InnoCentive drug-candidate market experiment

InnoCentive’s drug-candidate market experiment tested whether a market could combine employees’ dispersed knowledge to forecast which potential drugs would win FDA approval. Using realistic profiles and experimental data for six candidates—three known to be approved and three rejected—the market quickly separated winners from losers, offering a practical example of collective judgment applied to a consequential corporate decision.

Purpose and Stakes

For a pharmaceutical company, choosing which potential drugs to develop is among its most important decisions: profits depend on maximizing successful drugs while minimizing unsuccessful ones. A reliable way to anticipate FDA approval would therefore be valuable. The experiment asked whether employees could make that distinction through a market, rather than relying solely on a conventional hierarchy or a small group of decision-makers.

Design and Outcome

InnoCentive, a spinoff of Eli Lilly, constructed realistic profiles and experimental data for six different drugs. The company already knew the outcomes: three candidates would be approved and three rejected. Employees from a diverse mix of backgrounds made up the market. Once trading opened, prices for the eventual winners rose quickly, while prices for the candidates destined for rejection fell. The experiment thus showed the market’s collective judgment distinguishing the two sets of candidates using the information represented in the profiles and data.

Significance and Limits

The result supports the book’s broader argument that markets can help organizations aggregate knowledge scattered among people with different perspectives. Their appeal lies not in making corporate decisions automatically, but in providing information that can inform those decisions. Anonymity and a clear market outcome can reduce the influence of internal politics, sycophancy, and status, while giving participants an incentive to identify and act on useful information. The reported experiment is a demonstration of this potential in one drug-candidate exercise; it does not establish that markets should replace the judgment or authority involved in final decisions.