What is continuous discovery?
Continuous discovery is a way of doing product discovery as a weekly habit instead of a one-off project: the team building the product talks to customers at least once a week and runs small research activities in pursuit of a desired outcome. Teresa Torres defined the term in her book Continuous Discovery Habits.
- Updated
How does continuous discovery work?
Torres's Product Talk glossary describes it as "weekly touch points with customers by the team building the product." In practice, teams following her approach combine a few habits:
- A clear outcome. The team works toward a measurable outcome, not a list of features.
- Weekly customer interviews. Short, story-based interviews every week, run by the people who will build the product, usually the product trio of a PM, a designer and an engineer.
- Mapping the opportunity space. What the team learns is organized on an opportunity solution tree, linking customer needs to the outcome.
- Assumption testing. Before building, the team tests the riskiest assumptions behind each solution with small, fast experiments.
Continuous discovery vs project-based research
| Continuous discovery | Project-based research | |
|---|---|---|
| Cadence | Every week, indefinitely | A study every few months |
| Who talks to customers | The team building the product | Often a separate research team |
| Evidence freshness | Days old | Can be months old by the time it is used |
| Decision style | Many small, reversible decisions | Fewer, larger bets |
Common mistakes with continuous discovery
- Interviewing without synthesizing. Weekly calls that never become insights produce a pile of recordings, not decisions.
- Letting recruiting become the bottleneck. Teams that recruit each interview by hand stop within a month. Automate scheduling from your product or support queue.
- Asking about features instead of past behavior. Ask customers to tell a specific story about the last time they hit the problem.
- Starting each week from zero. If last week's findings are not in a shared, searchable place, the habit produces no compounding knowledge.
How does Praxiom AI help with continuous discovery?
Praxiom AI handles the synthesis side of the habit. Each week's interviews and tickets are turned into insights with quotes tagged to their source, and those insights stay in the workspace so the next week builds on the last. The Power plan adds overnight research cycles and autonomous web research.
Frequently asked questions
Who coined continuous discovery?
Teresa Torres, a product discovery coach who writes at Product Talk, defined and popularized the term. Her book Continuous Discovery Habits describes the practice in detail.
How often should you interview customers in continuous discovery?
At least weekly. The weekly rhythm is what makes it continuous. Teams that are not interviewing at all can start with fewer and build up to a weekly cadence.
Is continuous discovery the same as continuous delivery?
No. Continuous delivery is an engineering practice for shipping code frequently and safely. Continuous discovery is a product practice for learning from customers frequently. Many teams run both.
Do I need a researcher to do continuous discovery?
No. The point is that the team building the product talks to customers directly. Researchers can coach the team and run larger studies, but they do not replace the weekly habit.
Sources
Run your product discovery on autopilot.
Bring your interviews, tickets and surveys. Get scored insights, a PRD and tickets.