Rapid recruitment in retail: Leveraging AI in the hiring of hourly paid frontline associates during the Covid-19 Pandemic

Hunt, W. and O'Reilly, J. (2022), Digit Working Paper No. 3

Abstract

Increased demand due to the Coronavirus pandemic created the need for Walmart to onboard tens of thousands of workers in a short period. This acted as a catalyst for Walmart to bring forward existing plans to update the hiring system for store-level hourly paid associates in its US stores. The Rapid Recruitment project sought to make hiring safer, faster, fairer and more effective by removing in-person interviews and leveraging machine learning and predictive analytics. This working paper reports on a case study of the Rapid Recruitment project involving semi-structured qualitative interviews with members of the project team and hiring staff at five US stores. The research finds that while implementation of the changes had been successful and the changes were largely valued by hiring staff, lack of awareness and confidence in some changes threatened to undermine some of the objectives of the changes. Reservations about the pre-employment assessment and the algorithm’s ability to predict quality hires led some users reviewing more applications than perhaps necessary and potentially undermining prediction of 90-day turnover. Concerns about the ability to assess candidates over the phone meant that some users had reverted to in-person interviews, raising the risk of Covid transmission and potentially undermining the objective of removing the influence of human bias linked to appearance and other factors unrelated to performance. The impact of awareness and confidence in the changes to the hiring system are discussed in relation to the project objectives.

Citation

Hunt, W. and O’Reilly, J. (2022) ‘Rapid Recruitment in Retail: Leveraging AI in the hiring of hourly paid frontline associates during the Covid-19 Pandemic’, Digit Working Papers No. 3, University of Sussex, Falmer. Available at: https://dx.doi.org/10.20919/ALNB9606.

Download working paper

Share this: