Abstract

Digital matching platforms promise to reduce frictions on the labor market by providing low-cost information on available positions and candidates. As such, they may form a welcome addition to the toolbox available to Public Employment Services to bridge labor supply and demand. However, their adoption is not without challenges: vulnerable populations may face difficulties in utilizing digital tools, potentially widening existing inequalities in the labor market. In this study, we evaluate the effectiveness of a communication campaign by e-mail designed to encourage the use of an online profile platform maintained by the French Public Employment Service, France Travail. To drive job seekers’ engagement with the platform and address individual barriers to adoption, we design several e-mail templates that combine information, support, or motivational content. We implement an adaptive experiment (contextual bandit) to efficiently collect data on the impact of the different e-mails. This method leverages past job seekers’ take-up responses and characteristics to determine future e-mail allocation, gradually reducing the allocation of less promising templates. Additionally, we construct a personalized e-mail allocation strategy aimed at maximizing engagement and test it against a random allocation approach. Our findings reveal that e-mails positively impact the usage of the platform, as measured by a wide range of outcomes. However, attempts at learning a personalized e-mailing strategy do not manage to improve on a random allocation of e-mail templates significantly.

Co-authors : Morgane Hoffmann, Charly Marie, and Guillaume Bied

Practical information