470 visitors, 3 clicks. Continue or stop?

"Landing page visitors: 470. Users who took purchase action: 3. Given these metrics, should I continue pursuing this idea?" Someone asked that on Hacker News, and nobody could answer, because the number that would answer it was never written down. This episode is about setting your metrics before the test, so the result gives you a clear decision: invest more, stop, or change one thing and run it again. What you'll hear: Why deciding after the test goes wrong Pick a signal that costs people something Three lines to write before you build Where benchmarks come from, and why 200 views is the minimum for a rough signal How often the best teams are wrong: Microsoft, Booking.com and Supercell Try this week: for your next test, write what you'll invest if it works, the go line, the stop line and the middle. Send them to one colleague before the test goes live. Next episode: your CPO is busy with other teams. How do you get a decision out of them? Hand them a better hour. I'm Marcin Klimaszewski and I run ⁠Product Build Day⁠. Send me your question on ⁠LinkedIn⁠, and it might become an episode. Sources: Hacker News: Should I continue based on my fake door test results? (7 February 2025) Atlassian: The State of Product in 2026 (3 September 2025) Norton, Mochon and Ariely: The IKEA effect: when labor leads to love (Journal of Consumer Psychology, 2012) Kromatic, The Real Startup Book: Fake door test Kohavi and colleagues: Online experimentation at Microsoft (2009) HBR Cold Call with Stefan Thomke: At Booking.com, innovation means constant failure (3 September 2019) Ilkka Paananen, Supercell: The next chapter (15 February 2023) Indie Hackers: How do you validate your ideas? (20 September 2025)