One of the world’s best-known technology companies buys a business intelligence (BI) company to help uncover insights from its database. One of the world’s largest application companies buys a BI company to help it surface insights from its application data. Both transactions happen within days of each other.
I’m sure you’re thinking of the recent acquisitions of Tableau by Salesforce and Looker by Google.
But not me.
I’m thinking IBM buys Cognos and SAP buys Business Objects…and it’s 2007.
Who says history doesn’t repeat itself?
Here’s more relevant history.
In 2011, Gartner wrote a seminal research report that placed BI adoption below 30 percent. They recommended BI vendors improve usability for self-service analysis centered around interactive visualizations, search, mobile, data preparation, and performance.
The BI vendors responded with all of these and more.
In 2019, after eight years of BI vendors adding more capabilities, Gartner placed BI adoption at 35 percent. That’s compound annual growth rate of 1.9 percent. Put differently, that is adding roughly 0.6 percentage points or 60 basis points per year. Sure, that’s better than most bank accounts, but, like your bank account, it’s pathetic.
Basically, BI adoption has had no appreciable increase in eight years despite more tools from more companies with more features. If we are to believe Gartner, the new saviors of adoption will be features driven by artificial intelligence (AI) like natural language processing and conversational analytics.
Do you believe it? I don’t.
The bet Google and Salesforce—and IBM and SAP before them—placed with these acquisitions is that the best way to drive adoption of analytics is with better tools and more features. It clearly didn’t work then and there’s no indication it will work now. Better tools certainly make the lives of business analysts better, but that misses the point. Business analysts already have the tools they need to their jobs.
What about everyone else?
BI tools came into being because people were not getting insights from the applications they were already using to do their jobs. Given the anemic adoption rate, standalone BI tools clearly are not serving the needs of the vast majority of people who are making day-to-day decisions that impact the operations of an organization.
The answer is not to keep eking out miniscule improvements over years with no appreciable gains. Or, as Albert Einstein is often credited with saying, “The definition of insanity is doing the same thing over and over again, but expecting different results.” (As a side note, I’m sure he would not want an analytics tool named after him only to be replaced by another analytics tool. See: definition of insanity.)
The reason these acquisitions miss the point is because their approach to scalable and pervasive analytics is fundamentally flawed.
The answer is to embed analytics into the applications people use every day to do their jobs. For the past six years, we have conducted research on the embedded analytics market. Product teams that embedded analytics into their applications see adoption rates around 60 percent—nearly double that of standalone analytics tools.
When people get information in context of their daily responsibilities, they can make better decisions and can take action all in one place. And they don’t waste time switching between applications, because everything is already in the products they already rely on.
Of course, product teams don’t really care as much about analytics adoption as they do about the tangible business results the analytics provide. With embedded analytics, 67 percent of product teams surveyed say users spend more time in the application; 92 percent said it helped differentiate their product from competitors; and 68 percent said they can charge more.
But don’t believe us. The McKinsey Global Institute wrote a compelling report in 2018 titled Breaking Away: The secrets to scaling analytics where they researched over a thousand companies on their analytics investments. McKinsey writes that the toughest challenge is conquering the last mile of BI, which is embedding analytics into decision-making processes so insights become outcomes. Without completing the last mile, analytics investments risk going to waste.
To be sure, Salesforce will charge you extra for Tableau, and Einstein, and pretty much anything else they can. But the analytics still won’t be in context of the CRM application we all use (Logi included). The end result will (continue to) be frustrating, expensive, and much less effective.
Perhaps Google will do better with Looker, but don’t count on it. They’re trying to take on Microsoft in the battle for your desktop and cloud (documents, spreadsheets, presentations, standalone analytics, data storage).
The options are insanity or embedded analytics. Choose wisely.