<!-- https://letsfindwhy.com/use-cases/customer-discovery -->
<!-- title: Customer Discovery Interviews at Scale - LetsFindWhy -->
<!-- description: Run customer discovery interviews without filling your calendar. An AI moderator interviews each customer from a link, and Ask AI finds the patterns across every transcript. -->

[Use cases](https://letsfindwhy.com/use-cases)

# Customer discovery interviews, twenty at a time

LetsFindWhy lets a product team run discovery interviews without giving up its calendar. Describe what you want to learn and AI project setup drafts the screener and the discussion guide. The AI moderator then interviews each respondent one to one from a link, asking its own follow-up questions, and Ask AI and generated insights pull the patterns out of every transcript with quotes you can play.

## Why discovery stops after five interviews

Every product team agrees it should talk to customers every week. In practice each interview is a calendar slot, a note-taker and an hour of write-up, so discovery stops at five conversations and the roadmap gets decided on the most memorable one.

When the interviews run themselves and the transcripts are searchable together, twenty conversations take about the same effort as five, and a pattern across twenty carries more weight than any single quote.

## How to run discovery interviews in LetsFindWhy

1. **Start from a brief** Describe who you want to talk to and what you want to learn. AI project setup drafts a screener and a discussion guide, and you edit both before anything goes out.

2. **Recruit the right people** Share the signup link with your users, your community or a panel provider. The screener qualifies each signup automatically, so only the people you described are invited.

3. **Let the AI moderator interview** With Immediate AI interviews on, every qualified respondent gets a personal link and is interviewed one to one when it suits them. Set the open questions to “With follow-ups” so the moderator digs into the story behind each answer.

4. **Find the patterns** Generate insights for up to six findings, each resting on at least two passages from the transcripts. Ask AI answers your follow-up questions with quotes, and you can clip a quote to share with the team.

## Example discovery interview guide

1. Tell me about your role and what a typical week looks like. (Open ended)
2. When did you last run into this problem? Walk me through what happened. (With follow-ups)
3. How do you deal with it today, and what does that cost you in time or money? (With follow-ups)
4. What have you tried that did not work? (Open ended)
5. If you could change one thing about how this works today, what would it be? (With follow-ups)

## Tips for better results

- Ask what people did last time, not what they would do. Past behaviour is evidence; predictions are guesses.
- Keep your own idea out of the guide until the last question. Once people see a solution, they start reviewing it instead of describing their problem.
- When an interview surprises you, build the next study from it: LetsFindWhy can draft a reusable guide from a finished session’s transcript.

## Frequently asked questions

### How many discovery interviews do I need?

Patterns usually start repeating after 8 to 12 interviews with one kind of customer. The AI moderator runs interviews in parallel, so running more costs interview minutes rather than calendar time.

### Can I talk to some customers myself?

Yes. Schedule a live session for the conversations you want to lead, and invite colleagues to watch as observers. Live and AI-moderated interviews sit in the same project and are searched together.

### Can I trust the AI’s findings?

Every Ask AI answer and every insight cites the transcript passages behind it, and each citation opens the recording at that moment. Play the quotes before you act on a finding.

## Learn more

- [AI-moderated interviews](https://letsfindwhy.com/docs/ai-moderation)
- [Discussion guides](https://letsfindwhy.com/docs/discussion-guides)
- [AI analysis](https://letsfindwhy.com/docs/ai-analysis)
