High-Priority Resolution Required for GA4 and Integration Issues

[Sriram Suvaidhar]

Dear Google Cloud Community,

I hope this email finds you well. We are encountering critical issues with our GA4 implementation and its integrations that are significantly impacting our operations. Below are the details of the issues we are facing:
1. GA4 to BigQuery Linking Issue

• Upon linking GA4 to BigQuery, historical data  events_ 20*   from GA4 do not populate in BigQuery. Only the events_intraday_20* table is created, and data starts flowing from the day of the link creation. Historical data prior to the link creation remains missing.
• Additionally, it takes approximately 10 hours for the events_intraday_20* table to appear in BigQuery after the initial linking, further delaying data availability.

2. Audience Segment User Counts in GA4

• User counts for newly created audience segments are taking more than 24 hours to start populating.
• Once the data begins to populate, it remains inconsistent and does not align with the expected values or trends, causing concerns regarding data accuracy.
• 
  

3. GA4 Audience Segments Integration with Google Ads

• After linking GA4 with Google Ads, it takes up to 4 working days for audience segments to populate in Google Ads.
• Even after this delay, user counts within these segments are inaccurate and do not match the expected values, undermining confidence in the data.
• This is causing significant delays and inefficiencies in launching time-sensitive marketing campaigns.

Given the high priority and critical nature of these issues, we kindly request your assistance in resolving them within 2 business days.

Please escalate this matter to the relevant team, and let us know if additional details or clarifications are needed from our side. We would also appreciate timely updates on the progress and expected resolution timeline.
Thank you for your prompt attention to this matter.

[Audrey Stone]

Hi Sriram, sharing notes below in blue. This is more of a GA4 + Ads request vs. BQ in my opinion. Hope this helps!

 

Audrey Stone Data Analytics & AI | Enterprise Northeast audre...@google.com 650-704-0799

On Tue, Jan 28, 2025 at 7:55 AM Sriram Suvaidhar <sri...@squareshift.co> wrote:

Dear Google Cloud Community,

I hope this email finds you well. We are encountering critical issues with our GA4 implementation and its integrations that are significantly impacting our operations. Below are the details of the issues we are facing:
1. GA4 to BigQuery Linking Issue

• Upon linking GA4 to BigQuery, historical data  events_ 20*   from GA4 do not populate in BigQuery. Only the events_intraday_20* table is created, and data starts flowing from the day of the link creation. Historical data prior to the link creation remains missing. No, you cannot directly backfill data in GA4 using BigQuery; the native BigQuery export feature in GA4 does not support backfilling, meaning you cannot retroactively add historical data once the connection is established - you can only collect data moving forward from the date you set up the export. Use the GA API for historical data: To access past GA4 data, you need to use the Google Analytics API to retrieve historical data and then manually import it into BigQuery - Napkyn support article.

 

• Additionally, it takes approximately 10 hours for the events_intraday_20* table to appear in BigQuery after the initial linking, further delaying data availability.

2. Audience Segment User Counts in GA4 There could be a propogation delay, 24H delay is normal and also dependent on the quality on the inputs. 

• User counts for newly created audience segments are taking more than 24 hours to start populating.
• Once the data begins to populate, it remains inconsistent and does not align with the expected values or trends, causing concerns regarding data accuracy.
• 
  

3. GA4 Audience Segments Integration with Google Ads I would open a case with Google Ads if the quality and volume does not improve. The larger the sample set (re: all traffic source) and improvement of quality inputs and 1P identifiers, the better the output. [GA4] Why your audiences may not be populating in Google Ads

• After linking GA4 with Google Ads, it takes up to 4 working days for audience segments to populate in Google Ads.
• Even after this delay, user counts within these segments are inaccurate and do not match the expected values, undermining confidence in the data.
• This is causing significant delays and inefficiencies in launching time-sensitive marketing campaigns.

Given the high priority and critical nature of these issues, we kindly request your assistance in resolving them within 2 business days.

Please escalate this matter to the relevant team, and let us know if additional details or clarifications are needed from our side. We would also appreciate timely updates on the progress and expected resolution timeline.
Thank you for your prompt attention to this matter.

--
You received this message because you are subscribed to the Google Groups "Instant BQML and Vertex Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to instant-bqml-verte...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/instant-bqml-vertex-users/CAL3Edm5QDQDqVemaqTxpZVisatZxubJ%3DZXt1Ca8VjB0Zjj27bg%40mail.gmail.com.

[Shwetha Muthukumar]

Hi Audrey,

Sriram and I work together, and I sincerely appreciate your insights—they were incredibly helpful.

If possible, could you kindly provide a point of contact at Google so I can raise this issue directly with them? It would be greatly appreciated.

Best regards,
Shwetha M 

[Audrey Stone]

I'd start here with the Analytics help: https://support.google.com/analytics/gethelp?sjid=13919341637386544974-NA

They will request specific views, times of data loss, etc to troubleshoot the issue. Moving the BQML listserv to BCC as this is not related.

 

Audrey Stone Data Analytics & AI | Enterprise Northeast audre...@google.com 650-704-0799