Lattice

Explore your topic from multiple angles. Find at least 5 concrete signals that a real problem exists. Use articles, studies, interviews, and your own observation.

Map everyone who touches this problem. Who is most affected? Who has the most power to change it? Who might resist a solution?

Talk to at least 5 people affected by this problem. Prepare questions in advance, but let the conversation go where it needs to go. Document what they said and what it meant.

Break the problem into its root causes. Use a MECE structure (Mutually Exclusive, Collectively Exhaustive) — no overlaps, no gaps. Identify where the real leverage is.

What is the user really trying to accomplish? Go deeper than the surface request. Define the functional job, the emotional job, and the social job. When do they 'hire' a solution — and what are they 'firing' it for?

Build 3 distinct user archetypes. Each persona should have a name, a backstory, a day-in-life moment, specific pain points, and what success looks like for them.

Choose two axes that capture the most meaningful distinctions in your user base. Place your personas. Identify which quadrant holds the biggest untapped opportunity — and explain why.

How big is this problem? Use a top-down market estimate and an impact opportunity estimate. Don't just show numbers — explain what they mean.

What tools exist today for this problem? Map them. What do they do well? Where do they fall short? What gap does your solution fill?

Synthesize everything you've learned into a solution concept. Start with a 'How Might We' statement. Define your core design principles. Then describe what you'll build.

Synthesize everything from the last two weeks into a single product spec. A PRD is what engineers and designers build from — it has to be specific enough to act on. If it's vague, the prototype will be too.