When there is a Power BI report, and different users see different results, it could be due to a number of factors, including row-level security (RLS), caches, different software versions, as well as differences between Power BI desktop and service. Here is an outline of all these issues and the appropriate measures to take to eliminate the inconsistencies in results:
Row Level Security (RLS): it is typical for users looking at the same report to come across different figures owing to the row-level security that is implemented in the report. This means that if RLS is controlled at the dataset level in Power BI, data access is then controlled and customized for each role, and every user only gets to see the data that they are allowed to see. You may visit the dataset settings and check the applied roles to ascertain if RLS is present. The RLS should be reviewed or turned off if it’s possible when the report does not require it so that common accessibility to data can be guaranteed.
- Saves of reports are performed to enhance data analysis speeds in the Power BI application, but that kind of saving may, at times, cause some discrepancies. For instance, if the users are served with the old content that was not recent results they may rely on, users may have different results. This can be done by either forcing the system to refresh the data and clear the cache or changing the report settings so that data is refreshed after a short time. For example, in Power BI Service, scheduled refreshes can be implemented to ensure users are seeing updated data and not any stale information.
Version Updates: If some users work with an outdated version of Power BI Desktop or utilize the report within the Power BI Service, then some inconsistencies may arise regarding the updates of features and the way data is processed. For instance, Power BI Desktop and Power BI Service may utilize different formats of parsing dates, which depends on the regional settings, as in U.S. Format and European format on dates. This may be solved by ensuring that all the users are using the most recent version of the Power BI Desktop application and that the regional settings are the same. You may also want to look at how the report performs in Power BI Desktop and Power BI Service to check for differences. This means that you need to compare all the versions of the report with the actual one within different applications.
- Power BI Desktop and Power BI Service are different applications, and therefore, delivering the same report through either of them may not produce the same output due to the differences in the use of functions, datatypes, and even visual presentation. These may, in turn, create some small variances in the report where users cross-access the reports on different platforms. To avoid such situations, it is advisable that you focus on making and presenting your reports while considering the usage of both positions, such as Desktop and Service, and put them to the test. It is recommended to release Power BI Service to the authorized users for their consumption as it is usually the case that for any organization, most reports are mostly viewable in that environment.
To conclude, first, look into RLS and caching, make sure all software versions are up to date, and run your report in both Power BI Desktop and Power BI Service. Taking these measures will increase the chances of diagnosing and fixing the factors that contribute to differences in report data per user.