How to validate your startup idea

Todd Jackson, a seed-stage investor at First Round Capital, interviewed founders of eight companies (Vanta, Flexport, Cocoon, Good Dog, Snackpass, Rec Room, LaunchDarkly, Pinwheel) about how they chose and validated their startup ideas. He identifies three common starting points: a market or space, a better consumer experience, or a problem the founder has personally lived. Across the examples, the recurring pattern is that founders test their assumptions with real people early, using low-fidelity tools like spreadsheets, mockups, fake landing pages, and manual service, before writing much code. The strongest validation signals were unprompted inbound interest and customers willing to pay, not polite interest in hypotheticals. The piece matters because it offers a practical, evidence-based way to reduce the risk of committing to an idea.

01Key takeaways

• Talk to potential customers before writing code, since spreadsheets and mockups are far easier to change than software.
• Treat anything short of enthusiastic "I want this" as a polite no and dig for the underlying pain.
• Find the segment where pull is strongest and sell there first, rather than starting with a large theoretical market.
• Use cheap tests like fake landing pages, manual service, and waitlists to measure real demand before building.
• Validate both sides of a marketplace, since supply-side pain can be harder to crack than demand.

02Key sections

Three ways to find an idea

Ideas usually start from a market, a sense that a consumer experience should be better, or a problem the founder has lived. Each starting point suggests a different validation path.

Market-first validation

Vanta and Cocoon began by interviewing people in a space until a burning pain emerged. Early prototypes and mockups tested whether customers would actually act.

Experience-first validation

Good Dog and Snackpass started from intuition about a better experience, then used conversations, manual service, and campus-based tests to confirm pain on both sides of the marketplace.

Problem-first validation

Rec Room, LaunchDarkly, and Cover built from problems the founders had experienced, then tested whether others shared them through early users, content, and interviews.

Signals of traction

Strong signals included unprompted inbound requests, users inventing their own uses, and early paying customers. Founders noted that these signals often arrived gradually rather than dramatically.

03From the post

“1. What is a good growth rate 2. Preparing for a PM interview 3. Open thread: How is your product org organized? Q: I have an idea for a startup. How can I validate that it’s a good idea, and how will I know when an idea is promising enough to commit to? Committing to a startup idea is possibly the most consequential decision you’ll ever make. It’ll impact everything they do for the next 1 to 50 years, and either lead you to fame and fortune or a brick wall to endlessly bang your head against. No pressure. To help you navigate this critical time, Todd Jackson (partner at First Round Capital) spent 2+ months researching, interviewing, and synthesizing lessons from some of today’s most exciting companies to understand what gave them the confidence to commit to their idea. Below you’ll learn how these companies come up with their idea, validated it, and gained traction—along with a host of lessons learned along the way. In his nearly 20-year career, Todd has helped build some of the biggest products in…”

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.