Search press announcements

Bloomberg Launches Best (and Worst) Ranking Visual Data Product

May 22, 2013


Who doesn’t like a ranking?

Rankings help people gain control over information by creating order, or more precisely, ordered lists. Rankings can dispute (or support) “conventional wisdom” and for those who have a personal stake in a ranking, show them where they fit in. Is your state the best or the worst? How about your school, your company, your boss’ salary?

Still, rankings can be infuriating, particularly to those who perceive that they score poorly on them or to those who recognize that, well, there are rankings and then there are rankings. In a February 2011 New Yorker article, the celebrated and prolific observer Malcolm Gladwell took rankings to task, concluding at the end of his piece that “[who] comes out on top, in any ranking system, is really about who is doing the ranking.”

We agree. But we think there are ways to minimize bias and fairly evaluate data.

At Bloomberg, we started a Rankings team a little over three years ago. It has been mainly used as an internal resource as well as by those who subscribe to the Bloomberg Professional service. Today, with the help of our Visual Data team, we are launching a public showcase of the Team’s top 100 rankings at Bloomberg Best (and Worst) presents a wide range of rankings in categories including Business, Economics, Investing, Lifestyle, Personal Finance and Politics & Policy. (Yes, Bloomberg’s forte has been in business and financial data, but we’ve also created a Lifestyle section with lighter, fun rankings.)

Data permeates everything we do at Bloomberg. In all of our rankings, we use a strictly-enforced discipline that we believe has raised the bar on how rankings are constructed…and should be construed. There are many rankings, for example, that we just can’t, won’t and don’t do. We shy away from “Most Important” or “Most Influential” (or in either case “Least”). These are prime examples of rankings based on subjective information that we find impossible to quantify fairly.  Or, so far.

Hypothetically, consider a ranking of “Most Important Women in Tech Under 30 Years Old.” Indeed, we could readily compile a list of all the women holding executive positions in U.S. public companies in the tech industry who are under 30. But just how would we quantify “importance?”

That’s why we have strictly pledged to create our rankings purely out of data, and do not use surveys or polls.

Every one of our rankings carefully cites the source or sources of data used, the date it was created and includes a brief, but comprehensive methodology so that the reader, if so inclined, might recreate the ranking.

We hope, over time, our approach to rankings will not only help interpret data sets for our readers but also help them become sensitive to, and better discriminate between, questionable rankings and those that are solid.

Laurie Meisler, Head of Bloomberg Rankings