If your metrics are showing as zero in the "results" tab, but you can see the metric table, the most common causes are: 1. Identifier Mismatch: GrowthBook requires a user identifier (like

user_pseudo_id

or

anonymous_id

) to join experiment exposures to metric events. If you removed

AND user_pseudo_id is not null

because you don't have that field, you must ensure that the identifier you use in your experiment assignment query matches the identifier used in your metric queries. If you use a different column (e.g.,

user_id

), update both your experiment assignment and metric queries to use the same identifier consistently. Otherwise, GrowthBook cannot join the data, resulting in zeros for all metrics [[source](https://linen.growthbook.io/t/27125656/i-am-not-seeing-my-metric-in-my-experiment-the-values-stay-z)]. 2. Variation ID Type Mismatch: In GA4/BigQuery, sometimes

variation_id

is stored as a string, not an integer. If your data uses

variation_id_param.value.string_value

, but your query uses

.int_value

, the join will fail and metrics will be zero. Try changing:

sql    variation_id_param.value.int_value AS variation_id

to

sql    variation_id_param.value.string_value AS variation_id

and test if this returns data [[source](https://linen.growthbook.io/t/30129167/u07fj89lffb-i-can-see-experiment-data-in-bigquery-but-in-gro)]. 3. Missing or Incorrect Identifier: If you don't have

user_pseudo_id

, you must use another unique identifier (like

user_id

or a custom field) and ensure it's present in both your experiment and metric queries. If you remove the identifier filter entirely, GrowthBook may not be able to attribute metrics to users correctly [[source](https://linen.growthbook.io/t/15993334/hey-everyone-we-began-using-gb-two-weeks-ago-but-we-ve-encou)]. What to do: - Replace all uses of

user_pseudo_id

with your actual identifier column (e.g.,

user_id

) in both the experiment assignment and metric queries. - Make sure the

variation_id

type matches your data (try

string_value

if

int_value

returns no data). - Do not remove the identifier filter entirely; instead, update it to match your identifier column (e.g.,

AND user_id IS NOT NULL

). Example adjusted query (if using

user_id

and string variation_id):

sql SELECT   user_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   ... WHERE   ...   AND user_id IS NOT NULL

If you still see zeros, double-check that your metric query uses the same identifier and that your metric events actually exist in the data for the experiment period [[source](https://linen.growthbook.io/t/27125656/i-am-not-seeing-my-metric-in-my-experiment-the-values-stay-z)]. If you need more help, please share the identifier column you are using and a sample of your metric query.

To find the user identifier in your GA4 database in BigQuery, you should look for the column named

user_pseudo_id

. This is the default unique identifier that GA4 assigns to each user and is present in every event record in the GA4 export to BigQuery. If your implementation also tracks logged-in users, you might see a

user_id

column as well, but this is only populated if you explicitly set it in your tracking setup. You can confirm this by running a simple query in your BigQuery GA4 dataset, such as:

sql SELECT   user_pseudo_id,   user_id,   event_name FROM   `your_project.your_dataset.events_*` LIMIT 10

-

user_pseudo_id

is always present and is the main identifier for anonymous users. -

user_id

is only present if you set it in your GA4 implementation (typically for logged-in users). If you do not see

user_id

populated, you should use

user_pseudo_id

as your user identifier for experiments and metrics in GrowthBook [[source](https://linen.growthbook.io/t/16592082/hey-team-we-have-an-issue-with-receiving-data-from-a-gb-expe)]. If you have custom identifiers, you may need to check your event parameters or custom dimensions, but for standard GA4 setups,

user_pseudo_id

is the primary user identifier [[source](https://linen.growthbook.io/t/23126975/we-ve-implemented-growthbook-through-react-and-use-ga4-as-ou)]. If you are unsure which identifier to use, you can review your BigQuery table schema or run the above query to inspect the available columns.