The first_open event in FirebaseAnalytics is showing bloated data. How to figure out the reason for such data anomaly?

2
votes

Recently I noticed sudden jump in the count reported for first_open event on firebase dashboard for my app.

Earlier App details:

  1. For Earlier app versions first_open count used to be around 60k per week(observed for say last 3 months)
  2. Firebase Analytics version throughout this period was 17.4.0

App update details:

  1. After app update release recently, first_open count suddenly doubled to around 115k per week(monitored it for last 3 weeks).
  2. Firebase Analytics SDK version was upgraded to 18.0.0 in the new app release.

Note:

  1. I had many releases in last 3 months but never saw such spike in the number for first_open.
  2. I tried to look for the property values on firebase dashboard but it seems firebase does not provide those details on dashboard.
  3. Our QA team has tested it enough and have not reported any bug that would force user to reinstall it.
  4. During Same period I saw a significant Rise in MAU, however the DAU and retention numbers does not support the spike in MAU.
  5. I did refer to other posts related to first_open on SO like how first_open is counted but did not find any solution to my query.
  6. I cross-checked on other analytics platforms with the event similar to first_open, and they are reporting numbers which are very less than firebase and on expected lines.

So my questions are:

  1. Is it any tracking/reporting issue related to first_open event on firebase? Due to which is it showing bloated numbers for first_open?
  2. Is there any way that I can check the first_open event with all the property-values for certain time duration. Currently I am not able to see it on Firebase dashboard.

Any help in making me understand this data anomaly is appreciated.

1
androidfirebasegoogle-analyticsanalyticsfirebase-analytics

1 Answers

0
votes

Linking your Firebase project to BigQuery lets you access your raw, unsampled event data along with all of your parameters and user properties. Here, you can do custom analysis of your data.

Google Analytics for Firebase detects anomaly if the expected data (based on historical data) is way too far from the actual data on a given time period. This usually happens if you have recent changes in your app such as new versions, campaigns or other mediums that brought a significant amount of users or logged events in your app.

You can also check if the reporting identity on your analytics settings was changed.

Lastly, it could be a bug on the SDK or the Firebase product itself. If you think this is the cause of the issue, then I suggest reaching out to Firebase support.