Customer Research With No Users Yet: A Playbook

If you have no users yet, customer research becomes a sourcing problem: define the exact buyer you need, then reach verified strangers who match that profile through communities, cold outreach, referrals, or a research panel. You do not need an audience, a waitlist, or traffic to talk to the right people. You need a clear target and a channel that puts you in front of them fast.

This is the part most pre-launch founders get stuck on. Every discovery guide tells you to talk to customers, but you have no customers. Your network is friends and former colleagues who are too polite to tell you the truth. The real skill at stage zero is not running the interview. It is finding people who genuinely match your target and are willing to be honest with a stranger.

This playbook walks through the whole sequence: how to define who you actually need, where those people are, how to recruit them cold, how to screen out false matches, and what to ask so you avoid the politeness bias that produces fake validation. If you want the interview craft itself, pair this with our customer discovery interviews founder’s guide. This post is about the harder half: getting the right strangers in the room.

Start with the buyer, not the product

Before you look for anyone, write down who you are looking for in specific terms. Vague targets produce vague research. “Small business owners” is not a target. “Owners of 5 to 20 person marketing agencies who currently manage client reporting in spreadsheets” is a target you can screen for.

Get concrete on three things:

Role and context. What is their job title, company size, and industry? What are they responsible for day to day?
The problem behavior. What do they do today that your product would replace or improve? If they do nothing about the problem, they may not feel it strongly enough to buy.
The trigger. What event makes this problem urgent? A new hire, a lost client, a compliance deadline?

This becomes your screener. It is also the seed of your first buyer persona, though at this stage the persona is a hypothesis to test, not a conclusion. If you want structure here, our guide to buyer personas covers how to build one without inventing details.

Writing this down does two jobs. It tells you exactly who to recruit, and it stops you from counting the wrong conversations as validation later.

Where the right strangers actually are

With no list and no traffic, you have to go where your target buyer already spends time. Different channels trade off speed, fit, and cost. Here is how the realistic options compare for a zero-audience founder.

Channel	Speed to first talk	Fit control	Cost	Best for
Your personal network	Fast	Low	Free	Warm-up reps, not real signal
LinkedIn cold outreach	Slow	High	Free (time-heavy)	Named-account B2B targets
Niche communities (Slack, Reddit, forums)	Medium	Medium	Free	Understanding language and pain
In-person events and meetups	Slow	Medium	Travel cost	Deep conversations, hard to scale
Referrals from early interviews	Medium	High	Free	Reaching a specific role via snowball
Verified research panel	Fast	High	Paid per participant	Hard-to-reach or B2B buyers, fast

None of these is wrong. The right mix depends on how reachable your buyer is. If you are building for other founders or marketers, communities and LinkedIn can work with patience. If you are building for CFOs, compliance leads, or IT decision-makers, cold outreach conversion is brutal and you will burn weeks. That is exactly where a panel earns its cost.

Communities: listen before you ask

Before you post anything, read. Niche subreddits, industry Slack groups, and specialist forums are where your buyers describe their problems in their own words. That language is gold for your positioning later. When you do engage, ask for a short call to learn, not to pitch. People help founders who are clearly researching and not selling.

The limit is fit control. You cannot verify that the person replying is actually the buyer they claim to be, and community norms often ban recruiting outright. Treat communities as a place to learn vocabulary and find a few willing talkers, not as your primary sample.

Cold outreach: high fit, high effort

LinkedIn and email let you target exact roles at exact companies. That precision is the upside. The downside is volume. Reply rates for cold research invites are low, and every yes still needs scheduling, screening, and a no-show buffer. If your target is senior, this gets harder, which is why we wrote a dedicated guide on recruiting C-level executives for research.

Cold outreach works, but budget for the math. To run eight interviews you may need to contact a hundred or more people. For a founder with limited runway, weeks of outreach is often the real cost.

Referrals: the snowball

At the end of every good interview, ask: “Who else do you know who deals with this?” One well-matched participant can introduce you to two or three more. This is the cheapest way to reach a tight niche, and the introductions arrive pre-warmed. The catch is that it only starts working after you have your first few conversations, so you still need a way to seed the snowball.

When a panel is the fastest path

The uncomfortable truth for pre-launch founders is that sourcing verified strangers who match your target is slow and unreliable through free channels. This is the specific problem a research panel solves. A panel is a pool of people you can screen and recruit on demand, without a list of your own.

CleverX is a B2B research platform built for exactly this. Its participant pool is an 8M+ panel of professionals verified by work email and LinkedIn, plus B2C consumers, across 150+ countries. You define the role, seniority, industry, and behavior you need, and matched participants are typically delivered in about 2 to 5 days. For a founder who has no way to reach a VP of engineering or a hospital procurement lead, that verification and speed is often the only realistic route to the right buyer.

The reason verification matters so much at stage zero is false positives, which we get to below. When you recruit from open channels, you cannot always confirm that the person is who they say they are. A verified panel closes that gap before the interview starts. For a deeper comparison of sourcing options, see our roundup of B2B participant panels and the broader guide to finding quality research participants.

Paying for participants is not a shortcut that weakens your research. It removes the strongest bias in early-stage discovery: only hearing from people already in your orbit, who are the least representative and the most polite.

Screen for genuine fit, not eagerness

However you source people, the screener is what protects your research. The failure mode is recruiting someone who says they match because they want the incentive or want to be helpful. Screen on behavior and verifiable facts, not opinions or intentions.

Good screener questions:

Behavioral. “In the last three months, how many times did you do X?” Ranges filter out people who do not actually have the problem.
Role and responsibility. “Which of these are you personally responsible for?” with a list that includes decoys.
Verifiable. Confirm job title and company through work email or LinkedIn rather than self-report.

Avoid leading screeners that telegraph the right answer. If you ask “Do you struggle with client reporting?” everyone struggling to earn the incentive will say yes. Ask “How do you handle client reporting today?” and let their answer reveal whether they are a real match.

When you interview two different buyer types, screen and analyze them separately. Mixing a founder and an enterprise buyer into one sample hides the patterns in each. If segments are central to your product, our guide to recruiting for a segmentation study goes deeper.

What to ask so you avoid false positives

Sourcing the right people is half the battle. The other half is not corrupting their answers with the way you ask. Two biases wreck early interviews: politeness bias, where people tell you what you want to hear, and confirmation bias, where you hear what you want to hear.

The antidote is to ask about the past, not the future. Future questions invite speculation and flattery. Past questions surface evidence.

Instead of “Would you use a tool that does X?”, ask “Walk me through the last time you dealt with X.”
Instead of “Is this a big problem for you?”, ask “What did you do the last time this came up, and how long did it take?”
Instead of “Would you pay for this?”, ask “What are you spending on this today, in tools or time or workarounds?”

Notice that none of these mention your idea. In discovery, the moment you describe your solution, the conversation shifts from evidence to opinion. Keep your idea out of the room for as long as you can. When you do test willingness to pay or demand, do it deliberately as a separate step, which is what our pre-launch demand testing with real buyers guide is built for.

A few interview habits that keep answers honest:

Embrace silence. After they finish, wait. The next sentence is usually the real one.
Chase stories, not summaries. “Tell me about a specific time” beats “In general, how do you…”
Never sell. You are learning, not convincing.

For question banks you can adapt, see our 50 user interview questions that uncover real insights, and study the common interview mistakes so you can spot yourself making them. The Nielsen Norman Group has a well-known primer on interviewing users that reinforces the same behavioral principle.

Sequence discovery before you build

Order matters as much as method. A common mistake is building a landing page or a prototype first, then interviewing people about it. That measures reaction to your assumptions instead of testing whether the assumptions hold. Run the sequence in this order:

Problem discovery. Confirm the problem is real, frequent, and painful enough that people already spend time or money on it. No solution talk yet.
Solution shaping. Once the problem is validated, explore how people would want it solved and what they use today.
Demand and pricing. Only after the first two, test whether they would actually pay, and how much.

Most zero-audience founders should live in step one for longer than feels comfortable. If the problem is not real, nothing downstream matters. When you are ready to move into demand and pricing, the best product validation tools for startups roundup covers what to reach for next.

If discovery is going to be an ongoing part of how you build, not a one-time sprint, it helps to keep a light continuous interview program running so you always have fresh signal.

A realistic first-month plan

Here is what a focused four weeks looks like with no users and no audience:

Week 1. Write your buyer definition and screener. Read your target’s communities to learn their language. Line up your first five conversations from your network as practice reps.
Week 2. Start real sourcing. Post a recruit-to-learn message where allowed, send cold outreach, and open a panel study so verified matches arrive while your cold outreach percolates. On CleverX you can also run moderated conversations at scale with AI Interview Agents when your own calendar is the bottleneck.
Week 3. Run five to eight interviews per segment. Ask about the past. Take verbatim notes. Ask every participant for one referral.
Week 4. Look for patterns across interviews, not single quotes. If the same problem, trigger, and workaround show up repeatedly, you have signal worth building on.

Five to eight strong interviews per distinct buyer surface most recurring patterns. You are not running a statistical study. You are looking for evidence that a specific problem is real for a specific person, told in their own words.

The founders who get this right treat “no users” as a temporary sourcing constraint, not a reason to guess. Define the buyer precisely, reach verified strangers who match, screen on behavior, ask about the past, and sequence discovery before you build. When you are ready to reach people you could never find on your own, you can start recruiting verified participants on CleverX and have matched buyers in a few days.

Frequently asked questions

How do I do customer research if I have no users, no list, and no audience?

Treat it as a sourcing problem. Define the exact buyer you need, then reach verified strangers who match through cold outreach, communities, referrals, or a research panel. A panel is usually the fastest path because it delivers people screened to your criteria in a few days instead of weeks.

How many people do I need to interview before I have no users?

For early discovery, five to eight interviews per distinct buyer segment surface most of the recurring patterns. If you serve two very different buyers, plan for five to eight in each group rather than mixing them together.

How do I make sure the strangers I recruit are actually my target buyer?

Screen on behavior and role, not opinions. Ask what they did, what they bought, and what they are responsible for. Verify job title and company through work email or LinkedIn so you are not interviewing someone who simply clicked to earn an incentive.

How do I avoid politeness bias and false positives in early interviews?

Ask about past behavior instead of future intentions. Questions like would you use this invite polite yes answers. Questions like what did you do the last time this happened surface real evidence. Keep quiet, let silences run, and dig into stories rather than pitching.

Should I build a landing page or talk to people first?

Talk to people first. A landing page measures interest in a message you have not validated yet. Discovery interviews tell you whether the problem is real and worth paying to solve, which is what you need before you write a single line of copy or code.

Is it cheating to pay strangers to talk to me instead of finding organic users?

No. Paying for a participant’s time is standard research practice and it removes the bias that comes from only hearing from people already in your orbit. What matters is that participants match your target and answer honestly, not that they found you for free.