Question from Reddit user:
Hey question on Explore Reports, doing some KPIs for a subdomain, specific to 3 landing pages.
see if there is a better way to organize and also, I appear to be missing a couple sessions on two of the landing pages (these are internal website with low visibility so I can’t be missing sessions).
Row= landing page + query string
Values = about a dozen around engagement, page views and session count
Filter= landing page + query exactly match [url for about page]
I’ve played around with “contains”, “ends with” etc but still not seeing 2 sessions that I see in the traditional reports.
I copied the tab from the the above that I did for the homepage and that seems to work fine…
Answer from Nabil:
The short answer is:
The discrepancy you are seeing between your GA4 Explore report and your standard “Pages and screens” report is almost certainly due to GA4’s data thresholds or data sampling being applied in the Explore interface, especially given the low-visibility, internal traffic and the specific dimension you are using.
Explore reports, particularly when filtering on a high-cardinality dimension like ‘landing page + query string’, are more aggressive with applying privacy-driven thresholds than standard reports, which can lead to suppression of low count metrics like those two missing sessions.
To ensure complete session count accuracy without data suppression or sampling, the best solution is to use the Google Analytics Data API to extract the raw, unsampled data directly and then visualize it in Looker Studio.
The long answer is:
That is a frustrating issue, but what you are experiencing is a common headache when moving from the reliable standard reports to the more powerful, but more restrictive, Explore interface in GA4.
The key difference between the two reporting types is that standard reports are pre-aggregated and not subject to the same privacy thresholds and sampling limits as Explore reports.
The moment you use a high-cardinality dimension, like ‘landing page + query string’ or ‘page path + query string’ – which generates a large number of unique values – or apply a complex filter, GA4 starts aggressively applying privacy thresholds to the data in your Explore report.
This means that if a session count is very low (like the two sessions you are missing), GA4 may suppress that data point to prevent potential user identification.
Furthermore, since you copied the tab from the homepage, the issue may be related to the specific internal page path.
The homepage often has a much higher session volume, which makes it less susceptible to these low count thresholds than a low-traffic internal subdomain page.
A reliable fix is to abandon the Explore report for your KPI tracking and instead use the Google Analytics Data API to pull your data directly.
The Data API allows you to pull all your session and engagement data at the date and dimension granularity you need, and critically, it allows for the use of API tokens that can often access data without the heavy sampling or privacy-driven thresholds that restrict the Explore reports.
Once you extract this complete data set, you can push it into Looker Studio via its own connector or the Looker Studio API for visualization, where you will build a custom chart that mirrors the “Pages and screens” report, ensuring 100% data fidelity.
For a forward-looking solution, you could also use Google Tag Manager and a server-side environment like Stape or Google Cloud Platform to capture the ‘page path + query string’ not just as a standard parameter, but as a custom, dedicated event property.
This gives you more control over the data structure before it hits GA4, which can further reduce the chances of encountering cardinality-related sampling in the future.
About The Author
