When Less is More: Analysis and Empirical Evidence of Control Limit Strategies in a Diagnosis Crowdsourcing Platform
Jingxuan Geng, Guangwen Kong, Marco Shaojun Qin
Under 1st round review at Management Science
We consider a diagnosis crowdsourcing platform that allows patients to seek multiple medical diagnoses from doctors online. We find that using a commission-based pricing mechanism alone may yield a downward distortion on price to prevent the over-participation of doctors compared to a centralized benchmark. By imposing a control limit on the number of diagnoses received per inquiry, the platform can charge a higher price while maintaining the appropriate number of responses from doctors. When patients are sensitive to delays in receiving diagnoses, interestingly, a platform may benefit from a patient’s increased delay sensitivity because it plays a similar role as the control limit by discouraging late-arriving doctors from participating. As a result, the profit improvement from imposing a control limit mechanism may decrease with delay sensitivity. When doctors are heterogeneous in their service quality, the undesirable outcome of low-quality doctors driving out all high-quality doctors may occur. A control limit could not only increase high-quality doctors’ participation by increasing their chances of being rewarded by patients but also increase consumer surplus. We empirically test the predictions using data from a large diagnosis crowdsourcing platform and find that it supports the results from the model analysis.