This is the last email — I'm removing inactive subscribers tomorrow so I only email people who actually want this stuff.
Want to stay? Click here: {one link}
Otherwise: thanks for being here for a while, and good luck with {topic}. Genuinely.
The breakup email reliably outperforms the other two, because "you're about to lose this" moves people that "here's something nice" doesn't.
One rule: mean it. If you send "removing you tomorrow" and then don't, you've taught that segment your deadlines are theatre — and you've also kept the dead weight you were trying to shed. Both halves of that are bad.
The step everyone skips
If they don't click in email 3, suppress them.
Not a "maybe later" segment. Not "I'll try again in six months." Suppress.
I know this is the hard part. Watching your list count drop feels like watching your business shrink. So here's the reframe that makes it easier:
A 10,000-person list at 40% opens out-earns a 50,000-person list at 8% — and the small list's reputation keeps improving while the big one decays.
List size is vanity. Engagement rate is money. A subscriber inactive for six to twelve months has near-zero revenue value and negative deliverability value. You're not deleting an asset. You're removing a liability that's been quietly taxing every send you make.
The practical version: non-responders get suppressed, then deleted after another six months of silence.
Make it a routine, not an event
The reason people find sunsetting painful is that they do it once, dramatically, after years of accumulation — so it's a mass deletion rather than routine maintenance.
Schedule a monthly sunset review instead. One rescue sequence, then suppress, then delete. When it runs monthly, you're removing a trickle rather than performing surgery, and your Hot-plus-Warm percentage stays healthy on its own.
The alternative that saves more subscribers
One thing worth building alongside this: a preference centre.
A meaningful share of people who are about to leave will take "monthly instead of weekly" if you offer it at the moment they're deciding. Frequency options, topic options, and a pause button — plus a genuine one-click unsubscribe on the same page, because making leaving harder converts a clean exit into a complaint.
For your Cool segment specifically, "want fewer instead of none?" often outperforms a straight win-back attempt. They didn't dislike you; they were over-mailed. More on this in the list hygiene guide.
Frequently asked questions
When is a subscriber inactive?
90 days without engagement is a strong default, shortened to 30 or 60 if you send daily. But define engagement by clicks and replies rather than opens — Apple's Mail Privacy Protection pre-loads images and inflates open data, so sunsetting on opens keeps machine-opened Apple addresses while deleting real readers on desktop clients.
Should you delete unengaged subscribers?
Yes, after one rescue attempt. A subscriber inactive for six to twelve months has near-zero revenue value and negative deliverability value, because mailing them signals to providers that your email isn't wanted. Suppress non-responders after the sequence, then delete after another six months of silence.
Do win-back emails hurt deliverability?
They can, which is why timing matters. You're deliberately mailing your least engaged segment, which is a poor engagement signal. Run the sequence only from a healthy position — after confirming you're reaching inboxes and your complaint rate is under 0.1%. Never during an active deliverability problem.
What to do next
Before writing any copy, build the four segments — Hot, Warm, Cool, Cold — using clicks and replies rather than opens. Then check what percentage of your list is Hot plus Warm.
If it's under 40%, you've found the reason your open rates have been drifting, and re-engagement is the right project. If it's healthy, you may just need a monthly sunset routine rather than a campaign.
Free: The 60-Minute Email Authentication Fix — confirm your technical setup is sound before you mail a risky segment.
Go deeper: The Email Deliverability Playbook — the full re-engagement templates, sunset thresholds, and segment-by-segment send order.
Related guides
- Email automation: the sequences that make money — where re-engagement sits in the journey
- Why authenticated emails still land in spam — check this before you run a win-back
- Email list building and hygiene — the preference centre and the routine version
- Welcome email sequence — preventing dormancy in the first place
Join the Newsletter
Get practical marketing tactics delivered straight to your inbox.
Written by
Muhammad Basim
Related Articles
Klaviyo vs Mailchimp vs Kit: Which Fits Your Business
Note before we start: ConvertKit is now called Kit. Most people still search for the old name, so I've used both. The honest framing for this comparison is that these three platforms aren't really competitors. They're built for different businesses, and the "which is best" question mostly dissolves once you know which business you're running. […]
Abandoned Cart Emails: Why Timing Beats Discounting
Someone added your product to their cart. Then they left. Here's what people get wrong about that moment: they read it as rejection. It almost never is. Someone who reached the cart had already decided to buy — the deciding was done. Then a kid started crying, or a colleague appeared, or the card was […]
Email Automation: The Sequences That Make Money
Your broadcast emails go out when you have something to say. Your automations go out when the subscriber is ready to hear it. That's the entire difference, and it's why a welcome email sent thirty seconds after signup outperforms almost anything you'll send on a Tuesday afternoon to your whole list. Timing isn't a scheduling […]