Magazine article American Banker

Forget Six Sigma, Bankers Should Study CompStat

Magazine article American Banker

Forget Six Sigma, Bankers Should Study CompStat

Article excerpt

Byline: Sean Sposito

In the 90s, when Bill Bratton was New York City's police commissioner, he developed a system of continuously tracking crime that was given a name that sounds as nebulous and broad as the concept of big data: CompStat.

The method is credited with dramatically reducing murders in the Big Apple. It would later be adopted around the country.

You see, CompStat is a management style. One that involves not only statistical analysis and geolocation, but meetings between executives and low-ranking employees who might not otherwise get the chance to chat.

The process is so famous that it was even referenced on the HBO show "The Wire." Bankers might be smart to adopt similar tactics -- marrying high-level analysis with a culture of information sharing.

"The idea with algorithms is that they are being developed constantly, [honing] the ability to predict," says Bratton, an avuncular figure, speaking in a stream of consciousness. "In banking, it's the idea that you would want to be able to predict that there is an emerging problem by looking at certain trend lines and certain issues. For example, when JPMorgan Chase had the six or seven million dollar loss, wouldn't they have loved to pick up on what the [London] Whale was doing?"

Easy to say, but remember who's talking.

After working stints as Boston's top cop and Los Angeles' chief of police, Bratton is now a consultant running the Bratton Group, which provides leadership advice to public agencies and private companies. He's also the vice chairman of the Homeland Security Advisory Council.

The guy carries so much weight that you could hear a pin drop in a room full of hundreds of bankers during his keynote speech at CEB TowerGroup's annual conference in Boston on Wednesday.

There are, of course, regulatory issues that create divisions within banks, sometimes making it tough to share data and methods that might be beneficial to, say, the guy who deals with mortgages and another bank employee who evaluates transaction fraud. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.