Good problem discovery isn’t about asking customers what they struggle with. It’s about uncovering struggles naturally by focusing on specific workflows they’re already running.

When I ran problem discovery interviews for LEANSpark, I didn’t sit founders down and ask “What are your top 3 problems?” That kind of question invites rehearsed answers and surface-level responses. Instead, I asked them to walk me through how they were currently validating their idea — what was occupying their attention, the tools they used, and how they used them.

The struggles revealed themselves.

The Customer Forces Framework

Customer Forces is built on the Innovator’s Gift concept: innovation is fundamentally about causing a switch from some old way (the status quo solution people are using) to a new way (your product).

You don’t cause this switch by solving problems no one cares about. You cause it by solving existing problems customers are already facing with their current solutions.

The framework views a customer’s switching journey through four lenses:

1. Push

What’s pushing them away from their current solution? These are the frustrations, struggling moments, and unmet needs with the status quo. Without a push, there’s no motivation to switch.

2. Pull

What’s pulling them toward a new solution? This is the desired outcome — the promise of something better. Pull creates the aspiration to change.

3. Inertia

What keeps them stuck with their existing alternatives? Habits, sunk costs, familiarity, “good enough” — inertia is the gravitational pull of the status quo. It’s often the most underestimated force.

4. Friction

What makes them anxious about switching? Learning curves, migration costs, uncertainty about whether the new thing will actually work. Friction stops people at the last mile.

For a switch to happen, Push + Pull must exceed Inertia + Friction. Understanding all four forces is what separates effective interviews from superficial ones.

Broad-Match vs Narrow-Match

One of the biggest mistakes in problem discovery is going too narrow too early. Since LEANSpark is an AI product, it would have been tempting to only interview founders already using AI tools and try to get them to switch to LEANSpark.

But that creates selection bias. If you only talk to a narrow segment, you fool yourself into thinking the market is bigger — or shaped differently — than it actually is.

I intentionally went broad. Instead of “Tell me how you use AI,” the framing was: “Tell me how you’re currently validating your idea.” This opened the conversation to the full landscape of how founders work, not just the AI slice.

Once broad-match interviews revealed patterns, I shifted to narrow-match interviews — seeking out specific use cases where I had observed underserved problems that LEANSpark could address. Things like channel building, running experiments, and accountability tracking.

The goal of narrow-match is threefold: validate the broad findings aren’t false positives, confirm the existing alternatives list, and go deeper on struggles and workarounds to identify your axes of “better.”

The Context Switching Tax

The biggest finding from my interviews was something I labeled the Context Switching Tax.

Founders context-switch constantly — between building, talking to customers, fundraising, managing teams, and learning frameworks. That’s already hard. But now they also have to manage the context window of their AI tools.

I describe this as the Memento problem. In the movie, the protagonist wakes up every day with no short-term memory. He relies on notes, tattoos, and Polaroids to remember who he is and what he’s doing.

That’s what using AI tools feels like for most founders. Every conversation starts from scratch. You’re constantly re-explaining who you are, what your startup does, what you’ve already tried, and what you’re trying to do now. Even when you have a great brainstorming session, you’re left with artifacts you have to collect, organize, and turn into action.

This insight became central to LEANSpark’s design. Instead of another tool that demands constant re-explanation, we needed something that remembers — who you are, what you’re building, and where you are in your journey. Load context once. It remembers. And it learns new skills on the fly.

From “Tool” to “Co-Founder”

The interviews revealed a pattern I didn’t expect. Founders didn’t lack frameworks. They knew about Lean Startup, Jobs-to-be-Done, and other methodologies. The real problem was putting it to practice consistently.

That shifted how I thought about LEANSpark. It wasn’t just a coaching co-pilot that answers methodology questions or runs diagnostics. It was a daily co-founder you could use to run validation experiments.

That’s a much larger scope than where I started. But the customer interviews gave me confidence it was the right direction — because the signal came from customers, not from my assumptions.

The Takeaway

Customer interviews aren’t about validating your feature list. They’re about understanding the forces that drive switching behavior and uncovering problems worth solving by watching how people work today.

Start broad to map the opportunity space. Then go narrow to validate and deepen your understanding. Focus on workflows, not opinions. And pay attention to the struggles people work around but never articulate — those are your biggest opportunities.

I walk through the interview findings in Episode 5 of Building LEANSpark.