Skip to content
Zero2OneZero2One

Customer Interviews & Research

Talk to the people you're building for — and watch how they work today — before you decide what to build. The cheapest way to avoid making something nobody needs.

Cost
Free
Time to results
Days
Difficulty
Medium
Best for
B2B
0/7

Proof first

6 companies already did this.

Two of these examples come from Paul Graham's essay "Do Things That Don't Scale" — see sources.

Deep dives

Two case studies, start to finish.

Intuit — follow me home

Intuit · 1980s

Observation · "Follow me home"

What they built: watching a customer install and use Quicken

What they didScott Cook would go to software stores and, when someone bought Quicken, ask if he could follow them home and watch them use it. Intuit still runs "Follow Me Home" visits today.

Why it workedPeople can't tell you where they get stuck — they can only show you. Watching real setups on real computers exposed problems no survey or focus group would have caught.

Copy thisDon't ask "would you use this?". Ask to watch someone do the job today, on their own setup.

Superhuman — interview the people who'd miss you

Superhuman · PRE-LAUNCH

PMF survey · 22% → 58%

What they built: 4-question survey + follow-up interviews

What they didRahul Vohra asked users "How would you feel if you could no longer use Superhuman?" and tracked the share who answered "very disappointed". Only 22% did. The team then studied who those users were and what they loved, and what held everyone else back.

Why it workedFocusing on the users who already loved it lifted the score to 33% straight away. Building for them raised it to 58% within three quarters. The answers told the team what to double down on and what to fix.

Copy thisRun the "very disappointed" survey. Then interview the people who said it — they describe your real market.

The guide

7 steps, each anchored to a real example.

Tick each step off as you go. Your progress is saved in this browser.

Step 1 · ~1 hour

Pick one kind of person and three questions

Choose one specific type of person — "freelance designers billing 5+ clients" — and the three things you most need to learn. Vague targets give vague answers.

NotesThe Mom Test

How Pinterest did this step

Illustration

Ben Silbermann noticed many early Pinterest users were interested in design, so he went to a design bloggers' conference to recruit more of exactly them.

Source [6] ↓

Step 2 · ~2–3 days

Book 15 conversations in a week

Ask friends for introductions, post in the communities your people use, reply to people complaining about the problem online. Ask for 20 minutes — not a pitch.

LinkedInRedditSlack communities

How Slack did this step

"We begged and cajoled our friends at other companies to try it out and give us feedback."

Exact quote — see source

Slack's first testers came from friends at other companies — about six to ten of them to start with.

Source [4] ↓

Step 3 · ~30 min each

Ask about the past, not the future

"Tell me about the last time you…" beats "Would you use…?". Ask what they did, what it cost them, what they tried. Don't pitch, and keep your idea out of it until the end.

ZoomThe Mom Test questions

How Segment did this step

#CHANGELOG

Illustration

Segment's founders went through several products before talking with users pointed them to the one developers actually wanted.

Source [5] ↓

Step 4 · ~1 hour each

Watch them do the job

Ask to screen-share or sit with them while they do the task today. Write down every workaround, spreadsheet and sigh — that's your product spec.

ZoomLoomNotebook

How Intuit did this step

Illustration

Scott Cook followed new Quicken buyers home to watch them set it up — a practice Intuit still calls "Follow Me Home".

Source [1] ↓

Step 5 · ~Days

Go where they are

If your users cluster in one city, industry or event, go there. In person you see the context a call misses — and you find your next five interviewees.

CalendarTrain ticket

How Airbnb did this step

Illustration

Airbnb's founders went door to door in New York, recruiting new hosts and helping existing ones improve their listings.

Source [6] ↓

Step 6 · ~1 hour

Measure pull with one question

Once people use an early version, ask how they'd feel if they could no longer use it, and track the share who say "very disappointed". 40% is the widely used bar.

TypeformGoogle Forms

How Superhuman did this step

"How would you feel if you could no longer use Superhuman?"

Exact quote — see source

Superhuman's first score was 22%. Focusing on the users who loved it, and building for them, took it to 58% within three quarters.

Source [3] ↓

Step 7 · ~Half a day

Turn notes into decisions

After every five conversations, write the top three patterns and one decision — "cut X", "build Y first". Research that doesn't change a decision was just a chat.

NotionFigJamSpreadsheet

How Slack did this step

#CHANGELOG

Illustration

Slack learned quickly that the product worked very differently as team size grew — feedback from a much bigger tester, Rdio, changed what they built next.

Source [4] ↓

Before you start

Pros & cons.

Works in your favour

  • Free and fast — 15 conversations in a week
  • Stops you building the wrong thing
  • Gives you customers' exact words for your landing page
  • Your first interviewees often become your first users
  • Useful at every stage, not just the start

Go in with eyes open

  • People are polite — they'll say they love it and never buy
  • Easy to lead the witness without noticing
  • Small samples mislead; look for patterns, not one great quote
  • Scheduling takes time, so founders skip it

Make it yours

Three ways to run it.

The Intuit model

Observation · Follow me home

Watch the job being done in its real setting. Best for workflow tools and B2B software.

The Superhuman model

Survey + interviews · 22% → 58%

Survey for the "very disappointed" users, then interview them. Best once you have early users.

The Airbnb model

Go to them · Door to door

Travel to where your users cluster. Best for marketplaces and local businesses.

Copy and paste

Templates & resources.

Interview script

modeled on The Mom Test

Intro (1 min): I'm researching how {who} handle {job}. No pitch — I just want to learn.

1. Tell me about the last time you {did the job}.
2. What was the hardest part?
3. Why was that hard?
4. What have you tried to fix it? What did that cost?
5. What don't you love about those fixes?
6. Who else should I talk to?

Never ask: "Would you use…?" or "Would you pay…?"

Outreach message

modeled on Slack's friends-first testers

Hi {name} — I'm researching how {role}s handle {problem}. I'm not selling anything.

Could I ask you about it for 20 minutes this week? I'm talking to 15 people and happy to share what I learn.

Product-market fit survey

the four questions from Superhuman's engine

1. How would you feel if you could no longer use {product}?
   Very disappointed / Somewhat disappointed / Not disappointed
2. What type of people do you think would most benefit from {product}?
3. What is the main benefit you receive from {product}?
4. How can we improve {product} for you?

From the showcase

Be the first founder featured here.

The famous examples above are proof it works. The next founder reading this wants proof it works for someone like them — that could be you.

Used “Customer Interviews & Research”? Submit your product →

Discussion

Tried this? Tell us how it went.

Share a result, ask a question, or say what you'd do differently. The next founder to try this will thank you.

Loading the discussion…

NOW MAKE IT YOURS

You've seen how 6 companies did it. Your turn.

We're building an AI planner that turns this playbook into a step-by-step plan for your product. Leave your email for early access — one message when it opens, nothing else.

Already using this tactic? Show other founders what you built. Get featured →

Receipts

Sources & references.

Last reviewed

  1. 1.
    Intuit's "follow me home" program (opens in a new tab)

    Business Insider, 2015

  2. 2.
    Scott Cook and the first "Follow Me Homes" (opens in a new tab)

    Boxes and Arrows

  3. 3.
    How Superhuman built an engine to find product-market fit (opens in a new tab)

    Rahul Vohra, First Round Review

  4. 4.
    Slack's launch strategy and first test companies (opens in a new tab)

    Stewart Butterfield, First Round Review

  5. 5.
    How to find product-market fit (opens in a new tab)

    Peter Reinhardt (Segment), Y Combinator

  6. 6.
    Do Things That Don't Scale (Airbnb, Pinterest) (opens in a new tab)

    Paul Graham, 2013

  7. 7.
    The Mom Test — how to talk to customers (opens in a new tab)

    Rob Fitzpatrick

Company names and logos are trademarks of their owners, shown only to identify the companies discussed. Spot an error? Tell us below — we correct and date every change.

Corrections

No corrections so far.

Spot an error?

Tell us what's wrong — we fix it and log the change here.

Keep going

Related playbooks.

Landing Page + Smoke Test

$ · Days · Easy — test what you heard with real clicks

Read playbook →

Beta / Early-Access Program

Free · Weeks · Medium — keep learning with real users

Read playbook →

Build an Email List / Waitlist

Free–$ · Weeks · Easy — collect the people you spoke to

Read playbook →