Aggregating Unconsented Events Could Be Your Biggest GA4 BigQuery Mistake.
If you are using Google Advanced Consent Mode, your GA4 BigQuery export data should be in the following two major categories:
1) Consented events.2) Unconsented events.
Failing to do so can lead to significant errors in your data analysis and reporting.
Consented events:
(privacy_info.analytics_storage = ‘Yes’ OR privacy_info.ads_storage = ‘Yes’)
Unconsented events:
privacy_info.analytics_storage IS null OR privacy_info.analytics_storage = ‘No’ANDprivacy_info.ads_storage IS null OR privacy_info.ads_storage = ‘No’
When querying GA4 data in BigQuery, it’s important to consider the consent status.
Unconsented events include cookieless pings (when advanced consent mode is used), which make the data highly unreliable.
These “cookieless pings” from unconsented users contain limited, anonymized event data, which includes events without personal identifiers (such as cookies, device IDs, user IDs, client IDs, advertising IDs, etc).
So,
Unconsented events should be analyzed separately at the event level and should not be aggregated to the session or user level.
The following are the query considerations when dealing with consented and unconsented data:
>> Use separate SQL queries for consented and unconsented events.
>> Avoid using unconsented events for any session-based calculations, as these will be highly inaccurate.
>> When calculating user-based metrics (e.g., user count, user retention), exclude unconsented events entirely.
The following are the reporting considerations when dealing with consented and unconsented data:
>> Always clearly separate metrics derived from consented and unconsented events in reports.
>> For unconsented data, focus on reporting event-based insights rather than user or session-based metrics.
>> Clearly communicate the limitations of unconsented event data in your reports to prevent misinterpretation.
