Money BI

In my two previous Industry Insights, I discussed the books Freakonomics (Levitt and Dubner) and Outliers (Gladwell), and how they provide key lessons in the 'how to innovatively approach business intelligence'.

Freakonomics is more along the lines of how behavioural intelligence can be derived from raw data, while Outliers looks for above and below performance stories from data. As a final reading recommendation for those in the field of business intelligence, “Moneyball: The Art of Winning an Unfair Game” - a 2003 book by Michael Lewis - ties both of these concepts together in a real-life story about how the coach of an American baseball team puts his career and the team on the line in a high-stakes experiment in advanced data analysis.

BI operates in multiple, competitive business environments where companies of different sizes and budgets attempt to compete and grow.

David Logan is principal consultant at PBT Group.

From a BI industry point of view, the 2011 movie “Moneyball”, starring Brad Pitt, may be the first time Hollywood has glamorised the unglamorous industry of “data science”. Unfortunately for the number-crunchers out there, the attractive Brad is not actually the data scientist, instead he hires the fairly stereotypical slightly overweight geek in the form of Jonah Hill to do the hard work - I guess you can't have it all.

For those not familiar with the movie or the book, they detail the story of the underdog Oakland A's baseball team under their manager Billy Beane. As in most sport, there were large discrepancies in budget between the larger teams such as the New York Yankees, who in 2002 had a payroll of $126 million, while Oakland's was around a third of that. For the previous several seasons, Oakland had won more regular season games than virtually any other team, and teams with larger budgets had performed much worse. The author, Michael Lewis, wanted to understand and explain, how, in the big budget world of baseball, this could be possible.

The answer was that Billy Beane put his trust in data analysis over conventional wisdom. His assistant, Paul De Podesta, had not played pro baseball; he was a Harvard graduate with a distinction in Economics, but you could say his interest lay more on the Freakonomics side of the fence. This went against all conventional wisdom, which said that the baseball scouts with their years of experience and an “instinct” for talent would be the best source of players. However, scouts are people and prone to the same biases that affect most people, unathletic looking athletes are marked down, or high-profile activities (like hitting a home run) have a tendency to be over-valued. Consider the same thing in some companies where customer numbers and company size can be chased at the expense of long-term profitability.

By rating/measuring the value of potential players based on the attributes the statisticians assessed as important to a winning result, Billy and his team effectively purchased a number of players who out-performed the price the market was placing on them. One of the key biases in baseball was the concept of a batting average which roughly measures how often the batter hits the ball; however, actual runs achieved are a more accurate measure of how well the team does; the market was effectively placing an inflated price on a flawed statistic (batting average).

Based on work done by Bill James in his 1977 Baseball Abstract (ultimately a series of 12 Abstracts), Billy and the data analysts set out to change the measures used to rate potential players and by doing so challenging conventional wisdom.

The combination of buying players underrated by the general market based on specific and carefully thought out statistics and selling players at the peak of their careers results in Oakland experiencing incredible success per dollar spent and outperforming the baseball market as a whole.

This might as well be a story of how the business intelligence (BI) industry should be evolving from a report-driven environment to a value-based, results-driven, creator of insights one. BI is the creation of value from data. It operates in multiple, competitive business environments where companies of different sizes and budgets attempt to compete and grow.

So, I suggest you ask the following questions of your BI:

* What makes some businesses succeed and others fail?
* Why do apparently successful companies (and in recent times, banks) report good numbers year after year shortly before they go bust, while others operating below the radar, continue to flourish and only get recognition later in their corporate lifetimes?
* Are the right aspects being measured?
* Do we dig into the underlying conventional measures for weaknesses?
* Do we focus on the more visible and popular measures such as customer numbers and market size because measuring the quality of these numbers is more difficult and requires a blend of economics, psychology and statistics?

By answering these questions above it may assist in reaching your goal of “Money BI” and help your customer “Win an Unfair Game” in a very tough and diverse industry characterised by constant change. When looking at it from a BI perspective, this story can be seen as a great practical example of how BI can truly assist an organisation in standing out from the rest, adding value to the broader business and, of course, achieving set out goals.