High-volume recruitment—hiring a large number of people in a short time—is a challenging process, yet one that is common across industries, from hospitality to global manufacturing. A poorly managed process may overwhelm the recruiting team and lead to inadequate hiring decisions. Thanks to HKUST’s Qing Li and co-authors, recruiters now have guidance for optimizing their high-volume hiring with analytics.

“One way to spread out the workload for the recruiting teams and allow candidates flexibility,” say the researchers, “is to adopt a sequential, or rolling, process.” A recruiting season is divided into phases and candidates can apply at any time, allowing for comparison within phases. “For high-volume recruiters, there are two practical questions,” say the authors. How many offers should be made in each phase, and how does the number of phases affect recruitment success?

To answer these questions, they modeled a high-volume recruitment process as a large-scale dynamic program. The recruiter seeks to hire a specific number of candidates, who are assessed in phases. The success of the process is measured by the total assessment score of hired candidates minus a penalty cost if the total number of candidates hired deviate from a preset hiring target.

Based on this framework, the researchers developed a set of easily computable heuristics to determine how many offers to extend during each phase. The practical application of these heuristics was demonstrated through a case study involving graduate student recruitment. The findings show how advanced modelling techniques can enhance recruitment effectiveness and efficiency, ultimately transforming high-volume hiring practices.

“We are the first to provide an analytical tool for computing the number of offers to make in each phase and for evaluating the impact of the number of phases,” say the researchers. “Without such tools, recruiters would have to rely on their gut feeling, which can be time-consuming and could easily lead to poor and inconsistent decisions.”