Note: After publication of this post, the Journal’s graphics server was hacked. Graphics associated with this story may not be available.
For some people, the price of a Swingline stapler on the Staples.com website is $15.79, while for others, it’s more than a dollar cheaper, at $14.29. This is for the same stapler, at the same time, with the same shipping costs and taxes.
So what’s going on?
My colleague Jeremy Singer-Vine and I found that the Staples website was displaying different prices to people after it had estimated their locations — and that, specifically, Staples appeared to be offering discounts to people who were closer to rival brick-and-mortar stores.
The story, which was in The Wall Street Journal in December, identified several companies that consistently adjusted prices and offers online based on a shopper’s characteristics, including location. It was a fun story to write and to report. But perhaps the most important part of the reporting of this story is actually found in the technical methodology, which appeared only online.
This reporting was primarily a technical feat by Jeremy and another colleague, Ashkan Soltani. The two of them built custom software to analyze pricing on sites. And on Staples, Jeremy simulated visits from all the more than 42,000 ZIP Codes in the U.S. and analyzed the results statistically.
So why is this important? Because this type of work allows us as journalists to develop theories and test them ourselves. Without this, we would have to rely on others to tell us what’s happening — when too often, people simply don’t want anyone to know what’s going on.
Sure, it’s not possible to do this type of journalism on every subject. But I find it helpful to think regularly about whether a subject might benefit from data gathering and analysis.
And I’ve found that empirical journalism particularly helpful when studying the Internet. It’s something my editor Julia Angwin pioneered in the What They Know series, which reshaped the debate on digital privacy precisely because it produced empirical evidence in addition to a compelling story.
Empirical journalism isn’t a new idea, of course. For decades, journalists have developed stories by painstakingly investigating and compiling information. Recently, data-driven journalism has produced great work as well. But the type of reporting in the What They Know series is slightly different; it involves not just analysis of an existing dataset, but rather the gathering of entirely new data. And it’s the future of technology coverage.