KADODEHow to get 16 testers for a closed test (not just 12)
Updated 2026-11-14 · Checked against Google's official help pages
Recruit from people who already own an Android device and know you or your app, such as friends, family, coworkers and existing users, and aim for about 16 rather than 12. If you start with exactly 12, a single person leaving during the 14 days puts you under the bar. Google's own help page says personal and professional networks are the most common way to recruit.
Why 16 and not 12
A replacement tester only counts once they have been opted in for 14 days of their own, so you can't patch a drop-out late in the test. The buffer has to be there from day one. The table shows the chance that at least 12 testers are still opted in after 14 days, if each tester independently leaves with the given probability. These are our own calculations.
| Testers at start | 5% leave | 10% | 20% | 30% |
|---|---|---|---|---|
| 12 | 0.54 | 0.28 | 0.07 | 0.01 |
| 13 | 0.86 | 0.62 | 0.23 | 0.06 |
| 14 | 0.97 | 0.84 | 0.45 | 0.16 |
| 15 | 0.99 | 0.94 | 0.65 | 0.30 |
| 16 | 1.00 | 0.98 | 0.80 | 0.45 |
| 18 | 1.00 | 1.00 | 0.95 | 0.72 |
With 12 testers and a 10% drop-out rate, you fall short more than 7 times out of 10. With 16, you make it 98% of the time. Real drop-outs tend to cluster (once one person leaves, others follow), so treat these numbers as optimistic. If your group is loosely committed, aim for 18.
A recruiting plan
- List people with Android devicesWrite down 20 to 25 people: friends, family, coworkers, classmates, followers, and anyone already using your web or iOS version. iPhone-only users can't install the app.
- Set up a Google Group and the closed trackCreate a Google Group for testers, add it to your closed testing track, and publish. After the first release, the opt-in link can take several hours to start working. Don't add anyone you plan to count to the internal track.
- Send a clear set of stepsNon-developers often stop after joining the group without opting in. The tester page tool builds a single link with steps for your app.
- Check opt-ins in Play ConsoleBeing on the list is not the same as being opted in. Check the opted-in count the day after you send invites.
- Follow up on days 1 and 2Message anyone who hasn't opted in yet. Because late joiners start their own 14 days, the sooner you spot a gap, the less it delays you.
A message you can send
Use it as is or rewrite it in your own voice. Stating why you're asking and how long it lasts up front tends to keep people in.
I'm getting my Android app, (app name), ready to publish, and I need people to join a short test. The store requires at least 12 testers to stay in the test for 14 days in a row.
The steps are here: (tester page link)
Please don't uninstall the app or leave the test for 14 days. If you can open it now and then and tell me what feels off, I'll use that to improve it. Joining shows me your email address; if you'd rather not, you can use a separate Google account.
If friends and users aren't enough
There are a few common places to find additional testers. KADODE has no relationship with any of them and takes no referral fees. None of them can promise the outcome of Google's review.
| Where | Upside | Downside |
|---|---|---|
| Reddit and Discord test-for-test communities | Free. Lots of developers in the same position | Testers are other developers, not your target users. You owe tests back, often daily across many apps. Some subreddits filter or remove promotional posts |
| Forums of no-code and low-code tools (for example MIT App Inventor, FlutterFlow, Adalo) | Long-running threads where builders help each other; people often share a similar background | Slower and informal. Ask within each forum's rules; most are not meant for advertising |
| Test-exchange apps (earn a slot by testing others) | Fast to start, large pool | Frequent reports of people installing and never opening apps. Upsells to paid plans. Daily load of opening many apps |
| Paid testers (freelance marketplaces, tester services) | Numbers arrive quickly | Costs money. Reports exist of buyers receiving a list of email addresses and seeing no actual use. Testers are rarely your intended audience |
Whatever you choose, people recruited from these places aren't the users your app is for. The production access form asks whether tester behavior matched expected production users, so it is easier to answer if real users form the core and outside testers are the buffer.
Policy points to keep in mind
We couldn't find any policy that names tester exchanges or paid testers. The closest is the policy on ratings, reviews and installs.
"We prohibit any manipulation of ratings, reviews, or install counts."
— Developer Program Policy: User Ratings, Reviews, and Installs
The same paragraph lists incentivizing feedback as an example, and the policy also prohibits inflating install counts by illegitimate means. It explicitly forbids offering rewards or incentives for ratings or reviews, so don't ask testers for a five-star review in return for anything. Testers can't post public reviews of test versions anyway; their feedback goes to you privately.
Related guides
Sources
- Play Console Help: App testing requirements for new personal developer accounts (checked 2026-11-14)
- Play Console Help: Set up an open, closed, or internal test (checked 2026-11-14)
- Developer Program Policy: User Ratings, Reviews, and Installs (checked 2026-11-14)
- Probability table: KADODE's calculation (binomial, assuming testers leave independently)