Pricing · Two randomized controlled pricing field experiments: first trains heterogeneous price-response model; second tests recommendations out of sample.
Personalized Pricing and Consumer Welfare
Journal of Political Economy · 2023
Summary
Data-driven personalized pricing increased profits out of sample; first experiment revealed substantial unexercised market power.
For your marketing
Possible founder application
Use the study to frame a pricing or packaging hypothesis and identify outcomes to measure; estimate effects with your own customer mix and willingness-to-pay evidence.
Reported result
What the study found
Data-driven personalized pricing increased profits out of sample; first experiment revealed substantial unexercised market power.
Research question
What the authors investigated
Can randomized price experiments support personalized prices, and what are their profit and welfare implications?
Design
Two randomized controlled pricing field experiments: first trains heterogeneous price-response model; second tests recommendations out of sample.
Sample and setting
Large online service marketplace (ZipRecruiter); new consumers randomly quoted monthly prices.
Scope and transfer
Limitations to keep in view
Prices are not fully personalized because users with the same predicted response share a price; welfare/generalization depend on the service and model.
Research sources
Publisher/working-paper description reports two RCTs, out-of-sample validation, online-service setting and profit result. Direct publisher-page retrieval was recorded on 2026-10-09; this profile uses the publisher abstract/metadata and does not claim independent review of unavailable full text.