Where do the Review My Emails calculator's numbers come from?
Still have a question, spotted an error, or have a better explanation or a source we should cite?
Here's what I don't get. Your list is the most precious thing you have in email. No list, no email marketing. And yet I watch people email the whole thing, every week, for years, and never once check who's still there. You'd never pay rent on an empty shop. So why pay to email people who left years ago?
Your list is shrinking while you read this
Nobody unsubscribes from a job they've left. They just leave. The address dies, and your email marketing tool keeps it, counts it, and keeps sending to it.
Take a business list of 10,000 addresses. At the middle rate, 28% a year, about 2,800 of them are gone twelve months from now. Not unsubscribed. Gone. Two years from now, about 5,184 are left.
How long until half your list is gone?
| Careful | Middle | Aggressive | |
|---|---|---|---|
| Business | about 2.4 years | about 2.1 years | under 2 years |
| Mixed | about 5.4 years | about 4.3 years | about 3.1 years |
| Consumer | about 8.3 years | about 5.4 years | about 3.5 years |
Why three numbers, not one
Careful is the low end of what the evidence supports, aggressive is the high end, and middle is the likeliest. A range only gives you the two ends. The calculator needs a real number to do its sums, so it runs all three. You see the best case and the worst case, not one number pretending to be exact.
How we built them
Nobody has measured list decay directly in a way you can check. So we built it from the pieces that are measured.
The yearly rates count people who have left the address, bounce or no bounce, plus people who unsubscribed. They don't count people who are still there and ignoring you, so your real number can be worse.
Business lists: 25, 28 or 30% a year
Work addresses die when people change jobs. The median American stays with one employer for 4.1 years (BLS Employee Tenure, January 2026). About 3.2% of US workers leave a job every single month (BLS JOLTS, July 2026, preliminary). Every one of them leaves an inbox behind, and nobody tells you.
Consumer lists: 8, 12 or 18% a year
Personal inboxes fade more slowly, but they fade.
- Google deletes accounts that go unused for two years (Google).
- Microsoft closes them after two years (Microsoft).
- Yahoo stops delivering to a mailbox after 12 months (Yahoo).
Domains die too. Only 75.3% of .com and .net names were renewed in Q3 2025 (Verisign). Every domain that lapses takes all of its addresses with it.
Mixed lists: 12, 15 or 20% a year
Most lists are a bit of both, so this band blends the two. It's the default when you don't tell the calculator which kind you have.
And the people who leave on purpose
Unsubscribes run about 0.2 to 0.3% per campaign (Klaviyo). That's about 0.22% across Mailchimp users and 0.19% for ecommerce (Mailchimp). Small per send, but send every week and it adds up. As an upper limit only, paid newsletters lose 5.06% to 16.67% of their paying readers a month (beehiiv, 2026).
The numbers you've seen everywhere
You'll find decay figures all over the internet. Here are the two big ones, linked, so you can read them yourself. Neither says how it was measured in a way you can check.
- "22.5% a year." This one goes back to a single MarketingSherpa claim, repeated on HubSpot's Database Decay page. We couldn't find a sample size, a method, a place or a date behind it.
- "At least 23% a year." From ZeroBounce's email list decay report, built on over 11 billion addresses checked in 2025. That's a huge sample. But it measures how much of the lists people upload is bad, including lists that were bad on day one, not a clean list getting one year older. It also comes from a company that sells validation, like we do, so read it with that in mind.
We use the second one as one input for business lists, never on its own.
"But my dashboard says 98.5% delivered"
It probably does. Most email marketing tools count "delivered" as sent minus bounced. With bounces around 1.5%, that shows as about 98.5% delivered. (That's Validity's illustration, not a measurement.)
Delivered means a server accepted it. It doesn't mean a person saw it. In a survey of over 1,100 senders, only 12% could define delivery rate (Sinch Mailgun, 2025). If the experts can't agree on what the number means, it's not something to relax about.
Where the inbox figure comes from
Validity tests inbox placement with seed addresses at hundreds of mailbox providers, counting only senders who agreed to share. In 2024, 83.5% of marketing email reached the inbox, 6.7% went to spam and 9.8% went missing. Missing means it never showed up anywhere (Validity, 2025 report). For 2022 data the inbox figure was 84.8%.
By provider in 2024:
- Gmail: 87.2%
- Yahoo and AOL: 86.0%
- Apple: 76.3%
- Microsoft: 75.6%
So roughly one email in four sent to Microsoft and Apple inboxes didn't land in the inbox. These are seed tests from marketing senders, not your list. A real list mixes engaged people and quiet ones, so yours can land higher or lower.
Opens don't tell you who's alive either
Apple's privacy opens passed 50% of all opens by May 2022 (Litmus). A lot of your "opens" may be Apple's servers, not a person, and spam filters can open and click too. That's why the calculator doesn't count an open as proof someone is still there.
What it does to your bill
Most email tools charge by the size of your list, so people who left keep costing you. Here's what taking them out does on two tools, at the prices we checked on 1 Nov 2026: Klaviyo's Email plan and Mailchimp's Standard plan, at the regular price after the intro offer. These are monthly bills, before the cost of cleaning.
| List | Klaviyo | Mailchimp |
|---|---|---|
| 50,000 contacts, mixed, 15% gone | $720 down to $650, saves $840 a year | $450, no change |
| 100,000 contacts, mixed, 15% gone | $1,380 down to $1,205, saves $2,100 a year | $800, no change |
| 100,000 contacts, business, 28% gone | $1,380 down to $1,070, saves $3,720 a year | $800 down to $630, saves $2,040 a year |
See the two "no change" rows? Prices move in steps, so a smaller list doesn't always mean a smaller bill. And Mailchimp keeps counting unsubscribed contacts until you archive them, so the saving only comes once you do. Prices change and you may have add-ons, so check your own billing page.
What one email is worth
Klaviyo reports about $0.11 in campaign revenue per recipient (Klaviyo). Validity's model puts an inboxed marketing email at about $0.10, which is a model, not a measured study. Now multiply that by the people your list has quietly lost.
So what do you do about it?
Take the dead addresses out and your numbers go up. That's maths, not a trick. Fewer people who can't read you, so more of what you send reaches people who can.
Run the calculator to see what decay has likely done to your list. Then clean your first 1,000 free at Review My Emails.
The most useful email questions, once a month
One email a month. New questions worth reading, real deliverability changes (Gmail and Outlook updates, etc.), and any tool or report that landed since the last issue. No fluff, no upsell.
Contributors
Who worked on this answer
Every name links to their profile. Every company links to their site. Real people, real accountability.
