Data discrepancies
Discrepancies between measurement sources are expected. For mobile app stores like the Apple App Store and Google Play, differences in methodologies and attribution models can cause Adjust to report higher or lower values than Apple or Google.
There are common reasons for that, but there are two primary things to remember:
- Some differences normalize over time. After integrating the Adjust SDK into an existing app, initial spikes are common during the first weeks or months (depending on the app’s scale and user base). These values typically stabilize as historical traffic patterns level out.
- A certain level of discrepancy is normal. However, large or persistent gaps may indicate configuration or technical issues. For example, if Adjust reports significantly fewer installs than the app stores, the SDK configuration may require review.
Below is a list of typical concerns and possible reasons for the different behaviors.
What is data discrepancy?
A data discrepancy occurs when multiple comparable data sets are examined and the reported values do not match. Examples include Adjust reporting different event totals than a partner platform, installs being attributed as Organic instead of to a specific partner, or events not appearing on the partner’s side.
Some discrepancies can be resolved quickly by verifying configuration. Confirm that all of the following conditions are met:
- The integration with the specific partner is correctly set up and configured.
- The correct filters are applied in the Adjust dashboard.
- Adjust is instructed to share the relevant data with the partner.
- Adjust can identify the device.
- No additional data sources are sending duplicate events.
Why do discrepancies occur?
Discrepancies in app installs, user conversions, and in-app events are common and expected in attribution measurement. A mismatch does not indicate that either source is reporting inaccurately. Instead, it reflects differences in how each platform defines, captures, and processes data. These differences may stem from technical implementation choices, regulatory requirements, or platform-specific business logic.
As a neutral MMP, Adjust works closely with partner platforms to reduce discrepancies wherever possible. However, it is important to understand the potential causes, recognize common discrepancy patterns across platforms, and apply the appropriate steps to reconcile your data.
To learn more about the top mobile fraud types that impact datasets, read our blog post Discrepancies in data: Why don't the numbers always match up?
Common reasons for discrepancies
If configuration is correct, differences may result from variations in measurement approaches. Typical causes include:
Duplicated event count in the dashboard
If the same event is reported via SDK and S2S simultaneously, the event count will be duplicated on the Adjust dashboard. This happens because both data sources capture the same events under the same name.
Attribution window discrepancies
Adjust attributes installs to the source of install, unless specified otherwise under reattribution settings. However, partners have limited attribution windows. Google uses 30 days, Facebook uses 60 days, etc. This leads to discrepancies when comparing data over broader date ranges, such as 90 days.
Date-related differences
Self-attributing networks (SANs) attribute events to the date of the click, not the date of the events. This leads to different event numbers when comparing data across a given date range.
Reattribution settings
Depending on your reattribution settings, Adjust may attribute post-reattribution activity to a new partner, resulting in discrepancies on Adjust and the original partner's dashboard.
Temporary attribution
If a user installs an app using Partner A, which has a temporary attribution window of x days, all post-install activities performed by this user in this time frame are attributed to Partner A. Afterwards, activities are attributed to the fallback Organic channel.
Watch the Adjust webinar
In this recorded training session, Adjust's Implementation team teaches you everything you need to know to identify and resolve data differences.
Platform-specific data discrepancy troubleshooting
To learn how to resolve data discrepancies by platform, visit the relevant page of the platform you're interested in.