AI's Role and Limits in Assessing Potential

Thanks to HRBench for powering this episode. To find out more about the company building the future of people intelligence, reach out to book a demo at hrbench.com/directionallycorrect !

Check out this episode of the #1 people analytics podcast with special guest, Allan Church, Co-Founder and Managing Partner at Maestro Consulting!

In this wide-ranging and deeply insightful conversation, Cole Napper sits down with Allan Church to unpack one of the most debated and often misunderstood topics in the field of talent management: the distinction between performance and potential. Drawing on decades of experience at PepsiCo and beyond, Allan challenges the simplistic assumptions that many organizations still rely on—particularly the idea that high performance automatically equates to high potential. Instead, he emphasizes that success in a current role predicts performance at a similar level of complexity, not necessarily at the next level, a nuance that is frequently overlooked in practice.

The discussion explores how organizations continue to rely on reductionist frameworks like the nine-box grid, often without the rigor or clarity needed to differentiate performance from potential effectively. Allan explains that most leaders struggle to make this distinction, especially when assessments are conducted simultaneously, leading to flawed talent decisions and reinforcing biases. He introduces a more structured and science-based perspective on potential, outlining key components such as cognitive ability, personality, motivation, learning agility, and leadership capability—many of which can be measured through validated tools rather than intuition alone.

A key theme throughout the conversation is the tension between simplicity and accuracy. Organizations often seek clean, easy answers, but Allan argues that meaningful talent evaluation requires embracing complexity and using multiple data sources. He also highlights the importance of customized leadership models that reflect an organization’s strategy and future direction, rather than relying on generic frameworks. These models not only guide assessment but also align development, feedback, and succession planning in a cohesive way.

The conversation also dives into the evolving role of AI in talent assessment. While AI can aggregate and analyze large amounts of data, Allan cautions that it is only as good as the inputs and outcomes it is trained on. Without a clear definition of potential and robust underlying data, AI risks reinforcing existing misconceptions rather than improving decision-making. He also raises important ethical considerations around monitoring and data collection, noting the potential trade-offs between insight and trust.

Beyond assessment, Allan addresses broader challenges in talent management systems, including the persistent dissatisfaction with performance management. He describes it as a “no-win” system—necessary for differentiation and feedback, yet universally disliked. Rather than seeking a perfect solution, he suggests organizations focus on execution, consistency, and alignment with business needs. Similarly, he argues that many failures in succession planning stem not from flawed design, but from poor follow-through.

Throughout the episode, Allan brings a systems-thinking perspective, emphasizing that talent management must balance investment in high potentials with development opportunities for the broader workforce. He challenges the false dichotomy between elite talent focus and inclusive development, advocating for a more integrated approach that supports both organizational performance and employee growth.

This episode is packed with practical wisdom, candid reflections, and thought-provoking perspectives on how organizations can move beyond outdated assumptions and build more effective, evidence-based approaches to people analytics and talent strategy.

If you like this episode, you’d also love exploring prior episodes—visit colenapper.com for the full archive and show links.