How to Write a Sunset Policy for Inactive Email Subscribers

A sunset policy is a written rule for when you stop emailing subscribers who have stopped engaging. Since Apple's Mail Privacy Protection, "no opens in 90 days" no longer works as the trigger, because opens are recorded for mail nobody read. Define inactivity by clicks, replies, purchases or logins instead, and send one re-engagement attempt before suppressing.

Why you need one at all

Mailbox providers ask you to stop mailing people who do not engage, in writing. Google's email sender guidelines say to "send email only to people who want to get messages from you" and, more directly:

"Consider unsubscribing recipients who don't open or read your messages"

The same page asks senders to "keep spam rates reported in Postmaster Tools below 0.10% and avoid ever reaching a spam rate of 0.30% or higher". Inactive subscribers feed that number in two ways. Some of them eventually press "report spam" on mail they forgot signing up for. Others stop reading because the address is no longer theirs, and abandoned addresses do not stay abandoned forever.

On that second point, Google's own inactive account policy says: "An inactive Google Account is an account that has not been used within a 2-year period", and that for such accounts "all of its content and data can be deleted". A Gmail address you have mailed without response for years may not exist next month. Old dead addresses are also how spam traps get onto lists, which we cover in how spam traps get on your list.

Why opens stopped being a usable signal

An open is recorded when an email client downloads a tracking pixel. Apple's Mail Privacy Protection downloads that pixel whether or not anyone reads the message.

Apple describes the feature in its Mail user guide:

"When this option is selected, your IP address is hidden from senders and remote content is privately downloaded in the background when you receive a message (instead of when you view it)."

Remote content includes the open-tracking pixel. So for any subscriber reading in Apple Mail with this setting on, your platform logs an open at delivery time. This breaks an open-based sunset policy in both directions:

Subscriber	What they actually do	What open tracking reports	Open-based sunset outcome
Apple Mail user, MPP on, never reads	Ignores every email	Opens every email	Never sunsetted, keeps receiving mail
Apple Mail user, MPP on, reads	Reads some	Opens every email	Kept (correct, by luck)
Reader whose mail client blocks images	Reads text, never loads images	No opens	Sunsetted despite reading
Address abandoned, mailbox still accepting	Nobody there	No opens	Sunsetted (correct)

The first row is the dangerous one. The people an open-based rule protects most are the ones who may never have looked at your mail. The third row loses you real readers.

We are not giving a percentage for how much of a typical list uses Apple Mail with MPP, because it depends entirely on your audience and we have no sourced figure that would apply to your list. Your ESP's client breakdown report is the place to look for your own share.

The signals that still mean something

Use actions that require a human to do something after the email arrives. Rank them by how strongly they show intent.

Signal	Strength	Notes
Reply	Strongest	Rare for newsletters, common for B2B and cold email
Purchase or conversion attributed to email	Strong	Needs tracked links or matching by customer ID
Click on a content link	Strong	Exclude unsubscribe and preference-centre clicks
Login to your product or site	Strong	Comes from your app data, not your ESP
Form submission, survey answer	Strong
Open	Weak, unreliable	Use only to keep people, never as the sole reason to keep them

One caution on clicks: some corporate security systems scan links in incoming mail, which can register clicks that no person made. If you see clicks on every link in a message within seconds of delivery, treat them with the same suspicion as Apple opens.

A worked sunset policy

Below is a complete policy for a fictional business that sends one newsletter a week to a mixed consumer and B2B list. Every threshold here is a choice made for this example, not an industry rule. The reasoning behind each number is written next to it so you can adjust it for your own sending frequency.

1. Definitions

• Engaged: at least one reply, non-unsubscribe click, purchase, or product login in the last 120 days.
• Lapsing: no engagement signal for 120 to 180 days.
• Inactive: no engagement signal for 180 days.
• New: subscribed within the last 60 days. Excluded from sunsetting.

Why 120 and 180 days: at one send a week, 120 days is about 17 emails. A subscriber who has ignored 17 messages has had plenty of chances. A monthly sender would need a much longer window to give the same number of chances.

2. Lapsing stage: reduce frequency

Move lapsing subscribers from weekly to monthly sends, using your best-performing content. This lowers the volume of mail going to people who are not reading while keeping the door open.

3. Verify before re-engaging

Before the re-engagement email goes out, run the inactive segment through an email verifier. Addresses that have been silent for six months are the most likely part of your list to have been closed. Remove anything that returns invalid. Treat anything returned as risky, such as catch-all addresses, as a separate group and send to it last.

Doing this step first means your re-engagement campaign, which is by definition going to your least engaged people, does not also produce a burst of hard bounces.

4. Re-engagement: one or two emails, then stop

Send a short, plain message that asks one question: do you still want these emails? Make keeping the subscription a single click, and make leaving equally easy. A second reminder seven days later is reasonable; more than that starts to look like the behaviour the policy is meant to prevent.

Anyone who clicks to stay is reset to engaged. Anyone who does nothing within 14 days of the last re-engagement email moves to suppressed.

5. Suppress, do not delete blindly

Suppressed subscribers stay in your database with a status and a date so they are never re-added by an import, a form resubmission sync, or a CRM integration. If they sign up again themselves, through a fresh opt-in, they come back as new.

Whether you also delete their data after a period is a retention decision for your privacy policy, not a deliverability one.

6. Review

Check the policy twice a year. If your re-engagement emails get almost no clicks, your windows may be too long. If people you suppressed keep re-subscribing, they may be too short.

Putting it into practice in your ESP

Most email platforms let you build segments from last click date and last purchase date, even when they lead with open data in their dashboards. The logic, written out, is:

engaged   = last_click >= today-120
          OR last_reply >= today-120
          OR last_purchase >= today-120
          OR last_login >= today-120
          OR signup_date >= today-60

lapsing   = NOT engaged AND last_signal >= today-180
inactive  = NOT engaged AND last_signal < today-180

Where last_signal is the most recent of click, reply, purchase or login, or the signup date if none exist. Leave opens out of the formula entirely. If you want to use them at all, use them only as a tiebreaker for keeping someone, never as the reason.

If your ESP cannot see logins or purchases, export them from your app or shop weekly and import them as a custom field. That one integration is what makes the policy work without opens.

Cold email lists need a different version

For cold outreach there are no subscribers to sunset, but the same principle applies at the sequence level: a prospect who has not replied after your final follow-up should leave the sequence and go on a do-not-contact list for a defined period, and re-verification should happen before any later campaign. We discuss sequence length separately in how many follow-ups a cold email sequence should have.

Takeaway

Write the policy down, define inactivity with clicks, replies, purchases and logins, and treat opens as noise. Verify the inactive segment before you try to win it back, send one or two honest re-engagement emails, then suppress. Pick your windows from your own send frequency and say why in the policy, so the next person who edits it knows the numbers were a decision, not a law.

You can check single addresses from a lapsing segment with our free email verifier, and see how often a whole list should be re-checked in how often to re-verify your email list.