Here are the most common causes and quick checks when GA4/BigQuery shows data, but GrowthBook still says “No data yet”: - Confirm experiment exposures are reaching GA4 the way Script Tag SDK expects. - Script Tag SDK auto-sends experiment_viewed, so you typically don’t need a custom trackingCallback. If you added one, ensure it actually sends to GA4 (via gtag or dataLayer) and includes experiment_id and variation_id; otherwise, GrowthBook won’t be able to join metrics to exposures ([GA4 troubleshooting – trackingCallback](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#the-trackingcallback-doesnt-fire-or-is-misconfigured)). - Check for the GA4→BigQuery sync delay. - Free GA4 exports to BigQuery are daily; expect ~24h lag (and 24–48h initial delay when first linking GA4 to BQ). Seeing data in BQ means the export is working, but results in GrowthBook depend on the assignment queries returning that data for the experiment window ([BQ troubleshooting](https://docs.growthbook.io/kb/google-analytics/google-big-query-bq-troubleshooting)). - Test your Experiment Assignment Query (EAQ) in GrowthBook. - Go to Metrics & Data → Data Sources → your BQ source → Experiment Assignment Queries → Edit → Test Query. If no rows, try removing WHERE clauses for experiment_id/variation_id to see if anything returns. If still empty, the SQL likely doesn’t match your GA4 BQ schema (dataset/table names, field paths) and needs adjustment ([GA4 common issues – EAQ issues and testing](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)). - Verify no filters are excluding users. - In the experiment’s Analysis Settings, temporarily remove any Activation Metric or Segment; these can filter out traffic and produce “No data yet” even when exposures exist ([Troubleshooting experiments – lack of traffic](https://docs.growthbook.io/kb/experiments/troubleshooting-experiments#problem-1-no-or-little-traffic-flowing-into-the-experiment)). If the EAQ test returns data and GA4/BQ contain experiment_viewed with experiment_id and variation_id, GrowthBook should populate results after the next data refresh. If the EAQ test is empty, focus on fixing the assignment query to match your BQ export schema (dataset names and fields) and confirm the exposure events include the required properties ([GA4 common issues](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)).

A few likely fixes based on GA4/BigQuery + GrowthBook EAQ issues: 1) variation_id type mismatch Your query uses variation_id_param.value.int_value, but many GA4 setups record variation_id as a string. Change to string_value and test again: - variation_id_param.value.string_value AS variation_id (If this returns rows, keep string_value.) ([Forum examples](https://linen.growthbook.io/t/29091994/we-are-unable-to-get-any-result-for-our-experiments-we-ve-wr#29091994), [GA4 EAQ troubleshooting](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)) 2) Verify the GA4 dataset/table names and intraday suffixes Ensure the FROM points to the correct project.dataset and analytics property, and include intraday tables if you expect same-day data. Your query already includes events_* plus intraday_… range, which is correct, but double-check the dataset IDs match your actual BigQuery export location. ([GA4/BQ troubleshooting](https://docs.growthbook.io/kb/google-analytics/google-big-query-bq-troubleshooting)) 3) Narrow down by relaxing filters to confirm data presence - Temporarily remove: - experiment_id_param.key = 'experiment_id' - variation_id_param.key = 'variation_id' If rows appear, re-add them one at a time to find the mismatch (often the variation_id type). ([GA4 EAQ testing guidance](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)) 4) Confirm experiment_viewed events are actually in GA4/BQ If the test still returns no rows, ensure experiment_viewed exists in GA4 and flows to BigQuery (24h lag on standard GA4; up to 48h after first linking). ([GA4/BQ delays](https://docs.growthbook.io/kb/google-analytics/google-big-query-bq-troubleshooting#why-is-recent-experiment-data-not-yet-available-in-bigquery)) Suggested adjusted query (only change is variation_id to string_value):

sql SELECT   user_pseudo_id as anonymous_id,   TIMESTAMP_MICROS(event_timestamp) as timestamp,   experiment_id_param.value.string_value AS experiment_id,   variation_id_param.value.string_value AS variation_id,   geo.country as country,   traffic_source.source as source,   traffic_source.medium as medium,   device.category as device,   device.web_info.browser as browser,   device.operating_system as os FROM `learning-heroes-bq`.`analytics_372696867`.`events_*`, UNNEST(event_params) AS experiment_id_param, UNNEST(event_params) AS variation_id_param WHERE   ((_TABLE_SUFFIX BETWEEN '{{date startDateISO "yyyyMMdd"}}' AND '{{date endDateISO "yyyyMMdd"}}') OR    (_TABLE_SUFFIX BETWEEN 'intraday_{{date startDateISO "yyyyMMdd"}}' AND 'intraday_{{date endDateISO "yyyyMMdd"}}'))   AND event_name = 'experiment_viewed'   AND experiment_id_param.key = 'experiment_id'   AND variation_id_param.key = 'variation_id'   AND user_pseudo_id is not null

If this still returns no rows, test by removing the two key filters to verify events exist, then re-check dataset/table names and the GA4→BQ export timing. ([GA4 common issues](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook))

Thanks for checking. Here are the next targeted checks to resolve “No data”: 1) Switch variation_id to string_value and retest Many GA4 exports store variation_id as a string. Update and test the EAQ in GrowthBook: variation_id_param.value.string_value AS variation_id. [[GA4 common issues](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook); [Forum thread](https://linen.growthbook.io/t/29091994/we-are-unable-to-get-any-result-for-our-experiments-we-ve-wr)] 2) Relax filters to confirm events exist In the same EAQ test modal, temporarily remove: - AND experiment_id_param.key = 'experiment_id' - AND variation_id_param.key = 'variation_id' If rows appear, re-add them one by one to find the mismatch (often variation_id type). [[GA4 EAQ testing](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)] 3) Verify GA4→BQ export location and timing Double-check the project.dataset (analytics_XXXXXX) you’re querying matches where GA4 is exporting now, especially if you recently changed the linked dataset. Remember the free GA4 export has ~24h lag (and 24–48h after first linking). [[BQ troubleshooting](https://docs.growthbook.io/kb/google-analytics/google-big-query-bq-troubleshooting)] 4) Confirm exposures actually exist in BQ Run a minimal query directly in BigQuery to see any experiment_viewed rows (no param filters). If none, the issue is upstream (tracking or export). [[GA4 EAQ testing](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook)] Example EAQ adjustment (only change is string_value):

sql SELECT   user_pseudo_id as anonymous_id,   TIMESTAMP_MICROS(event_timestamp) as timestamp,   experiment_id_param.value.string_value AS experiment_id,   variation_id_param.value.string_value AS variation_id FROM `learning-heroes-bq`.`analytics_372696867`.`events_*`, UNNEST(event_params) AS experiment_id_param, UNNEST(event_params) AS variation_id_param WHERE   ((_TABLE_SUFFIX BETWEEN '{{date startDateISO "yyyyMMdd"}}' AND '{{date endDateISO "yyyyMMdd"}}') OR    (_TABLE_SUFFIX BETWEEN 'intraday_{{date startDateISO "yyyyMMdd"}}' AND 'intraday_{{date endDateISO "yyyyMMdd"}}'))   AND event_name = 'experiment_viewed'   AND experiment_id_param.key = 'experiment_id'   AND variation_id_param.key = 'variation_id'   AND user_pseudo_id is not null

If this still returns no rows, it almost always points to either a variation_id type mismatch or querying the wrong dataset/tables. Once the EAQ test returns rows, GrowthBook results should populate after refresh. [[GA4 common issues](https://docs.growthbook.io/kb/google-analytics/google-analytics-ga4-troubleshooting#common-issues-when-integrating-ga4-with-growthbook); [BQ timing](https://docs.growthbook.io/kb/google-analytics/google-big-query-bq-troubleshooting)]