The American Prospect Six years ago, I was at a conference at the University of Chicago, the intellectual heart of corporate-friendly capitalism, when my eyes found the cover of the , the business school’s flagship publication. “ ” the headline asked. I wasn’t, so I started reading.
The story looked at how online shopping, persistent data collection, and machine-learning algorithms could combine to generate the stuff of economist dreams: individual prices for each customer. It even recounted an experiment in 2015, where online employment website ZipRecruiter essentially outsourced its pricing strategy to two U of Chicago economists, Sanjog Misra and Jean-Pierre Dubé. ZipRecruiter had previously charged businesses one fixed monthly price for its job-screening services.
Misra and Dubé took the information ZipRecruiter asked prospective clients on an introductory registration screen about their location, industry, and employee benefits. In the initial part of the experiment, ZipRecruiter assigned a random price to each business. The researchers could then see which attributes correlated with a willingness to pay higher prices.
“There were enough things that people involved had disclosed at the registration stage that were associated with their price sensitivities that we could build a pricing algorithm around it,” Dubé told me in an interview. Sure enough, when ZipRecruiter deployed the algorithm, to deliver tailored prices based on the questions customers answ.