Inconsistent usage reporting between Report Usage Metrics and connecting to their Semantic Model

[SevsBo]

I'm trying to streamline Report Usage Metrics as we analyse them far deeper than the regular Report Usage Metrics report allows. We look at usage by report by user per time period, and the opposite, which pages on the report each user visited in time, etc.

If we extract brand new data straight from the Report Usage Metrics, we see 509 views of the entire report, and we can extract each page manually, which takes a lot of time.

If we connect to the Report Usage semantic model, and filter the results on the correct report, we suddenly see 616 views of the same report.

Focusing on a single day, for a single page of that report, here is the difference in reporting:

Data extracted from the charts on the standard Report Usage Metrics

Total 3 views.

Data as showing through the semantic model connection

Total 5 views, for the same user as above.

Across all pages and full timeline on the report this gives us a discrepancy of 107 views, despite the data coming from the same source.

What is the reason for this discrepancy and which data should we believe? Is there a hidden filter on the Report Usage Metrics that we should be using on the data from the Semantic Model for the same data?

[GilbertQ · Accepted solution]

Hi @SevsBo 

I would always trust connecting directly to the semantic model to get an accurate reporting number, because quite often, as you suggested, there could be some filters on the existing reports. That way it should be consistent for you to get the same number.

[SevsBo]

Thank you!