AI Won't Save Your Product. Saying No Will.

In the rush to embrace AI, too many enterprise platforms are prioritizing novelty over necessity. The result? Solutions in search of a problem, features that never scale, and teams pulled in a hundred directions with little lasting impact.

After 15 years building enterprise products at scale, I've found that the most effective way to build is not to chase the next trend, but to say no to it.

That's not an intuitive stance in our industry. Product culture still glorifies bold bets, flashy features, and hypergrowth. But in real enterprise environments (the kind with complex architecture, global users, and real operational risk) saying no is often the most strategic move a product leader can make.

Take AI. At ServiceNow, I led the design and deployment of foundational AI workflows now used across support and enablement functions, including agentic automation, case summarization, and generative guidance. These weren't experiments. They became core features used by thousands daily, reducing manual effort, improving resolution speed, and integrating seamlessly into legacy systems.

The key? We didn't build AI for AI's sake. We focused on operational integration, telemetry, driven iteration, and user outcomes. We chose to say no to standalone pilots, and instead built systems that disappear into the work itself. That's when AI becomes useful- not when it's visible, but when it's trusted and routine.

The same principle shaped our enterprise learning transformation. I led the launch of a first-of-its-kind AI-personalized learning platform built directly into workflow, designed not just for employees, but for partners and customers. Unlike traditional LMS systems, our platform tailors learning journeys based on role, behavior signals, and business context, driving measurable gains in time-to-competence and product adoption. This initiative has since been recognized as a model for scalable enablement within the enterprise ecosystem.

These contributions weren't the result of solo vision. They were built on rigorous prioritization, platform thinking, and the discipline to walk away from ideas that wouldn't scale. In fact, some of the most impactful decisions I've made were to stop projects early, when the effort couldn't justify the downstream complexity.

This is the neglected truth of enterprise product leadership: every "yes" creates long-term obligations — to support, scale, integrate, and govern. That's why I advocate for product strategy that isn't just about direction, but about operationalization. It's not enough to write the fancy deck. The real work lies in building the guardrails, rituals, and systems that make that strategy executable at scale.

As enterprise software evolves toward intelligent systems and agentic automation, product leaders will be judged not by the features they launch, but by the clarity they bring to complexity and the impact they make durable.

In that world, saying no is not a constraint. It's a superpower.