Acqui-hiring or Acqui-quitting: Data-driven Post-M&A Turnover Prediction via a Dual-fit Model: Completed Research Paper

Gaining highly skilled human capital is one of the key motivations for mergers and acqui- sitions (M&A), particularly in knowledge-intensive sectors such as the technology industry. However, the inherent cultural differences and organizational misalignments during the in- tegration process can lead to significant tensions and a high rate of talent turnover, which may ultimately result in integration failure. Hence, it is crucial for organizations to proac- tively anticipate and manage the potential effects of such events on employee turnover. The predominant perspective in existing literature focuses on the dyadic relationship between merging firms while a few other studies recognize the fit between employees and the firm. However, there has been a lack of endeavor to unify these two factors into a coherent frame- work. In this paper, we propose a novel data-driven neural network approach to predict the talent turnover trend during the post-M&A phase, by considering the compatibility between the merging companies as a key factor. Specifically, drawing on organizational theories, we develop a dual-fit heterogeneous graph neural network with 1) Organization to Organiza- tion (O-O) fit, which captures the relatedness and similarity at the firm level, and 2) Person to Organization (P-O) fit, which represents the compatibility and cultural closeness at the employee level. By leveraging this framework, we can effectively integrate multi-sourced, heterogeneous data to gain a more nuanced understanding of the compatibility between firm pairs. Our proposed approach is evaluated on a large-scale real-world dataset and bench- marked against state-of-the-art methods. Experimental results demonstrate the superior- ity of our approach in predicting talent turnover trends during the post-M&A phase. Our approach also offers interpretable results and provides valuable insights for organizations seeking to manage talent effectively during and after M&A events.