Product Discovery: How to Find What to Build Before You Build It

The most expensive mistake in product is building something well that nobody wanted. Product discovery is the work you do before building to make sure you're building the right thing. It's the difference between shipping confidently and shipping hopefully. Here's a practical approach that doesn't require a research department.

What product discovery is

Product discovery is the process of deciding what to build and validating that it's worth building — as opposed to delivery, which is building it well. Discovery answers: Is this a real problem? For enough people? Will our solution actually solve it? Is it worth the effort?

Skipping discovery is how teams end up with beautifully engineered features that get no usage. A little discovery upfront saves months of wasted delivery.

The core questions

Good discovery works through four questions, roughly in order:

1. Is the problem real? Do customers actually experience this, or do we just think they might?
2. Is it widespread and painful enough? One person's annoyance isn't a roadmap item; a pattern of real pain is.
3. Will our solution solve it? Does our proposed approach actually address the underlying job, or just the surface request?
4. Is it worth it? Does the value justify the effort, given everything else we could build?

Practical discovery methods

You don't need a formal research program. A few lightweight, continuous habits cover most of it:

Mine your feedback board. A feature request board is a discovery engine that runs itself: requests reveal problems, votes reveal how widespread they are, and comments reveal the context. This answers questions 1 and 2 almost for free.

Talk to users. Short, regular interviews are the highest-value discovery activity. Ask about their situation and goals, not your feature ideas. Ten conversations reveal patterns a survey can't.

Look at behavior. Analytics show where users struggle or drop off — problems they may never articulate. Pair this with what they say on your board.

Validate before building. For bigger bets, test cheaply first: a mockup, a fake-door test, a landing page, a prototype. Confirm demand before committing engineering time.

Continuous, not a phase

The modern view is that discovery isn't a phase before a project — it's a continuous habit running alongside delivery. You're always learning what to build next while you build the current thing. A live feedback board makes this natural: discovery signals arrive constantly, not just during a research sprint.

Once discovery tells you what matters, prioritization frameworks help you sequence it, and how to collect customer feedback covers the input channels.

Start discovering

The cheapest discovery tool is a place where users continuously tell you their problems and vote on them. Create a free feedback board on FeatureRequest and turn product discovery into an always-on habit rather than a guessing game.