From Data Scientists to Decision Scientists: The Value of Hybrid Talent in AI Transformation

Information Systems and Operations Management 

Speaker: Kejia Hu (Oxford Said)

Room Bernard Ramanantsoa

Abstract:

Despite massive investments in artificial intelligence and analytics, most firms fail to realize measurable returns. We argue that the bottleneck is organizational rather than technological, specifically in how firms configure human capital to bridge technical capability and business decision-making. Using a novel dataset of 25 million job postings from S&P 500 firms between 2011 and 2019, we develop a typology that distinguishes Pure Data specialists, Hybrid-D (data specialists with business skills), and Hybrid-B (business specialists with data skills). We find that firms pursuing Hybrid-B hiring strategies achieve 15-22 percent higher return on assets, while Pure-D strategies show no significant performance benefit. Through detailed mechanism tests, we demonstrate that Hybrid-B’s advantage stems from reduced coordination costs, proximity to decision-making authority, and embedded domain expertise. These effects are amplified in firms with longer business specialist tenure, faster decision cycles, and mature IT infrastructure. Extending our analysis to 2023-2024, we show that firms investing in Hybrid-B talent with AI capabilities are disproportionately positioned as AI-integrated leaders. Our findings challenge technology-centric approaches to digital transformation and highlight the critical role of business embedded analytics capabilities in realizing value from AI investments.