Power BI liberates users… But how can you maintain control of your platform over time?

 

Power BI liberates users…   But how can you maintain control of your platform over time?

Power BI is setting a precedent in the world of dataviz: practically all major companies in the world have deployed it as a corporate solution or,  at the very least , as a backup solution (for certain business lines)  – and it's understandable: attractive licensing, ergonomics, native integration with the Microsoft ecosystem (Excel, Azure, SQL Server, Microsoft 365) already present almost everywhere.   This accessibility has a downside: as the platform opens up to users, dashboards multiply, sources diversify, and complexity sets in.  Platform governance is already a challenge—or will become one. This is a shame when we know that adopting a new data visualization tool is an opportunity to rebuild on sustainably sound foundations, not to create debt at high speed. It is with this in mind that we have considered a tool-based approach, designed for data teams, but also for businesses. It allows you to maintain a clear vision of data quality and to manage the platform sustainably – without slowing down adoption   !

Control your Power BI platform - Axis #1: Continuously inventory and qualify your Power BI (and other) dashboard assets.

When a company uses thousands of dashboards in 2, 3 or 4 different technologies, it seems essential to have an overview...   The "Report Browser" that we developed within {openAudit}  allows you to view the entire Power BI portfolio  : dashboards, folders, last use/update, modification, dashboard ID, CUID, dashboard weight, number of views, response time, etc.  In the case of multiple technologies – which is almost always – all technologies are presented on the same interface.

Control your Power BI platform -  Axis #2:  An impact analysis in Power BI to develop the platform securely

Edit a column in a table? Delete an obsolete dimension? Before touching any element of the platform, {openAudit} 's impact analysis  indicates where and how this element is used throughout the Power BI environment (and beyond in the case of multiple tools). Result: fewer errors, fewer regressions, etc.

Control your Power BI platform -  Axis #3:  Replication analysis in Power BI to rule out clones

Some dashboards, although very recent, can already be widely replicated.  To avoid unnecessary duplication,  {openAudit}  compares each dashboard with all the others along five key axes. To do this, we'll rely on the {openAudit}   database, which stores the dashboard's intelligence and structure. The replication percentage reveals similarities (in green) and dissimilarities (in red), which can be analyzed in detail ("drill through").   The criteria analyzed: "Similarity"  : percentage of proximity of the contents (loaded data) between the master dashboard and its clones. "Formulas"  : comparison of formulas and variables, i.e. intelligence. “Structure”  : analysis of containers, i.e. graphic components. "Filter"  : measure of the replication of the applied filters. "Deep filter"  : evaluation of the replication of the filtered data itself.

Control your Power BI platform -  Axis #4:  Data lineage in Power BI, to have a clear vision of flows and dependencies.

Through  {openAudit} , we offer a highly intelligible data lineage, because it is modular and connected to sources. Each data item in each Power BI dashboard can thus be automatically traced back to its operational sources. Through this data lineage, we will be able to identify the sources used, the datasets involved, the expressions implemented, etc.     This data lineage can now be activated directly from the Power BI Online dashboard with our Chrome / Edge plugin!    Use cases: Understanding where the information presented comes from , especially when an indicator appears to be poorly calculated (or deserves to be calculated differently). Identify dependencies  when a system element changes. Document, if necessary, certain regulatory data, etc.

Bonus: some tools offered to make life easier for teams

To go further, we have developed various tools based on customer specifications, designed to enable data teams to move faster. Up-to-date backups:  we create Git versions of dashboard templates automatically.  Automated PBIT ⇄ PBIX conversion:  to facilitate workflows between local development and Cloud. Deployment to Power BI Cloud:  to automate and make production deployment more reliable. Mass archiving:  replicated or obsolete dashboards can be archived en masse with our tools, to keep the platform as simple as possible for the long term.  Furthermore, we  centralize management rules in databases,  which allows them to be factorized or even harmonized.

In summary, Power BI opens up immense possibilities, provided you keep control! Thanks to tools designed for data teams as well as for businesses, it is possible to reconcile agility and governance, innovation and stability.