<aside> ⚡

In short: the best founders talk to users for the whole life of the company, because users are the only stakeholders who pay you and will tell you the truth. Interview people about their problem (never your idea), watch what they actually do, ignore their feature requests but dig into the problem behind them, and check that the problem is worth paying for before you build.

</aside>

🎙️ Y Combinator · 👤 Gustaf Alströmer · ⏱️ 18 min · 📅 December 1, 2022 · ▶️ Watch on YouTube

🧠 Key ideas

1. Ideas come from conversations, not lazy Sundays. The first photo of Airbnb's founding isn't a hackathon: it's Brian Chesky with the very first guest. Great founders talk to future customers before they have a product.
2. Users keep you honest. They are the only stakeholders actually paying you. Chesky lived in 50 Airbnbs over a couple of months to talk to hosts daily, and the founders still get calls from hosts on their personal phones.
3. Friends answer, but may not tell the truth. People you know respond most, but can be less honest so as not to offend you. Former co-workers, LinkedIn, Reddit, Slack and Discord communities and in-person events are the other common sources.
4. Keep your idea out of the interview. Video, phone or in person (five minutes of video beat hundreds of survey answers), build rapport, and don't mention your solution until the end, or at all. Your job is to listen.
5. Six questions that work: how do you do X today, what's the hardest part, why is it hard, how often, why does it matter, what do you do about it now. Then watch them do it, ideally via screen share.
6. Questions that derail you: "would you use our product?", "which features would make it better?", yes/no questions, "what would a better product look like?", and two questions at once.
7. Good problems, bad solutions. Gmail users asking for inbox and email on one screen really meant "Gmail is slow". Airbnb guests asking for hosts' phone numbers really meant "I don't trust you yet".
8. Check it's worth paying for. Are people paying for other solutions? Are they happy with Excel or Google Sheets, which compete with hundreds of startups? How easy is this audience to sell to? Plumbers and contractors rarely switch tools.

📑 Chapters

00:00 · How startup ideas really start

00:42 · The best founders learn from their users

03:23 · Who should I talk to?

05:58 · How to interview potential customers

09:42 · Follow-ups and questions not to ask

11:15 · Problems, not features

12:51 · From notes to MVP, and is it worth paying for?

14:37 · MVP prototype sessions and keeping users close