AppsFlyer Industry benchmarks: Frequently asked questions

The benchmarks are based on certain data processed by AppsFlyer through its platform after it has been fully anonymized and aggregated. To ensure statistical validity, we follow strict volume thresholds and methodologies.

The data is updated on a monthly basis, usually by the 7th of each month to cover install data not from the previous month, but from the one before. For example, by June 7th, the complete data from April is added to the site (for more info on the reason behind this, please refer to the answer to the next question).

For app marketers, it’s all about post-install data. Therefore, we are showing cohorted data that provides insights into post-install behavior that covers at least 30 days.For example, data updated in the first week of May covers the full month of March. In this case, we can only show week 4 retention, but if you choose an earlier month in the drop-down menu, data for up to week 12 will be displayed — assuming of course it does not appear because it wasn’t statistically significant.

The percentage change compared to the previous month. Red indicates a drop, green an increase.

Our benchmarking methodology follows strict statistical thresholds including, among other things, a minimum number of apps with a minimum number of installs, and the exclusion of statistical outliers. If the drop-down menu does not include a country, platform, or date that you seek, it means that the data for that specific combination was not statistically valid.

Our benchmarking methodology follows strict statistical thresholds including, among other things, a minimum number of apps with a minimum number of installs, and the exclusion of statistical outliers. If the metric in question does not meet those thresholds for the requested data combination, it is not shown.

Every month the data is run and the tool is updated. To the extent there is enough data to surpass the strict thresholds, then previously missing metrics may be available.

The average represents the average among all apps included in the sample, excluding statistical outliers. The best is the 90th percentile or top 10% of performers.

We use the categories as defined by the apps themselves on the App Store and Google Play, with two exceptions: 1) Finance: To dive deeper into the general and highly fragmented Finance app store categorization, we grouped apps into the following sub-categories: Digital Banks: Banks with little or no physical presence. Traditional Banks: Banks with a physical branch and an established pre-digital presence. Financial Services: mobile payments, money transfers, budget management, credit reports, credit scores, credit cards and other financial tools Loans: Apps specializing in money lending. Investments: trade, cryptocurrency, stock market and other investment tools 2) Shopping: We separated between eCommerce, which covers most of the shopping category and includes general retail, fashion, and electronics, and all the other types of apps in this vertical — primarily groceries and marketplace.

Since Apple introduced its App Tracking Transparency (ATT) framework, and following the wide adoption of SKAdnetwork (SKAN) — the company’s aggregated deterministic attribution solution — an MMP can no longer accurately count the total number of non-organic and organic installs when the data is at a country level. This happens because both AppsFlyer as the MMP and SKAN measure non-organic and organic installs, but the data cannot be deduped on a country level because Apple does not, at present, provide any country data from SKAN (only campaign data which can be deduped via solutions like AppsFlyer’s SSOT).As a result, and unlike Android, the data displayed in iOS only covers overall installs, which is unrelated to the SKAN issue. A divide between organic and non-organic installs is not currently supported at this level of granularity. Similarly, CPI is also based on non-organic installs and cannot be shown at this level. Filter options in the iOS - SKAN & ATT drop-down menu only cover ‘Global’ for the same reason - Apple does not provide country level data from SKAN.

Filter options in the iOS - SKAN & ATT drop-down menu only cover ‘Global’ because Apple does not provide country level data from SKAN.