Once Again, Gartner Recommends Using a Data Warehouse, Just Not in so Many Words
Once upon a time the data warehouse reigned supreme. It wasn’t long ago, in fact, that any company making a serious attempt to enable reporting knew the first step was to develop a data warehouse. Many got it wrong, of course, but those who got it right found reporting nirvana: business users could mine the data and make fact-based decisions in real time with little or no effort.
Over the past decade, however, a seismic shift has occurred. The various companies selling business intelligence (now called data visualization) software, and the majority of industry analysts, stopped talking about the need for a data warehouse. They led us to believe that with the right software in hand, business people could do their own data preparation and visualization, no IT support required. Before long, the idea of building a centralized, curated data warehouse seemed antiquated, expensive and unnecessary. After all, data warehouse projects are difficult and have a notoriously high failure rate, so the appeal of a tool that could do data preparation along with visualization was intoxicating, and executives were easily sold on this new vision.
Fast forward to Gartner’s 2019 research paper, Critical Capabilities for Analytics and Business Intelligence Platforms, and suddenly the winds have changed direction again. Analysts James Richardson, Rita Sallam and Austin Kronz now emphasize that “Organizations that offer users access to a curated catalog of internal and external data will realize twice the business value from analytics than those that do not.”
If it isn’t immediately clear, Gartner is recommending that companies build a data warehouse as a necessary prerequisite in order to enable business users to reap the potential benefits of self-service data visualization.
This is an important shift in thinking, and a recognition that while modern data visualization tools like those found in Gartner’s “Leaders” magic quadrant have a lot to offer, without curated data they are insufficient to fulfill the promise of self-service reporting. In other words, the key to implementing a successful enterprise analytics function is developing a data warehouse that makes the data available, accurate and well organized before anyone attempts to visualize it with reporting software.
To companies that rely on a data warehouse to enable self-service BI and reporting this comes as no surprise. But, given Gartner’s level of influence on how companies approach analytics, this is a critically important admission that one hopes will be heard far and wide by companies still struggling to achieve success with their analytics function.