How to Talk to Customers When You're Shy or Nervous
If talking to strangers scares you, don’t force yourself into a big scary conversation — start with the lowest-stakes version (a text, a DM, a written question) to people who are one step removed from you, use a script so you never have to improvise, and treat each contact as a tiny experiment instead of a performance. Shyness doesn’t disqualify you from building a company. It just means you need a system that removes the parts you dread, and this post gives you one.
Customer discovery is the part of validation where you talk to real people to find out if your idea solves a problem they actually have. It’s the single most important skill in the Validate sprint, and it’s also the one that makes shy founders want to skip straight to building. That’s exactly the mistake that kills startups: you build for months, then discover nobody wanted it. So let’s make the talking part survivable.
Why does this feel so hard when you’re shy?
When you’re nervous, your brain treats a customer conversation like a test you might fail in public. It isn’t. Reframe it and most of the fear drains out.
You are not selling. You are not pitching. You are not on trial. You are a researcher collecting information about how someone lives. The other person is doing almost all the talking. Your only job is to ask a question and shut up.
Here’s the reframe that changes everything: the goal of an interview is to learn the truth, not to be liked. When you stop trying to make a good impression and start trying to fill in blanks about the other person’s life, the pressure lifts. You’re not the show. They are.
One more thing that helps: nobody can see your nerves as clearly as you feel them. The shaky voice in your head is not broadcasting. On a text or DM, it’s completely invisible.
Start with the lowest-stakes channel first
You do not have to walk up to a stranger. There’s a ladder of scariness, and you should climb it from the bottom. Each rung is easier than a face-to-face cold conversation, and every rung still gives you real answers.
| Channel | How scary | Why it’s great when you’re shy |
|---|---|---|
| Written survey / form | Very low | Zero real-time talking; async; you edit before sending |
| Text to someone you know | Low | Familiar person, one message, no eye contact |
| DM to a one-step-removed contact | Low-medium | You can draft it, no live pressure, easy to ignore a non-reply |
| Voice or video call | Medium | Live, but scheduled and one-on-one, no crowd |
| In-person, someone you know | Medium-high | Warm, but real-time |
| In-person cold (stranger) | High | Save this for last, or skip it entirely |
Most shy founders think discovery means the bottom of that list is off-limits and they have to do the scary in-person stuff. Wrong. You can validate an idea almost entirely through DMs, texts, and short calls. If you literally cannot approach strangers yet, we wrote a whole guide on validating an idea when you can’t talk to strangers yet that works around that constraint.
Write a script so you never have to improvise
The reason live conversation feels terrifying is the fear of freezing — not knowing what to say next. A script deletes that fear. You are allowed to read from notes. Nobody will know.
Good discovery questions ask about the past and the real, not the hypothetical. This idea comes from a short book called The Mom Test, and it’s worth learning cold before your first call. Here’s a script you can literally paste and adapt.
The DM / text opener (12 seconds to write, use verbatim):
Hey [name] — I’m a high school student researching how people deal with [problem]. Not selling anything, promise. Could I ask you 3 quick questions? Totally fine if not.
The call script (read it off your screen):
- “Thanks for the time. This is research, not a sales thing, so just be honest — you can’t hurt my feelings.”
- “Tell me about the last time you dealt with [problem]. Walk me through what happened.”
- “What did you do about it? Did you try anything to fix it?”
- “What was annoying about that? What did it cost you — time, money, stress?”
- “Has this come up recently, or is it kind of rare?”
- “Who else do you know who deals with this? Could you intro me?”
Notice what’s missing: you never describe your idea. That’s on purpose. The second you say “I’m building an app that…”, polite people start complimenting you, and compliments are useless. For the deeper version of this, read our beginner’s guide to customer interviews and the full breakdown of the Mom Test for teen founders.
What if I freeze or they say no?
They will sometimes not reply. That is not rejection — it’s the default state of the internet. If you send ten DMs and three answer, you’re doing great. Plan for silence so it stops stinging.
If you freeze mid-call, you have three escape hatches, and all of them are normal:
- Buy time out loud. “Give me one sec, I want to ask this right.” Silence feels like an hour to you and a blink to them.
- Read the next scripted question. You wrote it exactly so you’d never have to invent one on the spot.
- End early and gracefully. “That’s super helpful, I don’t want to take more of your time — thank you.” A short interview that happened beats a long one you avoided.
And if someone says no to being interviewed? “No worries, thanks anyway.” Move to the next name. You are looking for the small percentage of people who want to help a student. They exist, and they say yes more than you’d bet.
A one-week plan for the nervous founder
You don’t need to do this all at once. Spread it across the Validate week so each day has one small, defined action.
- Day 1 — List 15 names. People who plausibly have the problem: a teammate’s older sibling, a teacher, a coach, someone in a Discord about the topic. If you’re stuck on where these people are, use our guide on where to find people to interview.
- Day 2 — Send 5 written messages. Use the opener above. No calls yet. Just typing.
- Day 3 — Do one short call. Someone you already know. Read the script. Fifteen minutes, one person, no crowd.
- Day 4 — Send 5 more messages and book calls with anyone who replied.
- Day 5 — Do two more calls and write down what you heard, in their words, not yours.
Five to ten real conversations will teach you more than a survey of five hundred, because you can follow up and dig. Once you’ve done a handful, you’ll want to know how many customer interviews are actually enough so you know when to stop and start building.
The truth about doing this scared
Confidence is not a prerequisite. It’s a side effect. Nobody’s first customer conversation is smooth, and shy founders often run better interviews than loud ones — because they’re comfortable being quiet, and being quiet is how you let the other person tell you the truth. Your instinct to listen more than you talk is an advantage here, not a weakness.
You will be awkward on the first call. You’ll be fine on the fifth. By the tenth you’ll catch yourself enjoying it, which is a strange and specific feeling worth chasing.
This is exactly the muscle you build in the Validate sprint at batch0’s accelerator — with a script, a cohort doing it alongside you, and a coach who’s watched a hundred nervous founders make their first call. If you want a structured push to actually do the scary thing instead of researching it forever, applying is free and you only pay if you get in. But you don’t need us to start. Open your messages and send one today.