A lot of people insist that intuition beats analytics. Each of them seems to have a favorite story about an executive who made some decision that defied both data and common sense, yet led to great success. But then, who would brag about all the bold decisions that backfired?
Data won’t do you any good if it’s not gathered and analyzed appropriately for your needs. Still, the fact that analytics can be done badly is a poor excuse for pretending it has no value at all. You don’t hear stories of data-defying decision making at Amazon or Google, two companies known for their consistent use of measurement and testing, and they seem to be doing pretty well. Every executive in the world now knows about those successes and others like them, but not so many are emulating those disciplined data-driven management approaches.
Fear of the truth
The biggest barrier to successful use of analytics in business is simple to understand, and extremely difficult to overcome. When you commit to measurement and testing, you’re also committing to admit that some of your decisions go wrong. That’s a risk many executives aren’t willing to take.
Don’t judge them too harshly. Even executives fear losing their jobs. IE Business School professor Monika Hamori writes that executives now change jobs about once every three years, and that changing companies is not beneficial to their careers. Still, the personal goals and concerns of executives can be at odds with long-term interests of their employers.
MIT’s Michael Schrage attributes resistance to analytics to resentment of accountability, particularly when the associated risks seem to land only on the rank and file staff, rather than top-level management. Resentment, he says, is particularly strong when accountability does not cut both ways. In his words, ”Accountability flows up from the bottom; authority flows down from the top.”
There’s truth in what he says, although I have a somewhat different view. People at every level also resist using and being judged by measures they believe to be inaccurate or irrelevant to good outcomes. These concerns are often entirely rational. Even the strongest advocates of analytics know that not all data analysis is done well.