Fabric Killed the Power BI Star

It all started nearly 5 years ago when Salesforce bought Tableau in August of 2019. That was the seed that sowed the roots of Power BI's demise. Well, sure, "demise" might be a strong word but it's more dramatic this way. Read on and I'll explain what I'm talking about.

When Salesforce purchased Tableau, there was a considerable amount of concern among the Tableau faithful that the acquisition was the harbinger of impending doom for the product. A common fear was that Salesforce would gut Tableau and focus more on integration with the closed ecosystem of Salesforce versus new features and functionality for standalone Tableau uses. Unfortunately, for Tableau users, these fears have proven true, so much so that an Irish wake was held last year for Tableau.

At the same time, there was an ample amount of congratulatory snickering done by Power BI advocates. Finally, Power BI would soon be the new dominate, stand-alone data analytics visualization tool. It was only a matter of time. And this also proved true...for a time. However, a small number of astute industry analysts surmised that Tableau's exit from the stage was an extremely bad thing for Power BI. The theory went that without a substantive rival in the market, Microsoft would lose focus on Power BI and not be motivated to continue its rapid evolution as a product.

Unfortunately, these final prophetic fears have also been realized with the introduction of Microsoft Fabric. Microsoft Fabric has done to Power BI what Salesforce has done to Tableau. The architect of Power BI's destruction is none other than its creator. If you doubt this even a little, consider the following.

In the last year, the number of preview features in Power BI has ballooned from a relatively stable number of about one dozen to over double that amount, nearly 30. Before Fabric, features typically spent a relatively short 4 or 5 months in preview status before becoming generally available (GA). Now features essentially never make it out of preview, they are stuck there for a year or more and counting. The ballooning of preview features in Power BI is a clear indication of a lack of focus and commitment to truly improving the product.

Coinciding with this preview feature glut, for more than a year Power BI Desktop's monthly releases have been lackluster and some incredibly buggy. Little actually new functionality has been introduced with much of the innovation occurring only from third parties adding new custom visuals and custom data connectors to the product.

Consider that both the March 2024 and April 2024 versions of Power BI Desktop introduced a paltry 3 non-preview enhancements in terms of reporting capabilities and no non-preview analytics capabilities. For comparison with pre-Tableau purchase releases, the March 2019 version of Power BI Desktop included 10 new reporting features and 2 new analytics capabilities while the April 2019 release included 3 new reporting features and 4 new analytics features. It is obvious that much of the development horsepower once laser focused on Power BI has shifted elsewhere within Microsoft, most likely Fabric.

Speaking of the March 2024 Power BI Desktop release, the release was of such poor quality that it only lasted about a week and a half before being replaced with the April 2024 release. Poor quality software is, again, a telltale sign of an utter lack of focus and commitment to a product.

Moving on, it is clear that Microsoft feels so confident about its position in the market that it can change licensing terms to the substantial detriment of its Power BI Premium customers. The actual retirement of a Microsoft cloud SKU is unprecedented or nearly so but after July 2024, Power BI Premium SKU's will be unavailable. Only existing Premium customers and enterprise agreement (EA) customers are safe, but only until their contracts end. Then, they too will pay more for less Power BI functionality with increased costs for storage and the removal of Power BI Reporting Services license rights. Clearly, Microsoft does not value the loyalty of its early enterprise adopters, a substantive shift from pre-Tableau purchase days.

What is different about Salesforce killing Tableau versus Fabric killing Power BI is that the former was a straight-up purchase of a product for integration into a closed platform. The latter case, however, is more like trading on the goodwill of an entrenched user base in order to promote a product that ends up killing the original product that created the goodwill in the first place. It's all a little insidious and grimy and not something the original user base ever asked for or wanted. Fabric is not something that the vast, vast, vast majority of Power BI users even care about.

MTV famously played "Video Killed the Radio Star" as it's first music video on the channel, ushering in an age when video became the dominate platform for consuming music. While MTV may have stopped playing music videos, the trajectory of radio as a primary platform for consuming music has steadily declined over the intervening decades with most music lovers now using a streaming service or YouTube videos. Sadly, Power BI now finds itself in a similar state of dwindling focus and importance, upstaged by the shiny, glitzy new platform that is Microsoft Fabric. If only Fabric was even half as good or revolutionary as MTV.

In summary, Power BI is dead and the future is Fabric. Not because the people chose a superior medium/platform over another, but simply because Microsoft is hell bent on shoving Fabric down everyone's throats.