--- title: "Installs" description: "Track app installs and understand their impact on subscriptions with Adapty." --- The Installs chart shows how many users installed your app over the selected period. What counts as an install — and how each install is grouped — depends on the install-counting setting. This article explains how to select an appropriate counting mode, and how to [reconcile the potential discrepancies](#troubleshooting) between the different analytics sources. ## What counts as an install The Adapty SDK registers an "install" and sends it to Adapty when the user first launches your application. This has two consequences: - An install appears in Adapty when the user opens the app for the first time, which can be hours or days after they download it. - If a user downloads but never opens the app, Adapty doesn't count them. Your **reporting timezone** in App Settings governs which day each install lands on. An install at 23:30 UTC on June 1 lands on June 2 if your reporting timezone is +02:00, while App Store Connect or Google Play may show it on June 1. ### Counting modes The **Installs definition for analytics** setting determines what counts as a new install. To change it, open [App Settings → General → Installs definition for analytics](general#4-installs-definition-for-analytics). | Mode | What counts | Example | 3rd party metric | Potential discrepancies | | --- | --- | --- | --- | --- | | **New device_ids** (recommended) | **Every app install** — including reinstalls. Authentication, profile creation, and version upgrades don't add to the count. | One user across 5 devices = 5 installs.

Reinstalling on the same device = 2 installs. | App Store:
**Total Active Devices**

Google Play: **Devices** | **Higher than the number of app downloads** if reinstalls are common.

**Lower than the number of app downloads** if many users download the app without opening. | | **New customer_user_ids** | Only **the first install** per [identified user](identifying-users). Additional devices and anonymous users don't count. | One user across 5 devices = 1 install.

Reinstalling, logging back in = no new install.

Using the app without an account = no new install. | Sign-up stats from your app's authentication system | **Remains empty** if you don't identify users at all. | | **New profiles in Adapty** (legacy) | Counts every install and reinstall, **as well as anonymous profiles created on logout**. | One user, one device, 3 logouts = 4 installs. | None | **Higher than all external metrics**. Counts each anonymous profile created on logout as an install. | Use **New device_ids** unless you have a specific reason to switch. ## Troubleshooting ### Adapty's count is higher than App Store Connect or Google Play Two likely causes: - **Reinstalls.** If your [counting mode](#counting-modes) is set to **New device_ids**, Adapty counts both first launches and subsequent re-installs. App Store Connect's "Total Downloads" only counts the initial download. - **First launch date ≠ download date.** To. Stores attribute by download date. Late-opening users land on different days. To compare more cleanly, open **App Store Connect → Total Active Devices** or **Google Play → Devices**. Those metrics are device-level and closer to Adapty's **New device_ids** mode. ### Adapty's count is zero If your counting mode is **New customer_user_ids**, but you don't [iOS](identifying-users), [Android](android-identifying-users), [React Native](react-native-identifying-users), [Flutter](flutter-identifying-users), [Unity](unity-identifying-users), [Kotlin Multiplatform](kmp-quickstart-identify), [Capacitor](capacitor-quickstart-identify), Adapty won't register any installs In that mode, anonymous installs are excluded. Switch to **New device_ids** or implement user identification. ### Adapty's count differs from AppsFlyer or Adjust MMPs attribute installs by their own SDK init or first-touch event. Those fire on a different schedule than Adapty's first SDK launch — some discrepancy is normal. ## Available filters and grouping :::link Main article: [Analytics controls](controls-filters-grouping-compare-proceeds) ::: - ✅ Filter by: Attribution, Country, A/B tests, Segment, and Store. - ✅ Group by: Country, Store, Segment, and Attribution. ## Similar metrics For a side-by-side comparison of these metrics, see the [Metric comparison table](metric-comparison-table#subscribers-and-conversion). - [New subscriptions](reactivated-subscriptions) - [Active subscriptions](active-subscriptions) - [New trials](new-trials)