Paychex found that initially using AI to accelerate coding only created a backlog for product and design teams. This reveals that a holistic process change across the entire software development lifecycle is needed, rather than optimizing a single function in isolation.
When integrating teams with successful but conflicting cultures (e.g., "planful" vs. "aggressive"), don't force one model. Instead, engage leaders from both sides to co-create a new process that combines the best of both worlds, ensuring broader buy-in and a more resilient outcome.
To counter resistance to AI rooted in job-loss fears, frame it as an investment in your team's skills. This makes employees more valuable in their careers—whether at your company or elsewhere—reframing the change from a threat to a growth opportunity and building trust.
The word "failure" is loaded with negative connotations. A more productive mindset for innovation is to focus the team on reducing "time to learning"—the speed at which you discover a path is wrong. This encourages experimentation and small pivots rather than creating fear around outcomes.
Categorize all R&D work into four buckets: Health (maintenance), Innovation (new value), Parity (catch-up), and Modernization (tech debt). This "HIPM" framework clarifies where resources are going and enables leaders to make purposeful, annual budget shifts into the innovation category.
AI is not a feature but a complete reimagining of user experience. Product leaders must now design for interactions that originate outside their app and for users that are not human. The next wave of "users" will be other AI agents interacting directly with your system's agents.