By Marty Cagan · svpg.com · LinkedIn
Chris Jones of SVPG argues that many teams believe they practice product discovery but actually misuse it, falling into recurring anti-patterns. These include using discovery to confirm pre-chosen solutions, treating prototypes as the first experiment, running discovery with only some roles, relying on a single favorite technique, planning a few big experiments instead of many small ones, and outsourcing learning to agencies or innovation labs. The article matters because each pattern wastes time and money while producing solutions that may not work for customers or the business. It closes with principles for doing discovery well: stay open-minded, address value and business risks, stay cross-functional, learn fast, use mixed methods, and make it continuous.
01Key takeaways
- Check whether the ideas you test are the same ideas you ship; if so, discovery is confirming rather than learning.
- Test the riskiest hypotheses first, before investing in working code or polished prototypes.
- Include product management, design, and engineering together in discovery to share first-hand learning and better experiment design.
- Pair quantitative tests with qualitative methods like interviews, since numbers show what happens but not why.
- Run some customer-facing learning every week rather than planning a few large experiments for weeks.
- Keep discovery inside the product team instead of outsourcing it to agencies or separate innovation labs.
02Key sections
- Confirmation-Biased Discovery
- Teams use discovery to validate ideas they already committed to, centering the solution instead of the problem. A sign is when the ideas tested match the ideas shipped.
- Product-as-Prototype and Partial-Team Discovery
- Teams wait until working code exists before experimenting, and often run discovery with only PM and UX, leaving engineering out. Both slow learning and weaken handoffs.
- One-Dimensional Discovery
- Teams rely on a single favorite method such as A/B tests, usability tests, or surveys, which addresses only narrow risks and misses why and whether customers will value the solution.
- Big-Bang and Outsourced Discovery
- Long preparation of a few experiments undermines learning, and delegating discovery to agencies or separate innovation labs breaks the link between customer insight and the team that builds.
- Discovery Done Right
- The article summarizes principles: keep an open mind, test value and business risks, involve every discipline, prioritize fast learning, mix methods, and run discovery continuously.
03From the post
“A partnership dedicated to teaching best practices to product teams and product leaders”
04Frameworks mentioned
Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.