Victim of Your (Customer's) Own Success

43 Pages Posted: 26 Sep 2019 Last revised: 20 May 2020

See all articles by Yikun Jiang

Yikun Jiang

University of California, Berkeley

Nathan Yang

Cornell University

Date Written: May 19, 2020


Under the context of digital health, I explore the potential opportunity costs associated with giving customers a sense of success. Using large-scale data from a popular mobile fitness application, I show that many users who enjoy certain past successes need not necessarily upgrade to the premium version, or even continue using the app. These observed dynamics motivate a structural model of self-reputation, where users learn about their own ability at maintaining healthy behavior. The model estimates reveal that improvements in perceived ability do not necessarily increase their desire to upgrade, and in some cases, may even increase the incentives of quitting. Furthermore, I show that the observed heterogeneity in their prior beliefs about true ability matches patterns found in past research on self-confidence. Finally, counterfactual simulations reveal that improvements in the salience of personal performance metrics, while helpful for the user's learning process, would at the same time decrease the number of upgrades to the paid version.

Keywords: Behavioral Analytics; Big Data; Customer Relationship Management; Freemium; Goals and Motivation; Health and Wellness; Health Education; Learning Model; Particle Filtering; Self-Control; Self-Reputation; Subscription Model; Wearables

JEL Classification: I120, L84, M21, M31

Suggested Citation

Jiang, Yikun and Yang, Nathan, Victim of Your (Customer's) Own Success (May 19, 2020). Available at SSRN: or

Yikun Jiang

University of California, Berkeley ( email )

310 Barrows Hall
Berkeley, CA 94720
United States

Nathan Yang (Contact Author)

Cornell University ( email )

Dyson School, Warren Hall
Ithaca, NY 14853-6201
United States
6072551590 (Phone)


Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics