December 2007

Competing on Analytics or Black Swan?

Two business books getting a lot of attention these days are "Competing on Analytics: The New Science of Winning." and "Black Swan: The Impact of the Highly Improbable." (For those of you counting, "Competing on Analytics" has attracted 38 reviews on Amazon.com and "Black Swan" rates 217 reviews, more even than the 141 for Alan Greenspan's recent book). Both books were released within a month of each other this spring. But business professionals want to know, "Do these books have conflicting implications for our competitive intelligence or knowledge management efforts?"

Competing On Analytics
At first glance the two books wouldn't seem to have a lot to do with each other or with competitive knowledge management issues. The book derives a lot of its buzz from the current momentum at Google, now commonly seen as the prototypical analytic competitor, although about 30 organizations are mentioned as achieving this status. Competing on Analytics straddles strategy, operations and IT with its claim that becoming a stage four or stage five analytic competitor amounts to a core competence or a core strategy, like one of Michael Porter's original three generic strategies: overall cost leadership, differentiation or focus. Competing on Analytics stops short of making those explicit associations, however.

If you can picture an organization with ever more sophisticated algorithms slicing through rapidly accumulating databases driving real time, accountable decision-making, you've pretty much captured the end-state vision here. Competing on Analytics says that knowable stuff is good and more knowable stuff sliced and diced by ever more powerful algorithms is the wave of the future.

Black Swan
Black Swan, though, focuses on the problem and impacts of unknowable stuff and the misleading nature of the bell curve. Some of its buzz is associated with its contrary view of the current hedge fund boom. It is not a business applications book exactly; it is a discussion of three main ideas: that the most important events in history, the financial markets or our lives are Black Swans, unusual events that by their nature are not very probable and wouldn't be believed if predicted because their nature is so different from past experience, that the impact of these events is large and that our human nature tempts us to create explanations for these events after the fact so that we are lulled into a unjustified sense of orderliness about our world.

Black Swan essentially argues that all the stuff you've got in your ERP or CRM database is okay for incremental improvements but lots of analytic capability could make organizations complacent about the range of probable outcomes and is especially dangerous to rely on for predictive purposes. The Black Swan's author, Nassim Taleb, wouldn't argue for the presence of the 80/20 rule when assessing real value creation over time, he'd argue for something like the 95/5 rule. He decisively demonstrates the human tendency to sweep non-knowable problems to the side and under value their impact and implications.

Implications
The implications? Should you collect more or less data on your customers, competitors, employees and transactions? Use more or less resources for analysis? Where are the points of diminishing return for analyzing quantitative data? For analyzing the unstructured qualitative data that is often the focus of competitive intelligence and knowledge management departments?

These are important questions, but not either / or choices. The answers vary by industry and by market position in that industry, by organization structure and by organization culture. What's certain though, is that knowledge creators and knowledge users in any organization often need help articulating their knowable and non-knowable problems, communicating those to executives and developing the right tools for better decision making.

Like to receive more on this topic? Please contact Tom Davis at tdavis@cygnusassociates.com