Google Ads data transfer in bigquery: missing or partial cost for demand gen campaign

[giobreccia]

Hi everyone,

I'm encountering discrepancies in cost data for Demand Gen campaigns between BigQuery and the Google Ads platform. When visualizing costs in Looker Studio using BigQuery data (with simple custom query), the figures don’t match the values displayed directly in Google Ads.

Additionally, when exploring the tables in BigQuery, I noticed that for certain days or longer periods, the cost data for Demand Gen campaigns is completely missing—though clicks and impressions are present. This leads me to believe that the data might be corrupted.

I’ve checked the transfer configuration logs, and it shows that all data transfers to BigQuery were successful.

Has anyone else experienced this issue or know of any potential causes or solutions?

Thanks

[tal_azriel · Accepted solution]

we had the same problem. we did a scheduled backfill since the beginning of the discrepancies (around July) few days ago and it was fixed. still have days with minor gaps of ~$2

[giobreccia]

Thank you,

Do you notice if discrepancy is only for the "demand gen" campaign?

[tal_azriel]

yes it was only for demand gen campaigns. its a known issue

[Snoshone07]

Hi @giobreccia Looks like you encountered a frustrating data consistency issue between BigQuery and Google Ads for your Demand Gen campaigns. Random things you might do to try and debug, then fix this mismatch:

Check Data Schema: Check that your schema in BigQuery matches the latest Google Ads API version because it might be sometimes not updated If google adds new changes to Google Ads.

Ensure There Are No Data Filters: Certain types of campaigns or cost metrics may be omitted or the data treated differently in BigQuery, so always check for them. Re-examine your query so that Demand Gen costs are not filtered by an accident.

Check Time Zones and Conversion Windows: There are occasions where time zone εach data create matching issues, mainly due to the variation of attribution settings εboth nature or εeven conversion windows within Google Ads and BigQuery.

Evaluate if using an alternative data integration tool makes sense: If this continues and you can’t find a fix, it may be worth investigating other data connectors or ETL (extract, transform and load) tools that support Google Ads more robustly and manage inconsistencies better. I’ve heard that Windsor. For example, ai allows to use its more reliable Google Ads connectors, which often ensure greater control over data extraction and transformation, preventing problems like this one.

Hope this helps!

[giobreccia]

Hi @Snoshone07   I'm using the official "data transfer" tool for Google Ads inside BigQuery, which automatically inserts data into the dataset.

When I query these datasets, the costs for search, display, video, and performance campaigns match perfectly with the Google Ads UI. The only cost that appears incorrect is for demand gen campaigns.

It seems strange that I would need a third-party tool to fetch accurate data in BigQuery when there is an official tool available.

I will try with Google Ads Api call.

Thanks for you response!