We recently noticed that some of our drilldown pages receive a very low number of views. To improve adoption and usability, we would like to experiment with different report designs and track how those changes impact user behavior. However, Power BI currently has no built-in capability for A/B testing report experiences.
If we want to test different layouts, navigation approaches, or drilldown page designs, there is no native way to split users between variants and measure engagement metrics such as page views or interaction rates. In an era where dashboards are increasingly treated as tech products, the ability to iterate based on user behavior is critical.
Product teams routinely rely on A/B testing to validate design decisions, yet BI developers lack similar experimentation tools inside Power BI.
A native A/B testing feature could enable scenarios such as:
- Splitting report users between two or more design variants
- Measuring page view rates, drilldown usage, and interactions
- Comparing engagement and navigation patterns across versions
- Making data-driven decisions about report UX improvements
This capability would help teams move from static dashboard development to continuous improvement of analytics products, ensuring reports are actually used and provide value.
on:
Consistent TMDL serialisation across developer mac...
on:
Allow Users to Configure the Refresh Frequency of ...