I recently had the opportunity to talk with R. “Ray” Wang, one of the enterprise technology industry’s foremost thinkers. Ray is founder and chairman of Constellation Research, and his new book, “Disrupting Digital Business,” examines why 52 percent of the Fortune 500 have merged, been acquired, gone bankrupt, or fell off the list since 2000. Most importantly, Wang explores how today’s companies can avoid the same fate.
One of the ways companies can maintain relevance is through the insight gained by analytics, so I wanted to get Wang’s take on what business leaders should be thinking about when it comes to the accelerating evolution of enterprise analytics. Here are the highlights of our discussion:
The number one priority for business leaders is to ask the right questions, which are often the big questions. Take the Voltaire quote, “You don’t judge people by the answers they give, you judge people by the questions they ask.” Business leaders should be asking things like, “Will I get more productivity if I hire higher-skilled workers in this plant or in this office, or would I get better results by cutting costs?” The problem is that most organizations base the questions they ask on the data they have in front of them. Because of that, the questions tend to get smaller and smaller—not bigger, as they should be.
What needs to happen, in a world that’s moving into explainable AI, is if you ask a question and your software can’t find the answer, its response should be, “We’d love to solve the problem, but here are the things that are missing. Can you help me find the right data?” That’s much more useful than, “Unavailable, beep-boop-beep.” Explainable AI means that the computer will be able to tell you how it arrived, or didn’t arrive, at a certain decision so that you can weigh AI input as you would from a human source.
It helps democratize decisions. To do this, you need context—what we call mass personalization and scale—to correctly answer whatever question someone might pose. Putting analytics inside the system of action helps you get to this context.
We should start thinking about the outcomes we want to achieve. The first thing is, “Tell me what the hell is going on,” right? The second thing is, “Give me an alert or a notification.” Okay, we can do that. The third thing is, “Make a recommendation.” That’s a little bit harder. And then the fourth thing is, “How do we automate those types of decisions?”
“To prepare for the future of analytics, decide what are your most important questions. Then, work backward to find out how to get the best answer.”
When you can start predicting outcomes, preventing tragedies, meeting compliance requirements, and ultimately getting real-time situational awareness, this gives you an inherent advantage over competitors who aren’t doing these things.
Go to each department and ask them what questions they’ve always wanted to ask, but never could. This will help organizations understand what they are not getting out of their analytics system today. Start there, because then the discussion is centered on the art of the possible.
Put your business issues first, and then you can look at different technologies, or redesigned processes, or a number of other things that can help you answer those questions you’ve always wanted to ask, but couldn’t. In other words, to prepare for the future of analytics, decide what are your most important questions. Then, work backward to find out how to get the best answer.
Learn more about Workday Prism Analytics and how it helps with self-service analytics.