Dormant Email List Reactivation: The Safe Playbook (Eligibility, Warm-Up, Stop Conditions, Measurement)

Updated 17 October 2026. This playbook describes the process we run inside List Revival. It works with any ESP and does not require migrating your list.

What "dormant" actually means (and why opens can't define it)

A dormant subscriber is a contact you are still allowed to email who has shown no evidence of a relationship for longer than your inactivity window. "Evidence of a relationship" means a click, a purchase, a reply, a site visit tied to the address, or a preference update. It does not mean an open, because Apple Mail Privacy Protection pre-loads images and registers opens that no human made.

The inactivity window depends on how often you send. A daily newsletter can call a reader dormant after 30 to 60 days of silence; a monthly sender needs 180 days or more before the label means anything. The table below is the one we use, and it matches what the major ESPs recommend in their own documentation.

Scroll for more → On small screens, choose cards for wide tables.

Two platform reference points: Klaviyo's own guidance treats a profile as unengaged when it has received at least five emails and has not opened or clicked in 90 days, and only suggests suppressing profiles created at least 180 days ago that received five or more emails in 72 weeks with zero engagement of any kind, ever. beehiiv's product guidance suggests 45 days for daily newsletters and 90 days for weekly ones.

Notice what these windows are not: they are not a rule to delete anyone. They define who needs a different treatment. The sunset-policy guide covers the deletion question separately.

Why blasting a cold list is how domains die

The arithmetic is short and unforgiving.

Suppose you have 100,000 addresses that have not been mailed in 14 months. Industry list-scanning vendors consistently report that 15–30% of such files are invalid or risky: dead mailboxes, abandoned domains, role accounts, and addresses that mailbox providers have quietly converted into recycled spam traps after months of bouncing. Take the low end: 15,000 addresses.

Send to all 100,000 at once and three things happen the same day:

1. Bounces spike. Thousands of hard bounces in one hour are a pattern mailbox providers associate with purchased or scraped lists. Your ESP may throttle or suspend the account before the send finishes.
2. Spam traps fire. A recycled trap doesn't bounce; it accepts the mail and reports the sender. One pristine trap hit can trigger a blocklisting; a cluster of recycled traps tells Gmail and Yahoo that you mail addresses nobody has used in a year.
3. Complaints cross the line. People who forgot they subscribed hit "Report spam." Gmail's limit is 0.30%: three complaints per thousand delivered. On a cold list, a complaint rate of 0.5–1% is not unusual. You cross the threshold on day one, and Google's FAQ says a sender above 0.3% stays ineligible for mitigation until the rate "remains below 0.3% for 7 consecutive days."

The damage is not confined to the dormant segment. Reputation is tracked per domain and per IP, so the engaged subscribers who open every week start landing in spam or Promotions too. Recovery from a bad Gmail domain reputation takes weeks of clean sending, and Microsoft's rejection code (550 5.7.515) means "we are not accepting this domain," not "this one message looked spammy."

The point of a playbook is to make the same 100,000 addresses safe to mail by changing four things: who gets mailed at all, in what order, how fast, and what stops the process.

For help putting this plan into practice, EngageMKTG offers dormant email list revival among its email marketing services.

Step 1 of 9

Step 1. Establish eligibility before planning

Before any segmentation, answer four questions for each record and store the answers as fields:

Scroll for more → On small screens, choose cards for wide tables.

"Hold" means the record is excluded from the revival segment until a human decides what to do with it. Missing data is recorded as "unknown," not assumed to be fine.

Legal floor in the US: CAN-SPAM lets you send commercial email without prior opt-in, but it requires a working opt-out honored within 10 business days, an accurate From and subject, and a physical address. Gmail and Yahoo add their own floor: visible unsubscribe link plus one-click unsubscribe headers, and spam rate below 0.3%. Twenty US states now have comprehensive privacy laws; if your list includes residents of those states, check that your consent records cover marketing use. For EU or UK residents, prior consent is required and a "re-permission" email to people who never opted in is itself a marketing email you cannot send.

Step 2 of 9

Step 2. Freeze the baseline

Reactivation is measured against a cohort that does not change after you define it. Do this before the first send:

• Export the eligible dormant segment with a cohort ID and a snapshot date.
• Record the baseline for that exact cohort over the previous 90 days: delivered, clicks, click-to-delivered rate, orders, revenue, unsubscribes, complaints. For most dormant cohorts the baseline is zero clicks, which is fine; zero is a baseline.
• Record the same metrics for your engaged segment. This is your control for "did the revival hurt the people who were already reading?"
• Decide the measurement window now: we use 45 days from the first revival send for the primary read-out and 90 days for attributed revenue.

Do not use open rate as a primary metric. Apple Mail's privacy proxy opens messages on the user's behalf; depending on whose data you use, Apple clients account for roughly 49% to 62% of recorded opens in 2026. An "open rate improvement" on a dormant cohort can be nothing more than a change in the mix of mailbox providers.

Step 3 of 9

Step 3. Reconcile identity and suppressions

Most dormant lists live in more than one system: the ESP, a CRM, an old ESP export, a support tool, an e-commerce platform. Before building the revival segment, merge them on the email address (lower-cased, trimmed) and apply every suppression you can find:

• Unsubscribes and complaints from every system, including the ESP you left two years ago.
• Hard bounces from any system. Klaviyo, for example, suppresses a profile automatically after a hard bounce or after more than seven consecutive soft bounces; other platforms have similar rules, and an address one system has given up on should not be re-tried from another.
• Role addresses (info@, sales@, admin@) and disposable domains.
• Customers who asked to be contacted only transactionally.

The output is two lists: the suppression list (never mail, keep forever so nobody re-imports it) and the candidate list. Keep the suppression list even after the project ends. Most "we got blocklisted" stories we see begin with a well-meaning re-import of an old CSV.

Step 4 of 9

Step 4. Segment by evidence of relationship

Consent and engagement likelihood are different things. A customer who bought in 2024 and never opened a newsletter has strong consent and weak engagement evidence; a newsletter signup from last month who clicked twice has both. Segment on evidence, keeping consent as a separate gate.

Scroll for more → On small screens, choose cards for wide tables.

Everything in E stays out of the revival stream. It is where most of the invalid addresses and traps live, and the revenue potential is lowest. If you only revive tiers B and C, you have already captured most of the value at a fraction of the risk.

Step 5 of 9

Step 5. Scan before you send

Run the candidate list through a verification and risk scan. A good scan returns, per address: deliverable / undeliverable / unknown (catch-all), known trap or trap-like pattern, dead domain (no MX), disposable, role account, and a domain-level risk score. Suppress everything that is undeliverable, trap-like, dead-domain or disposable, and hold catch-all addresses for the last cohorts.

Know what a scan cannot do. It cannot identify pristine spam traps (addresses that were never real users), and it cannot tell you whether a deliverable mailbox is still read. That is why Step 6 exists: the stream itself is the final test, run small enough that a bad surprise stays small.

Record the scan results as a snapshot: percentage invalid, percentage risky, percentage held. These numbers become your "dormant share" and "recoverable share" inputs for planning, and they are the first thing a mailbox provider's postmaster will ask about if you ever need mitigation.

Want the scan done for you? The free List Health Audit returns a bounce, trap and dead-domain scan, a dormancy map by tier, a domain risk score and a recoverable-revenue estimate within 72 hours. Get the free audit →

Step 6 of 9

Step 6. The 1/30 warm-up schedule

After the scan, the sendable segment is streamed, not blasted. The schedule we use:

Scroll for more → On small screens, choose cards for wide tables.

Why 1/30 rather than "start small and double"? Doubling schedules are built for new domains, where the goal is to build volume. A dormant list is the opposite problem: volume already exists and the goal is to keep daily risk flat. A fixed fraction means that if cohort 12 turns out to be full of bad addresses, you have exposed 3.3% of the list, not 50%.

The daily check is the part most teams skip. Our implementation automates it: a planner we call Manactor reads yesterday's bounce, complaint and engagement signals and decides whether today's cohort ships, shrinks or waits, and a status layer we call Live Marketing Status keeps every profile's current eligibility (consent, suppression, bounce state, frequency) in one place so a profile that bounced in cohort 3 cannot be re-queued in cohort 20. You can run the same logic by hand with a spreadsheet and a daily 15-minute review; the important thing is that the rules exist before day 5.

Step 7 of 9

Step 7. Stop conditions in numbers

Write these down before the first cohort and put them where the person pressing "send" can see them. These are our operating thresholds; they are deliberately stricter than the provider limits, because by the time Postmaster Tools shows a problem you are already days behind.

Scroll for more → On small screens, choose cards for wide tables.

Unsubscribes are a feature of this process, not a failure. A one-click unsubscribe is the cheapest signal you can receive, far cheaper than a complaint. Make it prominent.

Download the stop-conditions sheet: PDF · CSV · shareable PNG.

Step 8 of 9

Step 8. What to send: re-permission vs value reminder

Two kinds of first email cover almost every case. Pick by tier.

Value reminder (tiers B and C): a short, plain message from the brand people actually bought from or subscribed to, reminding them what they signed up for, showing the single most useful thing you have published or offered recently, and giving an obvious way out. No discount in the first email; discounts to people who have not engaged in a year train the list to wait for discounts and attract complaint-prone recipients.

Re-permission (tier D, and E after review): an honest question. "You subscribed to X on [date or context]. We haven't emailed you in a while. Do you still want these?" with one button to stay and one to leave, and a statement that no reply means you will stop. This email is a legal and reputational safety net: it documents renewed consent, and it removes the people most likely to complain before they can.

Both emails need: the usual From name and domain (not a new "friendly" sender), a subject line that says what the email is, a visible unsubscribe link in the body, one-click unsubscribe headers, and a physical address. Keep images light and links few; a long image-heavy template to a cold list raises filtering risk for no benefit.

Templates for both, with annotated examples, are in the re-engagement email examples guide.

Step 9 of 9

Step 9. Measure recovery: fixed cohorts, attributed vs incremental

At day 45, read out the frozen cohort against its own baseline and against the engaged-segment control:

• Reactivation rate: share of the cohort with at least one click or order in the window.
• Click-to-delivered rate and orders per thousand delivered for the cohort.
• Attributed revenue: orders placed by cohort members who clicked a revival email within the attribution window (we use 90 days). Label it attributed.
• Incremental revenue: attributed revenue minus what the cohort would have produced anyway. For a dormant cohort with a zero-click baseline, most attributed revenue is incremental, but say so explicitly rather than claiming it.
• Cost: suppressions, scanning, sending, your time, and any performance fee.
• Collateral: did the engaged segment's clicks, complaints and inbox placement change over the same period?

Two measurement traps. First, the Return Path finding: in their study of 33 retailers, 45% of people who received a win-back email went on to read a later message from the brand, but only 24% of those had read the win-back email itself, and the average gap was 57 days. The study could not prove causation, but it shows that judging a revival on the first email's open or click rate undercounts what you got back. Keep the cohort flag on these profiles and read the 90-day number.

Second, don't let opens creep back in through the side door. If your ESP's "engagement" score weights opens, build the read-out from raw click and order events instead.

Worked example

The numbers below are illustrative. They are not a forecast and not a measured lift; they show how the arithmetic works end to end.

A 10,000-record snapshot of contacts not mailed in 14 months.

Scroll for more → On small screens, choose cards for wide tables.

Sensitivity: at a 2% reactivation rate the attributed revenue is $1,700; at 8% it is $6,800. The cost side is mostly fixed (scan, setup, daily review), so the decision to proceed depends on list size and revenue per subscriber, which is what the calculator estimates before you start.

When not to revive

Stop before Step 1 if any of these is true:

• The list, or a large part of it, was purchased, scraped or acquired without consent evidence. Revival cannot fix provenance.
• Consent is older than 24 months, there is no relationship evidence, and you have no way to run a lawful re-permission email (EU/UK residents, for example).
• Your authentication is not in order: SPF, DKIM and DMARC must pass and align before any volume goes out. See the sender requirements tracker.
• Your engaged program is already in trouble (spam rate above 0.1%, visible Promotions or spam placement). Fix the live program first; a revival on top of a weak reputation accelerates the decline.
• You cannot commit to the daily review for 30 days. A revival without stop conditions enforced is a blast with extra steps.

Benchmarks

Public reference points, with the caveat that most come from vendors:

Scroll for more → On small screens, choose cards for wide tables.

As of 17 October 2026, the Dormant List Index has not been published. No measured median dormant share, invalid share or reactivation rate is available here; the defaults below are planning assumptions, not findings from that index.

Before you plan, estimate what is actually at stake with the dormant list value calculator, then confirm it with a scan.

FAQ

How long can a list sit before it's unsafe to email? It depends on how often you used to send. Past the inactivity window for your frequency (30–60 days for daily senders, 180+ days for monthly), treat the list as dormant and follow this playbook. Past 12 months, assume 15–30% of addresses are invalid and plan for re-permission copy on any tier without relationship evidence. Past 24 months without consent evidence, hold the records.

Do I need a re-permission campaign or just a re-engagement email? Use a value-reminder re-engagement email for contacts with a purchase or click in the last 36 months and consent on file. Use a re-permission email for contacts with consent but no relationship evidence. Hold everyone else until a human has reviewed the source.

What percentage of a dormant list is recoverable? Public data is thin. Consultants report up to 15% of inactive newsletter readers; the Return Path study found 45% of win-back recipients reading later messages. Our worked example uses 4% click-or-order reactivation as a conservative planning figure. The scan in Step 5 gives you the only number that matters: how much of your list is sendable at all.

Should I use a new domain or subdomain for reactivation? Not a new domain. Gmail applies an accelerated enforcement timetable to domains without sending history, and you would be building reputation from zero on your riskiest traffic. A dedicated subdomain of your main domain is acceptable if your ESP supports it and it is authenticated and aligned, and it can isolate the risk from your engaged program.

Can I do this inside Klaviyo, Mailchimp or beehiiv without migrating? Yes. Every step here is a segment, a schedule and a daily review. Build the tiers as segments, create one segment per day's cohort (or one segment with a rotating date condition), and send the same email to each. Platform-specific guides for Klaviyo, beehiiv and Substack walk through the clicks.

How long until I see results? Bounce and complaint signals arrive within hours of each cohort. Postmaster Tools lags by days, so expect the spam-rate dashboard to reflect a change 5–10 days later. Primary measurement is at day 45; attributed revenue closes at day 90.

Start with the free List Health Audit. Bounce, trap and dead-domain scan, dormancy map by tier, domain risk score and a recoverable-revenue estimate, delivered within 72 hours. No migration, no blast. Get the free audit →

Continue with these guides

• List revival
• Dormant email list calculator
• Email sender requirements
• Email sunset policy
• Re engagement email examples
• Emails going to spam all of a sudden
• Google postmaster tools v2
• Klaviyo inactive profiles
• Beehiiv inactive subscribers
• Substack open rate

Sources

• Google, Email sender guidelines and Email sender guidelines FAQ: spam-rate thresholds, 7-day mitigation rule, November 2025 enforcement ramp-up, error code 5.7.26, accelerated timetable for new domains. https://support.google.com/a/answer/81126 and https://support.google.com/a/answer/14229414
• Microsoft, Strengthening Email Ecosystem: Outlook's New Requirements for High-Volume Senders (April 2025, v4.0): 5,000/day, SPF/DKIM/DMARC, 550 5.7.515. https://techcommunity.microsoft.com/blog/microsoftdefenderforoffice365blog/strengthening-email-ecosystem-outlook’s-new-requirements-for-high‐volume-senders/4399730
• Yahoo, Sender Best Practices: 0.3% complaint rate, two-day unsubscribe, reconfirmation of inactive subscribers recommended. https://senders.yahooinc.com/best-practices/
• Klaviyo Help Center, Understanding active profile management: unengaged definition (5 emails, 90 days), suppression candidates (180 days, 5 emails in 72 weeks, zero engagement), automatic suppression after hard bounce or 7 consecutive soft bounces. https://help.klaviyo.com/hc/en-us/articles/24263920096027
• beehiiv, Automate a healthy email list (product update) and support articles on subscriber statuses and re-engagement automations: 45/90-day guidance. https://product.beehiiv.com/p/automated-reengagement-campaigns
• Return Path, win-back study of 33 retailers (2014), as reported by Validity, MarTech and MarketingProfs: 12% read rate, 45% subsequent reads, 57-day average lag.
• Inbox Collective, Not a Newsletter (April 2023): up to 15% of inactive readers won back (consultant experience).
• EmailListValidation (2026), vendor guides on dormant list revival: 15–30% invalid or risky addresses. Vendor figure.
• Stripo B2B benchmark (49.29% Apple MPP opens) and Litmus email client market share (July 2026, Apple family 62.26% of tracked opens, via Webtonic): Apple share of recorded opens.
• Mailpro, Omnisend, Suped: inactivity window ranges by send frequency (vendor guidance).
• FTC, CAN-SPAM Act: A Compliance Guide for Business: opt-out within 10 business days, physical address, accurate headers. Verify link at publication.