Data Discrepancies between Meta Ads and Airbridge
Written by Jusang Song,Data discrepancies may occur between Meta and Airbridge due to varying reasons. Read on to learn the primary causes of the discrepancy. For more details, you may also refer to this article from Meta.
Attention
Make sure the channel integration with Airbridge is completed before checking up on the causes of data discrepancy.
The differences in the attribution rules of Meta and Airbridge may cause the data discrepancy. In particular, the definition and configuration of the attribution window differ.
The following attribution settings are supported by Meta.
Click-through: 1 day, 7 days
View-through: 1 day
Engaged-view: 1 day
For more details, refer to this article from Meta.
The winning touchpoint is determined from the touchpoints collected within the lookback window.
Lookback window: 7 days from the day a click occurred by default
Attribution window: 30 days by default
The lookback window and the attribution window can be configured on the [Management]>[Attribution Rule] page. For more details, refer to this article.
Meta attributes conversions to Meta when there has been any interaction with an ad from Meta ads within the attribution window. Airbridge, however, collects touchpoints from various ad channels and attributes conversions to a winning touchpoint based on the attribution model used in Airbridge. Therefore, Airbridge may attribute some of the conversions claimed by Meta to have been driven by Meta ads to ads from other ad channels.
For more details on this issue, refer to this article.
When you start using Meta’s AEM, unusually large data discrepancies may occur in your iOS campaign data, as Meta will use additional data for attribution, whereas the AEM won't have any influence on the attribution model in Airbridge.
For more details on Meta’s AEM, refer to this Airbridge or this article from Meta.
Meta aggregates event data using web pixels, mobile SDKs, conversion API, or conversion modeling and performs attribution. Airbridge mainly uses ADIDs and Airbridge Device IDs to recreate the user journey by connecting the event data with the winning touchpoints.
For more details on the user identifiers Airbridge is using, refer to this article.
Meta Ads Manager reports ad performance based on the date of the ad impression or click. By default, Airbridge reports ad performance based on the event timestamp. This discrepancy in the reporting of aggregate data may cause data discrepancies.
In the Actuals Report, you can use the Date Option feature to freely configure how you want to view the ad performance results. When selecting the is based on Touchpoint Date option, ad performance results will be reported by the date of the touchpoint and mitigate the data discrepancies. Refer to the following article to learn more.
When it comes to receiving conversion event data, Meta uses postback which causes a delay of several hours before the data is displayed in the Meta Ads Manager. On the other hand, Airbridge generally displays conversion events in real-time on the dashboard.
When the time zones for Meta and Airbridge are set differently, the data will be displayed differently, causing discrepancies. Refer to the following article to understand the app time zone used in Airbridge.
At first glance, the iOS campaign performance may appear more successful in Airbridge due to differences in SKAN data display compared to Meta Ads Manager because Meta Ads Manager displays only the click-through attribution data in the reports, whereas Airbridge displays the sum of the click-through and view-through attribution data in the reports.
For more details, refer to this article.
Airbridge identifies ad fraud based on the fraud validation rules you configure in your Airbridge dashboard. The fraudulent conversions and touchpoints are processed according to the prevention level you set - they can be marked as fraud but still attributed or not attributed at all. Refer to this article to learn more about the prevention levels.
Meta, on the other hand, is known not to have a fraud validation rule, which may cause a discrepancy in data.
このページは役に立ちましたか？