How To Talk To Users | Startup School

Building a successful startup begins with a fundamental truth: great founders talk to their users and future customers before they even have a product, and they continue doing so throughout the entire lifetime of their company. This comprehensive guide, derived from Y Combinator's Startup School, breaks down the exact frameworks needed to find users, conduct high-signal interviews, extract real problems, and translate those insights into a Minimum Viable Product (MVP).

1. The Myth vs. Reality of Startup Origins

Most people harbor completely incorrect assumptions about how startups are born. Popular culture (such as Hollywood films dramatizing Silicon Valley origins) often portrays breakthrough startup ideas as sudden epiphanies born during a late-night coding session or a lazy Sunday financial planning meeting.

In reality, sustainable companies are forged through relentless, direct human contact with the people experiencing a problem.

Note: A famous historical parallel involves Airbnb. While movies depict isolated genius, the real Airbnb was born when co-founder Brian Chesky hosted Amal (Airbnb's very first guest) for a weekend in San Francisco, engaging in deep, two-way conversations that laid the foundation for the entire business.

Why You Must Talk to Users Continuously

• Keeping You Honest: Users and customers are the ultimate reality check because they are the only stakeholders actually paying you money.
• Uncovering Hidden Pain Points: Anonymous feedback forms and "do-not-reply" email aliases insulate founders from the brutal, valuable truth.
• Deep Empathy: Early Airbnb founders Brian Chesky, Joe Gebbia, and Nathan Blecharczyk put their personal cell phone numbers directly on their website, refusing to hide behind automated systems so they could learn daily from hosts.

2. Real-Life Analogy: The Airbnb Living Experiment

To understand the lengths to which world-class founders go to connect with users, consider Brian Chesky’s 2010 Living Experiment.

Real-Life Analogy: Living in 50 Airbnbs

Instead of sitting behind a computer screen trying to optimize anonymous Google Ads for scalable growth, Brian Chesky gave up his permanent apartment and spent months living in 50 different Airbnb properties across the country.

• The Superficial Take: Observers thought it was just a PR stunt or a crazy travel hassle.
• The Founder Reality: By living in those spaces, Brian was forced into a daily, mandatory two-way conversation with 50 distinct hosts. This intimate, ground-level feedback loop provided invaluable qualitative data that saved Airbnb from stagnation and deeply shaped its core product evolution.

3. Finding and Sourcing Your Early Users

Before you can interview anyone, you must identify where your target audience lives and how to reach out without sounding like generic spam. Founders often make the mistake of looking exclusively within their immediate echo chambers.

The User Sourcing Hierarchy

1. Personal Network & Friends: Easy to reach out to and likely to respond, but they carry a high risk of being overly polite or dishonest to avoid hurting your feelings.
2. Co-workers and Former Colleagues: Highly reliable if you are building B2B software for startups or industries where your past colleagues possess deep domain expertise.
3. Out-of-Network Communities: The gold standard for objective feedback. Source these users via targeted channels:
   • LinkedIn searches using industry-specific keywords (e.g., "carbon", "climate", "sustainability").
   • Subreddits and niche forum boards.
   • Slack and Discord developer or professional communities.
   • In-person industry meetups and events.

4. Crafting High-Response Outreach Messages

When reaching out to prospective interviewees, keep your message concise, personal, and strictly focused on learning rather than selling.

Outreach Template 1: Warm Contact (Former Colleague / Acquaintance)

Hi [Name],

Hope you're doing well! It's been a long time since our days working together at [Company]. 

I’m currently exploring a new project in the [Industry/Problem Space] space and would love to catch up for 20 minutes to hear about your perspective on how things work over at [Their Current Company]. 

Would you be open to a quick video call sometime next week? No pressure at all!

Best,
[Your Name]

Outreach Template 2: Cold Contact (LinkedIn / Email)

Hi [Name],

I came across your profile while researching how companies handle [Specific Problem, e.g., carbon accounting compliance] and noticed your extensive background in sustainability. 

I’m currently researching workflow challenges in this space and would love to ask you a few questions about how [Their Company] approaches this today. 

Would you be open to a brief 20-minute phone or video call next week? I'd greatly appreciate your insights.

Best,
[Your Name]

5. Anatomy of a High-Signal User Interview

Conducting a successful customer interview requires strict discipline. Your goal is to listen, not to pitch.

Golden Rules of Customer Interviews

• Medium Matters: Conduct interviews over video calls, phone calls, or in-person. You can learn infinitely more from a 5-minute face-to-face video chat than from 500 superficial survey responses.
• Build Rapport: Establish a comfortable environment since interviewees are often answering questions no one has ever asked them before.
• NEVER Pitch Early: Do not introduce your product idea or solution until the very end of the call (if at all). Introducing an idea prematurely biases their answers and makes them tell you what you want to hear.
• Watch Behavior Over Opinions: Do not just ask what people think; observe what they do. Whenever possible, ask them to share their screen and walk you through their current manual workflow.

The Problem-Discovery Script Walkthrough

The following mock dialogue demonstrates how an effective founder interviews a potential user without ever revealing their product concept:

Founder: Hi Amy, great to see you! Thanks for taking the time. I'm exploring an idea in the carbon emissions space and wanted to ask you a few questions about how your company, Montevue Inc., handles this today. Do you care about your carbon emissions?

Amy: We do actually care, and we use an outside consultant who creates these PDF reports for us once a year.

Founder: Interesting. What do you actually do with that report once you get it?

Amy: Right now, I don't think we take much action on it, unfortunately.

Founder: Why not? 

Amy: It's just really hard to know what to do with the information. The data is super dense, and it uses words we don't really understand.

Founder: Tell me more about that.

Amy: Well, we hired this consultant, and they asked us to fill out a massive Word doc. Most of the data going into it feels inaccurate anyway, and I feel like we need another expert just to figure out what to do with the data.

Founder: Got it. Why is it important for your company to track carbon emissions at all?

Amy: Well, our leadership cares about sustainability, and plus, I think there are new regulations coming down the pipeline.

6. Questions to Ask vs. Questions to Avoid

To extract pure, unadulterated signal from your interviews, use open-ended exploratory questions and ruthlessly eliminate leading questions.

High-Signal Questions (Ask These)

• "Tell me how you do [X] today?" (Focuses on current workflows and baseline behavior).
• "What is the hardest thing about doing [X]?" (Pinpoints active friction and pain).
• "Why is that hard?" (Uncovers underlying system limitations).
• "How often do you have to do [X]?" (Measures frequency and urgency of the pain).
• "Why is it important for your company to solve this?" (Evaluates business value and budget potential).
• "What do you currently do to solve this problem for yourself?" (Identifies actual incumbent competitors, such as Excel or manual scripts).

Low-Signal / Dangerous Questions (Avoid These)

• ❌ "Would you use our product?" (Users will politely say yes, but their affirmative answer means nothing).
• ❌ "Which features would make product X better?" (Delegating product design to the user; their job is to report problems, your job is to invent solutions).
• ❌ Yes/No Questions (Cut off detailed storytelling and fail to yield concrete examples).
• ❌ "How would a better product look like to you?" (Assumes the user is a trained software designer or engineer).
• ❌ Asking Two Questions at Once (Confuses the interviewee and dilutes their answers).

7. Synthesizing Interviews and Evaluating Problem Value

Once you complete 5 to 10 user interviews, compile your notes systematically to extract objective takeaways rather than emotional biases.

The Value Evaluation Framework

Before writing a single line of code, evaluate whether the problem you are solving is genuinely valuable using three criteria:

1. Existing Spend: Are people already paying money for mediocre solutions in this space today (e.g., paying human consultants to generate PDF reports)?
2. Incumbent Competitors: Are users currently hacking together a solution using spreadsheets or Google Docs? Moving someone off Excel requires a dramatic, 10x better user experience.
3. Sales Friction: Consider how hard it is to sell to your target audience. For instance, traditional industries like plumbing or construction notoriously resist software changes, whereas early-stage startups eagerly try new tools.

8. Transitioning Conclusions into an MVP and Prototype Testing

Once you have formed a solid hypothesis based on user interviews, build a lightweight Minimum Viable Product (MVP)—such as an interactive design prototype in Figma or InVision—and put it directly back in front of your users.

The Silent Testing Technique

• Hand Over Control: When testing a prototype, hand your phone or laptop directly to the user. Do not touch the device for them.
• Give a Goal, Not Instructions: Tell the user what outcome to achieve (e.g., "Try to complete a booking on this screen"), but do not narrate how to click through it.
• Think-Aloud Protocol: Ask users to verbalize their inner monologue as they navigate the prototype. Pay close attention to which UI labels and terminology confuse them.

9. Maintaining Lifelong Customer Loops

The best founders do not stop talking to users after the initial launch phase. They maintain permanent, high-touch feedback loops.

• Exclusive Communities: Create an exclusive Slack workspace or WhatsApp group for your earliest users. Make them feel special by granting them front-row access to a world-changing product in development.
• Rapid Iteration Trust: When users suggest a valid problem and you ship a fix or new screen days later, it builds immense customer trust and loyalty.
• Peer Connection: Enable your users to connect with other professionals experiencing the exact same operational hurdles—something they rarely get to do outside of your ecosystem.

Summary and Key Takeaways

• Continuous Engagement: The best founders talk to customers throughout the entire lifetime of their company, treating user feedback as a perpetual operating system.
• Reject Hollywood Myths: Startup ideas do not come from solitary late-night coding sessions; they are born from intense, two-way conversations with real people.
• Emulate Airbnb's Early Hustle: Legendary experiments—like Brian Chesky living in 50 Airbnbs—prove that personal connection beats anonymous growth hacking every time.
• Source Beyond Echo Chambers: Reach out to cold users via LinkedIn, niche communities, and professional networks without fear of rejection.
• Focus on Problems, Not Solutions: During interviews, your job is solely to listen to the user's pain points; never pitch your product idea prematurely.
• Observe True Behavior: Watch what users actually do rather than what they say, asking them to walk through current manual workflows and spreadsheets.
• Avoid Leading Traps: Never ask users "Would you use this?" or "What features do you want?"—they will say yes to everything while providing zero validation.
• Evaluate Economic Value: Check if customers are already paying for imperfect workarounds or human consultants, proving that a budget exists for a better solution.
• Silent Prototype Testing: When testing MVPs, hand the device to the user, give them a goal, and stay silent while observing their confusion and friction points.
• Build Lifelong Loops: Keep early users engaged via exclusive WhatsApp or Slack groups, shipping rapid fixes to build unbreakable trust and product loyalty.