LinkedIn
Twitter
I'm currently using data transfer to upload data into BigQuery from a google ads account. I've noticed that the video quartile metrics (video_quartile_X) in the "AdCrossDeviceStats" table don’t match up with the figures in the report generated by the Google Ads UI.
Here are the images for the same ad on the same day:
Data from the Data Transfer Report:
Data from Google Ads UI:
To calculate the video quartile metrics in the Data Transfer (tables in bigquery) report, I use this weighted average formula: sum(video_percentile_x * views) / sum(views)
So far i have tried:
Checking the Math: I made sure the weighted average was calculated properly. Turns out, the discrepancies are there even before using the weighted average.
Testing Timezones: I tweaked the timezones to see if that was the issue, but no luck there.
Looking for Fixes: I searched to see if others had the same problem and found some potential fixes. Unfortunately, none fit my situation since I need the data grouped by Ad ID.
