Segment Decay: Why Your Best Segment Goes Stale in 90 Days
The engaged segment you built in January is not the engaged segment you have in April. People unsubscribe, go quiet, and change jobs, and the segment does not tell you.
An email list decays: at 5 percent monthly churn, only 54 percent of a segment is still active after a year, a half-life of 13.5 months. HubSpot puts average annual database decay at 22.5 percent, and Apple Mail Privacy Protection inflates open rates, so the segment looks healthier than it is.
The half-life of a list
An engaged segment is a snapshot, not a permanent asset. Every month, a slice of it leaves: unsubscribes, hard bounces, quiet inboxes, and people who changed jobs and left their work address behind. At 5 percent monthly decay, the math is brutal and simple: after twelve months, 54 percent of the segment is still active, which is a half-life of 13.5 months. At 8 percent monthly, the half-life drops to 8.3 months.
HubSpot puts the average email database decay at 22.5 percent per year, and B2B lists lose 23 to 30 percent of addresses annually to job changes alone. The segment you built in January has lost a quarter of its members by December, and the members it lost are not the ones who were never going to buy. They are a random slice, including your best customers.
Where the members go
Some leave loudly: the unsubscribe, the spam complaint, the hard bounce. Those are easy to see and easy to remove. The quiet ones are the problem. They stop opening, stop clicking, and stop buying, and the segment keeps them because the segment only knows what they did, not what they stopped doing. A segment built on a 90-day engagement window will hold a member for three months after they went quiet, and the campaigns sent in that window go to someone who is not listening.
The job-change churn is the hardest to catch, because the address still works. The inbox belongs to someone else now, or to nobody, and the mail gets delivered, counted as sent, and ignored. B2B lists are especially exposed: a quarter of the addresses can turn over in a year, and the segment has no way to know.
Why opens lie about engagement
Apple Mail Privacy Protection, introduced with iOS 15 in 2021, preloads tracking pixels, so an open is recorded whether or not the person read the email. The open rate went up, and the meaning of the open went down. A segment built on opens now includes people who never saw the message, and the campaigns sent to it look like they perform better than they do.
Clicks, replies, and purchases are stronger evidence, because they require an action the privacy protection cannot fake. A segment built on clicks is smaller and more honest than a segment built on opens, and the honest segment is the one worth mailing. The open-based segment is a vanity metric with a send button.
Rebuilding the segment on clicks and purchases
Rebuild the engaged segment on behavior that cannot be faked: a click in the last 30 days, a purchase in the last 90, a reply, a site visit with a login. The segment will shrink, and that is the point. A smaller segment of people who actually engage will outperform a larger segment padded with preloaded opens, on revenue per send and on deliverability alike.
Then set the refresh cadence. Rebuild the segment monthly, not quarterly, because the decay runs monthly. And run a re-engagement check on the members who fall out: one win-back email, then suppression. The segment is not a list to grow, it is a list to keep honest.
Keep reading
Related guides on this site: building segments from orders, not pixels RFM segmentation by recency, frequency, and monetary value when to refresh an audience segment More from Creative that adapts to every customer segment.
Frequently asked questions
What is segment decay?
Segment decay is the slow loss of active members from a segment through unsubscribes, bounces, quiet inboxes, and job changes. At 5 percent monthly decay, only 54 percent of a segment is still active after a year.
How fast does an email list decay?
HubSpot puts average annual database decay at 22.5 percent. B2B lists lose 23 to 30 percent of addresses a year to job changes. At 8 percent monthly decay, a segment's half-life is 8.3 months.
What is the half-life of an email list?
The time for half the list to go inactive. At 5 percent monthly decay it is 13.5 months; at 8 percent monthly it is 8.3 months. The half-life is a better way to think about list health than the total size.
Should I trust open rates for segmentation?
No, not since Apple Mail Privacy Protection started preloading tracking pixels in 2021. Opens are inflated whether or not anyone read the email. Build segments on clicks, replies, and purchases instead.