Helping jobseekers with recommendations based on skill profiles or past experience: Evidence from a randomized intervention

Author: Mirjam Bächli (University of Lausanne)Rafael Lalive (University of Lausanne)Michele Pellizzari (University of Geneva)
Posted: 30 October 2025

Abstract

Searching for jobs is challenging, and online platforms increasingly aim to improve outcomes by offering personalized job recommendations. In a randomized controlled trial with over 1,250 participants, we evaluate recommendations based either on prior experience or on skill profiles assessed at study enrolment. We find that both types of recommendations tend to improve job finding rates. Profile-based recommendations are particularly effective for individuals with limited experience and mismatch in their previous employment. These findings highlight the importance of aligning job search advice with jobseekers’ skills, especially for disadvantaged groups.
JEL codes: J24, J62, J64
Keywords: Jobseekers, online job search, job recommendations