When discovery makes sense
Discovery is the right approach when you're starting fresh. Maybe you're exploring a new market. Maybe you've noticed something interesting and want to understand it better. Maybe you've been handed a vague brief like "figure out what's going on with [industry/audience]."
The common thread: you don't yet have a specific problem or solution in mind. You're trying to build a mental model of how people think, what they struggle with, and what matters to them. This is different from validation research, where you already have a hypothesis and want to test it. Discovery comes first. It's where hypotheses come from.
Who should you talk to? People who fit the general profile you're exploring, and ideally, a mix of perspectives within that group. You're not looking for "ideal customers" yet; you're looking for variety. Different roles, different contexts, different levels of experience.
What you're trying to learn
In discovery, you're not looking for answers to specific questions. You're looking for the questions themselves.
Good discovery interviews help you understand:
- How people currently handle a situation or area of their life
- What frustrates them about it
- What they've tried before and why it did or didn't work
- What "good" looks like to them
- Where they spend time, money, and attention
What discovery interviews should feel like
Discovery interviews are deliberately open-ended. You're following the participant's lead, not steering them toward your assumptions.
A good discovery interview:
- Starts broad and lets the participant define what's important
- Follows the energy, digging in when something sparks frustration or enthusiasm
- Pushes for specifics and real stories, not generalizations
- Avoids planting ideas or hinting at solutions
The goal is to leave each conversation with a richer understanding of their world, not confirmation of your own ideas. That takes restraint. It's easy to accidentally guide people toward the answers you're hoping for.
Common mistakes
Going too broad. "Tell me about your life" isn't useful. You need a domain to explore: a specific part of their work, their routine, their decision-making. "How do you handle [X]?" gives people something concrete to start from.
Pitching too early. If you already have an idea, it's tempting to mention it, "just to see what they think." Don't. The moment you introduce your solution, the conversation shifts from discovery to validation, and you've lost the open-endedness that makes discovery valuable.
Stopping too soon. Five interviews might feel like enough, but patterns often only emerge around interview seven or eight. Keep going until new conversations mostly confirm what you've already heard.
Asking about the future. "What would you do if..." or "Would you want..." invites speculation. People are bad at predicting their own behavior. Focus on what they've actually done.
From conversations to direction
After eight or ten conversations, you'll start noticing patterns. Three people mentioned the same frustration. Two described identical workarounds. Someone's offhand comment echoes what you heard last week. These are themes: recurring concerns that show up across different conversations.
Not every theme is a problem worth pursuing. A problem becomes interesting when it's frequent (happens regularly), painful (people notice it and care), and unsolved (existing options aren't cutting it). If you heard it once, it's an anecdote. If you heard it five times with similar intensity, you might be onto something.
The raw output of discovery is often a messy document: notes, quotes, half-formed observations. That's fine. What matters is that you can point to specific patterns and say: "This came up repeatedly. Here's how people describe it. This is worth investigating."
When sharing with stakeholders, lead with the surprises. What did you learn that you didn't know before? What patterns emerged that no one was talking about? That's usually more compelling than a list of findings.
A practical example
Imagine you're exploring the world of freelance designers. You don't have a product idea yet. You just know this is a group you want to understand better.
You start with questions like: "Walk me through how you find new clients." "What happens after a project ends?" "How do you decide what to charge?" You're not testing anything. You're mapping terrain.
Over ten conversations, patterns emerge. Several people mention the anxiety of irregular income. A few talk about the awkwardness of raising rates. One problem keeps coming up: following up with past clients without feeling salesy.
Now you have something. Not a solution, but a problem worth investigating further. That's what discovery gives you.