Testers opted in but your count is short: 6 causes
Updated
Fourteen people told you they opted in. Play Console says nine. Nobody is lying, and the console is not broken.
There are six causes, and they account for essentially all of these cases. Work down the list in order — the first two are far more common than the rest combined.
1. They installed but never opted in
This is the cause about eight times out of ten. Opting in is a specific action: follow the opt-in URL, sign in, press accept. Getting the app onto the phone by any other route — a direct store link, an APK you sent, searching the store — produces an install and no enrolment.
How to tell: ask one of the missing testers to open the opt-in link again and describe what they see. If it says they are already a tester, they are fine and the problem is elsewhere. If it offers them a button to join, they never pressed it.
Fix: resend the opt-in URL specifically, with the instruction to press the accept button. The full method is in how to verify a tester actually opted in.
2. Wrong Google account
The tester gave you one address and is signed into the Play Store with another. They opt in on the account that reads the email, then install on the phone signed into a different one — or they simply gave you the address they use for email rather than the one on their phone.
How to tell: ask them to check which account is signed into the Play Store app on the device that will keep your app, and compare it to what you were given.
Fix: add the correct address and have them opt in again on that account.
3. You are looking at the wrong track
Internal testing, closed testing and open testing are separate tracks with separate tester lists. Testers added to internal testing do not appear in your closed testing count, and internal testing does not satisfy the requirement at all — see closed testing vs internal testing.
How to tell: check which track the count you are reading belongs to, and which track your opt-in link came from. If you have been sharing an internal testing link, everybody followed it correctly and none of it counts.
4. Google Group changes have not propagated
Attaching a Google Group to a track is convenient and it is not instant. Group membership changes, and the group-to-track association itself, can take hours to be reflected. A group whose members joined ten minutes ago will not show up yet.
How to tell: the change is recent, and the count is rising slowly rather than not at all.
Fix: wait, then re-check. If it has been a day and nothing has moved, switch to an explicit email list — it is less elegant and it is immediate.
5. Country availability
A closed testing track is available in the countries you selected. A tester outside them cannot reach the opt-in page, so they cannot opt in.
How to tell: the missing testers are the ones not in your own country, and they report that the link does not work or the app is unavailable.
Fix: add their countries to the track's availability, or recruit testers where your track is live. If you are trading tests with developers worldwide, setting availability broadly during the closed test costs nothing.
6. They opted out again
The count is current, not cumulative. Someone who opted in on day two and left on day five is simply gone from it, and Play Console will not tell you it happened.
How to tell: the count went up and then came down, which you only know if you were checking. If you were not, you cannot distinguish this from cause 1.
Fix: replace them, and start checking every few days. What happens if a tester opts out covers the cost of the gap.
The pattern underneath all six
Five of the six are the same mistake in different clothes: the developer trusted a report instead of reading the number. The count in Play Console is the only thing Google measures the requirement against, so it is the only thing worth treating as true.
Build the habit of reconciling your own invite list against that number every few days, and five of these six become things you catch on day two instead of day fourteen.