Closed Testing Statistics: Installed Audience vs DAU

Which Play Console number counts toward the 12-tester rule?

Only one of them, and it is not on the Statistics page. Google writes the requirement in terms of opted-in testers: "At least 12 testers must be opted in to your closed test when you apply for production access, and they must have been opted in continuously for the preceding 14 days." Installed audience, DAU, installs and retention appear nowhere in that sentence. The number that gates production access is the counter in your app's Dashboard checklist, and 0 testers currently opted in walks through it.

The Statistics page serves a different purpose. Google defines Installed audience and Daily Active Users for every app on Play, closed test or not, and the definitions do not change because twelve people are testing your build. Treating those charts as pass/fail marks for your test is the confusion this article is about.

• Opted-in tester count - the only one that gates production access. Dashboard checklist item, and the Testers tab on your closed track: Google counts testers who completed the opt-in and stayed opted in.
• Installed audience - a diagnostic. Statistics page: users with the app installed on at least one device used in the past 30 days.
• Daily Active Users (DAU) - a diagnostic. Statistics page, metric picker: users who opened your app on a given day.
• Testing feedback - evidence, not a criterion. Ratings and reviews area: private feedback testers sent from the test listing, which you summarize in your application.

What does Installed audience actually count?

Google's definition is one sentence: Installed audience is "the number of users who have your app installed on at least one device and have used the device (though not necessarily your app) in the past 30 days." Three parts of it do the work. It counts users, not installs. The device must have been used in the past 30 days. And the user does not have to have opened your app. A tester who installs your build and never taps it still sits in Installed audience, which is exactly why that chart can look healthy while DAU sits flat beside it.

Two further details are on the same help page. The day boundary is not yours: "Installation data is based on Pacific Time (PT)," so the column you read as yesterday is a Pacific day, still being assembled while your own day is already half over. And the numbers are adjusted for privacy choices: "Some metrics are calculated based on data from users who have agreed to share their data with developers in aggregate. The metrics that we provide in Play Console are adjusted to compensate for users who have opted out of data sharing, so that metrics better reflect your entire user base."

Why is installed audience lower than your opted-in count?

Because opting in and installing are two separate actions, and the Statistics page only sees the second one. Google describes the sequence: after clicking the opt-in link, testers "receive an explanation of tester responsibilities and a link to opt in," and only then can they install - from the Play Store URL you sent them, because during a closed test "testers cannot find your app by searching Google Play. You must share the app's Play Store URL with testers so they can download it."

The counter and the chart are therefore reading different steps. The causes, in the order we check them:

• Opted in, not installed yet. The opt-in is the click on the web link; the install has to happen afterwards from the store link, on the same device and account.
• Installed under a different Google account than the one that opted in. The two events never join up and the chart stays where it was.
• The data has not been processed. Community reports put the wait for Installed audience at two to four days; Google publishes no number.
• The value is below Google's display threshold. Google states that it will not display data under certain minimum thresholds, and with twelve testers you are living near them.
• Someone uninstalled, or the device went quiet. Google's User loss metric covers users who "uninstalled your app from all of their devices or stopped using any devices that your app is installed on for over 30 days."
• You are reading one country row. Developers report the chart opening split by country or region with several rows selected at once; set the dimension to all countries and regions before comparing it with your tester count.

This gap is the mismatch we investigate most often in our own campaign logs - the records behind the 1,500+ closed-testing campaigns we have analyzed - and in almost every case it closes by itself once installs land and processing catches up.

Why is DAU blank on some days while installed audience has data?

DAU has a one-line definition: "The number of users who opened your app on a given day." A blank column therefore has three possible readings, and only one of them is bad news.

• Nobody opened your app that day. With twelve testers this is common, and it is the reading to act on: ask for the daily open rather than an explanation from Google.
• The value exists but sits below Google's display threshold. "The metrics that we provide in Play Console are adjusted to closely reflect data from all of your users. However, Google will not display data that falls under certain minimum thresholds." Blank is not the same as zero.
• It has not been processed yet. DAU is commonly reported as arriving later than installs, which is why developers see one metric populated and the other empty on the same date.

What Google does not publish is the threshold itself, or a per-day completeness guarantee. We re-read View app statistics, Changes to your Play Console data and Troubleshoot app statistics problems on 24 October 2026: none of them states a number, a delay, or a rule for blank engagement days.

None of this changes eligibility. The requirement is continuous opt-in, and Google's stated reasons for continued testing are about tester counts and engagement - not about a gap in a chart.

How delayed is closed testing data?

Google publishes no freshness guarantee for daily metrics. The closest it comes is an answer in its own metrics FAQ. Asked whether everything is available from day one, it replies: "No. Because of how data aggregation works for unique devices and users, weekly metrics will start to appear two weeks after launch. Monthly metrics will be available once there's a single full month's data." That statement is about weekly and monthly rollups rather than the daily chart, but it is the official acknowledgment that these reports are assembled, not streamed (Understanding recent changes to report metrics).

For daily data the numbers in circulation are community ones. Developers in Google's own help forum describe Play Console data running "around 4 days" behind, with one member reporting a known case of eleven days. A tester-service thread reports roughly four days for install audience to appear. A developer blog puts the usual lag at "two or three days," and AppTweak's documentation of the Play reporting API notes "usually a 3-day difference between the last available data in the console and today." None of those is a Google statement.

The rule we work to: judge Installed audience and DAU on data at least three days old, and treat the newest column as provisional. A chart that has not moved in two days is a chart that is behind, not a test that has failed.

Does Google read these statistics when it reviews your application?

No Google page says so, and we could not find anyone who can prove it. What is published is the application itself: Part 1 asks you to "provide details about tester engagement during your closed test," including "whether testers used all available app features" and "whether tester usage matched expected production user behavior." When Google asks for more testing, it names two reasons: "having fewer than 12 opted-in testers or insufficient tester engagement during the testing period."

So the engagement judgment is documented; the data source is not. Google does not say whether a reviewer opens Statistics, reads an internal dashboard, or works from your written answers. In one help-forum thread, a developer who asked Play Developer Support exactly that was told the representative was not on the team responsible for production access reviews and could not explain how they are conducted.

The honest reading is the one we gave when we asked do testers need to open the app every day?: absence of documentation is not proof that nothing is measured. Google is free to look at signals it never publishes, and the review is a review, not a formula. What you control is the record you keep and the description you will write - which is what the production access form questions are asking for.

What should you watch during the 14 days?

Four numbers, checked on different schedules:

• The opted-in counter, daily. It is the requirement, it lags by a few hours after each opt-in, and it is the only one of the four that can stop you applying.
• Installed audience, twice a week, read three days behind. Steady growth is what you want. A plateau while opt-ins keep climbing means installs are missing, not that testers left.
• DAU as a shape, not a number. A line alive on most days is the pattern you can describe honestly in your application. Google publishes no threshold to aim at.
• Your own tester sheet. In our campaigns we keep one row per tester separating invited, opted in, active and reporting - "active" is our definition, not Google's - because testers do not announce when they stop opening the build.

Ship an update during the window and keep the feedback you can point to; both feed the questions you answer at the end. And if lining up twelve people yourself is the blocker, how a campaign runs shows the alternative.

Closed testing statistics: symptom, cause, what to do

The six patterns that account for most of the messages we receive, with the first check for each.

• Installed audience reads 0 while the counter says 12. Usually testers opted in without installing, installed under another account, or the data is still processing. Send the Play Store link again with the instruction to install from it, then wait three or four days before judging.
• DAU has gaps mid-test. Either nobody opened the app on those days, or the value fell under Google's display threshold. Ask testers for a daily open, then re-read the chart once processing catches up.
• The Statistics page is empty for a new app. Reports are assembled rather than streamed, and weekly or monthly metrics take longer still. Check back after a few days and keep dated CSV exports as your own record.
• Numbers fall partway through the test. An uninstall, an opt-out, or a device that stopped being used - Google's User loss. Check the opted-in counter first, then see what happens if a tester uninstalls your app during the 14 days.
• The chart seems to show a single country. The chart is split by country or region with rows already selected. Set the dimension to all countries and regions, then compare with the counter.
• Nothing has updated for a week. Processing delays of days are reported in community threads, occasionally longer. Do not re-recruit off a stale chart; the counter and the chart update independently.

Frequently asked questions

Do I need 12 in Installed audience to apply? No. The requirement is 12 opted-in testers for 14 continuous days. But if Installed audience sits near zero while the counter says 12, your engagement evidence is thin - and insufficient tester engagement is one of the two reasons Google gives for asking you to keep testing.

Why does my Statistics page show nothing at all? New apps have little to aggregate, reports are assembled rather than streamed, and Google withholds values below its display thresholds. On weekly metrics Google is explicit that they "will start to appear two weeks after launch."

Can a blank DAU day break my 14 days? No. The condition is continuous opt-in, and Google's continued-testing reasons are tester counts and engagement, not gaps in a chart.

Is Installed audience the same as installs? No. Installs are events; Installed audience counts unique users who still have the app on a device used in the past 30 days, so one person with two devices counts once.

Can I download these statistics? Yes. Google's statistics guide describes configured reports you can save and download as CSVs, "generated daily and accumulated in monthly CSV files" - the simplest way to keep a dated copy.

Will Google tell me my engagement numbers are too low? Only through the outcome. "Review usually takes seven days or less, but can occasionally take longer" - see how long a production access review takes - and the answer may be a request to keep testing, as in Google Play production access rejected. There is no live gauge that says your DAU is sufficient.