How to Talk to Users — Gustaf Alströmer's YC playbook, field-tested.

Executive summary

This lecture cuts through the theory of "talking to users" and gives founders a concrete, repeatable system: who to contact, how to reach them, what to ask, what NOT to ask, and how to convert raw user conversations into an MVP. The thesis is simple — the best founders never stop talking to users. The discipline starts on day one and continues forever.

1 — Why the best founders talk to users continuously

Most people carry a wrong mental model of how startups begin. The lone-genius eureka narrative is fiction. The Social Network, for example, dramatises Facebook's founding in a way that omits the real engine behind successful products: two-way conversations with real customers.

Gustaf uses two Airbnb stories to make the point visceral:

• Brian Chesky and Amal, the first Airbnb guest. Brian spent an entire weekend in San Francisco with Airbnb's very first guest. The image of that weekend is a real photo, not a stock shot, and it symbolises the two-way conversation that built the company.
• The 50-Airbnb experiment (2010). Brian gave up his apartment and lived in 50 different Airbnbs over several months. The mainstream press missed the point. The real purpose was to speak to every one of those 50 hosts every single day, generating an uninterrupted stream of honest feedback. Gustaf is blunt: "Without the Airbnb hosts, Airbnb would not exist today."

Why users keep founders honest

• Users are the only stakeholders actually paying you anything.
• They have the least incentive to protect your feelings — they will tell you the truth.
• Brian and Joe still receive calls on their personal cell phone numbers from hosts — numbers they published on the Airbnb website early on, choosing not to hide behind anonymous email.

The contrast Gustaf draws is stark: most founders spend their time in front of a computer chasing scalable growth channels like Google Ads, and hide behind donotreply@ email addresses. He suggests you search your own inbox for "donotreply@" and count how many companies are actively avoiding their customers.

2 — Who should you talk to?

Gustaf breaks potential interviewees into three groups, in order of accessibility vs. honesty:

Group 1 · People you already know

• Easiest to contact, most likely to respond.
• Risk: they may soften feedback to avoid offending you.
• Still worth doing — don't be afraid of the ask or the rejection.

Group 2 · Co-workers and former co-workers

• Excellent source if they have domain knowledge relevant to your problem.
• If you're building software for startups, former colleagues are often directly in the target audience.

Group 3 · People outside your network

• The most valuable and most underutilised group.
• YC founders most commonly find early users through LinkedIn, Reddit, Slack/Discord communities, and in-person events.

Practical example — carbon-emissions startup

Gustaf walks through a live example across the lecture. Hypothesis: "Companies want to reduce their carbon emissions but for whatever reasons don't."

Target outreach personas: founders, CEOs and CFOs at startups and larger companies; LinkedIn profiles with job titles containing "carbon," "climate" or "sustainability."

Research goals for the interviews:

1. Does the company care about carbon emissions?
2. Why do they care — or why don't they?
3. Who inside the company cares the most, and why?

Outreach message framework

Two template structures — one for someone you know, one for a stranger.

To someone you know (e.g. a former Airbnb colleague who is now a CFO):

1. Brief personal introduction / mention shared history.
2. One sentence: "I'm starting a new project" — briefly describe it, not in detail.
3. Simple ask: a 20-minute phone or video call.

To a stranger: same structure, slightly adjusted tone for someone who doesn't know you.

3 — How to run a great user interview

Format

Always run interviews over video call, phone call, or in person. Gustaf is explicit: you can learn more from a 5-minute video interview than from 500–5,000 survey responses. Surveys compress the richness of real conversation into binary choices.

Build rapport first

The interviewee will be asked questions no one has asked them before. Create a safe, trusting environment before diving into the problem space.

The golden rule — do NOT introduce your idea early

"Do not introduce your idea until maybe at the end of the call, or maybe not at all. Doing this too early can bias them in their answers. Your role here is to listen, not to talk."

Once you pitch your idea, the interviewee shifts from honest reporter to polite evaluator. You lose the signal you came for.

Notes and recording

If not recording, take written notes as thoroughly as possible during the call. Even if you record, you'll need notes anyway — so note-taking in real time doubles as your translation layer.

Live demo — the carbon-emissions interview

Gustaf acts out a mock interview with "Amy" from "Montevue Inc." The key moments:

4 — What questions to ask

These six questions form the core interview script:

1. "Tell me how you do X today." (X = the specific problem, task, or goal you want to solve.)
2. "What is the hardest thing about doing X?"
3. "Why is it hard?"
4. "How often do you have to do X?"
5. "Why is it important for your company to do X?" — Gustaf flags this as especially critical. Understanding the motivation is the pivotal insight.
6. "What do you do to solve this problem for yourself today?" — Understanding the current workaround is crucial because you are competing with their existing behaviour, not just other products.

Power follow-up phrases

When a user gives a partial answer, use these open-ended bridges to draw out more:

• "What do you mean by that?"
• "Can you tell me more about that?"
• "Why is that important to you?"

5 — Questions you should not ask

These four question types kill the quality of your research:

The biggest danger — focusing on features instead of problems

"Your brain is thinking through all the ways you're going to solve this problem. But what you're here for during this interview is to deeply understand the problems — not actually come up with solutions."

Two vivid examples from the transcript:

• Gmail, early days. Users asked Paul Buchheit to show the inbox and email on the same screen simultaneously. The real problem wasn't layout — it was that Gmail loaded too slowly. The feature request was a symptom, not the root cause.
• Early Airbnb. Guests asked for hosts' phone numbers. The real problem wasn't communication access — it was that the platform hadn't built enough trust for guests to feel safe booking without speaking to the host first.

Gustaf's principle: users generally have good problems but bad solutions. Your job is to diagnose the problem; the prescription is yours to write.

Also: users have no incentive to say no to additional features. Ask them if they want Feature A, B and C — they'll probably say yes to all three. You, as the founder, have far more incentive to determine which one actually matters.

6 — Processing interview insights into an MVP

After 5–10 interviews

1. Consolidate all notes — sticky notes or equivalent software work well for organising themes.
2. Bucket the problems into clusters and identify which problem is most prevalent or most painful.
3. Write your conclusions — what are you actually learning across all conversations?
4. Generate a hypothesis for the solution — not an over-intellectualised framework, just a starting point.
5. Move to MVP design as fast as possible, using accurate information from the interviews.
6. Test the MVP with the same users who gave you the original insights.

Evaluating whether the problem is valuable

Three filters Gustaf uses to determine if a problem is worth building for:

7 — MVP prototype sessions

Once an MVP or even a clickable design prototype (InVision, Figma) exists, test it with real people using these techniques:

• Don't tell them what to do. Give a goal ("Try to make a booking on Airbnb") but let them navigate independently. In a real product, you won't be standing next to them.
• Ask them to think aloud. Have them narrate every screen as they interact: what they see, what the words mean to them, what they expect to happen when they tap a button. This reveals gaps in language and UX clarity.
• The Airbnb team used to go downstairs, find people waiting in their office lobby, hand them a phone with an InVision prototype of something they hadn't even built yet, and watch how they used it.

8 — Keep early users engaged long-term

Gustaf's final, often-overlooked recommendation is to keep interviewees as an ongoing community, not a one-time data source:

• Create a Slack instance or WhatsApp group for your early users — many YC companies do this.
• Make them feel special and exclusive: they have early access to a world-changing product.
• Ship product updates in response to their feedback visibly and quickly — builds trust and demonstrates responsiveness.
• Some users will love connecting with other people in the same situation; you become the enabler of that community.

Quick-reference — the Gustaf framework

Key quotations

"Users and customers will keep you honest. They are the only stakeholders actually paying you anything. If anyone will tell you the truth, it will be them."

"Your role here in this interview is to listen, not to talk."

"Users generally have good problems but also generally bad solutions."

"Without the Airbnb hosts, Airbnb would not exist today."

"Excel or Google Spreadsheets is actually a competitor to many hundreds of startups — quite a formidable one. To move someone off Excel, you need to make the experience of your solution dramatically better."

Field notes — how this maps to a UAE operator

The framework maps directly onto platform and marketplace businesses being built in the GCC right now:

• Chef & food marketplaces (EatCookJoy & co.): before building more features, run 10–15 interviews with chefs and diners using the six core questions. The one to obsess over is "What do you do to solve this problem for yourself today?" — the answer surfaces your real competition (WhatsApp groups, Instagram DMs, direct catering contacts).
• LinkedIn is the Gulf B2B channel. Gustaf explicitly names LinkedIn as a top channel YC founders use to find early users — directly applicable to UAE consulting and SaaS outreach.
• WhatsApp is the local user-community channel. Gustaf's recommendation to build a Slack group of early users maps cleanly to WhatsApp here — it's the dominant business channel in the GCC.
• Multi-venture operators win on validation speed. Convert raw interview insights into a prototype as fast as possible and bring it back to the same users. Ideal for lean operators juggling multiple ventures at once.
• Excel is your real competitor. For Stackbirds-style automation pitched at Gulf SMEs, the incumbent is almost always an Excel sheet on someone's desktop, not another startup. The bar is dramatic improvement on that, not parity.

Source: How To Talk To Users, Gustaf Alströmer (YC Group Partner) — Y Combinator Startup School Library, December 2022.