recently, I’m getting more and more questions from customers why the same data in Looker Studio does not match when using a scorecard and a table!
The issue in detail: I created a simple report in Looker Studio showing GoogleAnalytics4 (GA4) sessions over time (current month). I added a scorecard, table, and bar chart to the report page. The bar chart data is wrong and underreports compared to the scorecard, the data table, and Google Analytics 4 itself.
Debugging:
There are no report filters in place, no custom data, no formulas, no blends, nothing. Just a simple date range and reference to actual sessions. I’m connecting directly to GA4.
I tried setting the bar chart to display as daily, weekly, monthly, etc and it always comes up short.
When I change the bar chart ‘chart type’ from bar chart, to scorecard, the correct number displays, and when I change back to a ‘bar chart’ it goes back to the wrong number.
I see the same issue in other reports I’ve created for other websites in separate dashboards with entirely separate data sources.
I also see a similar issue between scorecard/chart data discrepancies when creating a calculated field to measure conversions by total users. In that case, the bar chart overreports vs the scorecard.
Is this common? A bug? Any way to resolve it?
It is very specific problems with problem different solutions. Some details and screenshots about how the “setup” tab is configured can help to understand.
One of the common mistake is the presence of a wrong field for the sorting,
provoking side effect on the aggregation, so the value of the displayed metric.
For each scorecard there is a time series chart with the same metric. Everything is connected to the same GA4 property. Only for the new users metric the scorecard an time series chart match. For the other two metrics only the scorecards match with th GA4 property.
For your use case, the problem is related to data aggregation. Keeping in mind that new users is a session-scoped metric (state changing regarding the session) whereas active users and total users are user-scope metrics.
When your scorecard is displaying the right number of active users for a specific month, your series is summing the daily number of active users for the same month. Using the dimension “Date” and displaying it as a month format (data type) will not deduplicate the daily number of users to monthly users.
With GA4, you should solve this problem by replacing “Date” with the built-in dimension “Year Month” available in the GA4 connector. The connector will send one monthly row with the correct aggregation, and not sending 30 daily rows summed by Looker Studio.
I am having a similar issue, where the time series chart is reflecting the correct data. But in the scorecard it is reflecting “No Data”. It’s only when it’s a scorecard, if I change it to a table or other chart time the correct data is represented. What could be causing this?
