When to use this: Before you build anything, when you’re pre-PMF, or when you keep launching to crickets. OR, when your marketing is falling flat even after you have 10+ paying customers and have interviewed them.
Honestly, if you haven’t interviewed them, doing this practice is just hypothetical. Not necessarily a waste of time, but also, you’re just delaying the Customer Development you should be doing.
The Problem
You’re building for “small businesses” or “busy professionals” or “tech founders.” That’s everyone and no one. Your landing page sounds generic. Your positioning doesn’t land. You talk to someone about your product and they say “interesting” but never buy.
The issue isn’t that you need better marketing copy. You don’t know who you’re building for at a level that changes what you build and how you talk about it.
What “Customer Avatar” Means
A customer avatar is a reference document that captures who experiences the problem you’re solving most acutely, what their life looks like, and how your solution fits into it.
This isn’t demographic data (age 30-45, lives in cities, makes $75k). That’s useless for product decisions. Demographics don’t predict whether someone will pay for your product.
You need to know:
- What triggers them to go looking for a solution RIGHT NOW
- What they’ve already tried that failed
- What they’re currently doing to cope (their workaround)
- What would have to be true for them to switch to your thing
- Where they spend time, who they listen to, what they trust
If you can’t answer these, every product decision is a guess.
Mistakes to Avoid
Creating the avatar first
You sit down, imagine your perfect customer, write down everything about them, then go looking for people who match. That’s building before validation. You’re creating confirmation bias.
Instead, do some interviews first.
Asking people if your idea is good
“Would you use this?” and “Do you think this would work?” are worthless questions. Everyone lies to you because they’re being polite. Your mom will say it’s great. Your friends will encourage you. Strangers will say “interesting idea” because they don’t want to hurt your feelings.
This is basically the biggest violation of The Mom Test. Read the linked article to see other ways of framing this same type of question that lead to bad results and wasted time.
Stopping at demographics
Knowing someone is 32, works in tech, and lives in Austin tells you nothing about whether they’ll pay $50/month for your product. You need to understand their problem and what solving it is worth to them.
Even more, you need to know WHEN they need that problem solved. Arguably, the “when” question is more important than just about anything else when trying to understand your avatar. When does this type of person need your product?
It’s hard to sell ice cream in the winter. It’s much easier to sell to that same person in the summer.
Building the avatar solo in a vacuum
You write it yourself based on assumptions, never validate it with real humans, then wonder why your positioning doesn’t resonate. You’re building a fiction.
It’s fine to do that, but only if you tag it as a “hypothesis” and then go out to prove/disprove it with short small testing. Again, do interviews to test it.
Talk First, Document Second
Rob Fitzpatrick’s Mom Test teaches you to ask questions even your mom can’t lie to you about. The trick: don’t ask about your idea. Ask about their life.
3 Rules for Conversations
-
Talk about their life, not your idea
Don’t pitch. Don’t explain what you’re building. Ask about how they currently handle the problem.
-
Ask about specifics in the past, not hypotheticals
Not “would you pay for this?” (speculative) but “what did you do the last time you hit this problem (factual)? How long did it take (factual)? Did you pay for anything to help (factual)?”
Phrasing things like this takes practice. Listen to your own interviews and note where you can get better. This is a muscle and you have to exercise it. So, practice lots and you’ll get better at it.
-
Talk less, listen more
Your job is to understand their world. If you’re doing most of the talking, you’re doing it wrong.
Read The Mom Test for more details.
Questions
Examples:
“Walk me through the last time you dealt with {this problem}”
“What have you tried to fix it?” (Then dig: why didn’t that work? What was missing? Why did you stop using it?)
“How are you handling it now?” (Their workaround tells you how urgent the problem is. If they have an acceptable workaround, your product needs to be 10x better, not 10% better.)
“If you could wave a magic wand and fix this, what would change?” (This reveals what they actually care about, not what you think they should care about.)
“Who else has this problem?” (Helps you find more people to talk to and tests if they’re really in a community of people with this pain.)
The Process
1) Talk to 10-15 people (minimum)
You need enough conversations to see patterns. Three people don’t show you patterns. They show you three individuals.
Who to talk to: People who might have the problem you’re solving. Not people who might buy your product someday. People actively dealing with this right now.
Where to find them: Slack communities, Discord servers, Reddit, LinkedIn groups, Twitter, niche forums, anywhere people gather who have this problem.
Cold outreach template:
“I’m researching {problem space} and would love to learn about how you currently handle it. 30 minutes? Happy to share what I learn.”
Pro tip
You’re not selling. You’re learning. Make that clear.
2) Document what you hear
After each conversation, write down:
- Exact phrases they used to describe the problem (verbatim quotes matter - this is how they talk about it, not how you talk about it)
- What they’ve tried and why it failed
- Their current workaround and how much time/money it costs them
- What finally pushed them to try fixing this problem
- Other tools they’re paying for in this space
- Who they follow, what content they consume, where they learn about solutions
- What words and phrases they use when they get emotional about the problem
- Words they use to describe themselves
Don’t interpret yet. Just capture.
3) Summarize post-interview
Quickly summarize and add these to your record:
- What surprised you
- What you were right about
- What you were wrong about
- What you didn’t know
- What you want to learn more about
- Any new ideas
4) Synthesize after 10+ interviews
Now you analyze and synthesize. What keeps coming up? Any patterns yet?
- 4 out of 10 people mentioned the same failed solution
- 7 out of 10 described the problem using nearly identical language
- Everyone’s workaround involves spreadsheets + manual work
- All of them went looking for a solution after a specific triggering event (deadline, client complaint, personal breaking point)
These patterns are your signal. This is what the avatar captures.
5) Create the reference document
Once you’ve identified patterns, document everything in a format you can reference during product decisions.
Fill this out based on patterns from your conversations, not assumptions. Use exact quotes where relevant and possible
Here’s what to capture:
Demographics (but don’t stop here):
- Age range, gender, income level, education, location, marriage status, work status
Psychographics (this is where it matters):
- Fears: What keeps them up at night related to this problem?
- Goals: What are they trying to achieve?
- Content they consume: What newsletters, YouTube channels, podcasts, blogs do they read/watch/listen to?
- Hobbies: What do they do for fun?
- Who they subscribe to: What creators, experts, companies do they follow?
- Offline buying behavior: What do they spend money on? How do they make purchase decisions?
- Beliefs & values: What do they care about? What principles guide their decisions?
- Online hangouts: What communities, forums, Slack groups, Discord servers are they active in?
- Habits: What’s their daily routine? What tools do they use every day?
- Motivators: What drives them to take action?
- Where they spend time: Both online and offline
- What they do on weekends: How do they recharge?
- What they wish they were doing: The gap between current state and desired state
- Who they want to be: Their aspirational identity
Additionally…
Current State:
What are they doing now? What sucks about it? What have they tried? Include quotes - use their words, not yours.
Trigger Event:
What makes them finally go looking for a solution instead of living with the problem? This is critical. If there’s no trigger, there’s no urgency. No urgency = no sales.
Desired State:
Not some fantasy utopia. What specific outcome would make them feel like the problem is solved? What does “better” look like to them?
Barriers:
What’s stopping them from getting there? Is it time, money, knowledge, existing tools, confidence, something else? Your product needs to remove or reduce this barrier.
Information Sources:
Where do they learn about solutions? Who do they trust? What do they read/watch/listen to? This tells you where to find more people like them and how to reach them.
Company Killer Assumptions:
Based on these conversations, what assumptions - if wrong - would kill this business? Example: If nobody’s willing to pay more than 50/month to work, that’s a killer assumption you need to test early.
Give this person a name. Make them real. When you’re deciding what to build next, ask: “Would Alexis care about this feature?“
6) Segment & converse
Now, once you have your avatar, go back to your interviewee list. If you can contact them again, do it. Thank them for the time they spent with you previously, and then ask them the next riskiest assumption question on your list.
Doing this gives you two benefits:
- You keep in contact with someone who you might be able to sell to in the future. You now have warm leads. These conversations lead to trust. Trust leads to sales.
- These people could become your ICP, and you likely have more assumptions than what you asked about in your interview. This gives you the ability to keep the conversation going and get even more information from them where as otherwise you’d have to find another person to interview.
Tip
Specifically and concisely describe your ideal avatar. Write it down.
Then, thank anyone who doesn’t fit your avatar/ICP and ask them if they know anyone who is {avatar description}.
Otherwise, don’t waste any more time with those that don’t fit your avatar. Their advice and suggestions will be counter-productive as it pertains to understanding and solving the problem of your avatar.
When Customer Avatars Don’t Help
You talked to 10 people and got wildly different answers
Your segment is too broad. Narrow it. Instead of “small business owners,” try “solo service providers with 0-2 employees who schedule client appointments manually.” If it’s still too broad, narrow more. Go through Self-Selection process to get more specific here.
Everyone says the problem isn’t urgent
They have acceptable workarounds or don’t care enough to pay. This is good information - you just learned not to build this. Pivot or kill the idea before you waste six months building.
You can’t find 10 people to talk to
This is your distribution problem showing up early. If you can’t find people to interview, you won’t find people to sell to later. Either you’re looking in the wrong places or the market is too small. Figure out distribution first.
Warning
If you can’t find 10 people to talk to, DO NOT MOVE FORWARD. The reason we go through this process is precisely this. Notice that if you DO find 10 people to talk to, you’ve not only found good information, but also a potential source of new customers.
However, if you can’t find 10 people to talk to in one place, you likely also don’t have an early distribution channel figured out. And THAT, my friend, is a company killer assumption.
You must figure out company killer assumptions first, before moving on to anything else. Otherwise, you’re wasting your time on this business if it’s wrong or can’t be worked around.
Red Flags for Wrong Avatar
You’re getting “interesting” responses but no one’s buying. This means the problem isn’t urgent enough or your solution doesn’t remove the barrier.
People say “I’ll think about it” and then ghost. They’re being polite. The problem isn’t painful enough for them to act.
Your positioning sounds generic. If you could replace your company name with a competitor’s and the copy still works, you don’t understand your customer well enough.
You’re constantly explaining what you do. If it takes more than two sentences and they’re still confused, your avatar is based on what you think they need, not what they told you they need.
Feature requests are all over the place with no pattern. This means you don’t have a clear avatar. You’re talking to too many different types of people with different problems.
Anti-Avatar: Who to Avoid
Equally important: document who you should NOT build for.
From your conversations, you’ll notice some people don’t fit:
- The problem isn’t urgent for them (they shrug it off)
- They’re price-sensitive and would never pay enough to make your economics work
- Their workflow is so different that serving them requires totally different features
- They’re comparison shoppers, not problem solvers (they’ll churn the moment something cheaper appears)
Write these down too. When someone like this reaches out, you know to politely decline instead of wasting any time trying to make them happy.
Read Self-Selection for more details on this.
What Changes with an Avatar
Product decisions get easier
Feature request comes in. Check it against your avatar. Would Alex care? Does it solve the problem you documented? If not, say no.
Positioning gets clearer
You use their language from the conversations. Your landing page doesn’t sound like a SaaS template anymore. It sounds like them talking to themselves.
Distribution gets obvious
You know where they hang out, who they listen to, what they read. You’re not guessing about whether to focus on Twitter or LinkedIn. You know.
Pricing makes sense
You understand what they’re already paying to solve related problems and what solving this is worth to them. You’re not pulling numbers from competitor pricing pages.
Marketing copy works
You’re not writing generic value props. You’re describing the exact gap they told you about, using words they used. When they read it, they think “this person gets it.”
Mind Map Your Customer’s World
After documenting the avatar, create a mind map to see who and what your customer interacts with. This reveals distribution channels and partnership opportunities you might miss otherwise.
Start with your customer type in the center, then branch out into categories:
Service Providers:
- What tools do they pay for? (Email providers, payment gateways, hosting, analytics, CRM, project management, etc.)
- What services do they use? (Legal, accounting, marketing automation, etc.)
Partners:
- Who do they work with regularly? (Business partners, technology partners, strategic alliances)
Competitors:
- Who else solves similar problems? (Direct competitors, alternative solutions, software vendors they currently use)
Internal Teams:
- If they’re in a company, what departments do they interact with? (Dev, Design, Marketing, Sales, HR, IT Support, Ops, Finance)
- What’s their org structure look like?
Client Types:
- Who are their customers? (Solopreneur, Digital, Physical, Funded, Bootstrapped, Nonprofits, SaaS clients, etc.)
Innovation & Trends:
- What emerging tech do they care about? (AI, new frameworks, industry trends, specific tools)
- What standards orgs or regulatory bodies affect them?
Industry Networks:
- What communities do they belong to? (Professional associations, networking groups, conferences they attend, online forums)
- What awards programs or certification agencies are relevant?
Marketing Channels:
- Where do they discover solutions? (Search engines, social media platforms, content marketing sites, paid advertising channels, influencer marketing, email marketing, newsletters, blogs)
Educational Resources:
- Where do they learn? (Online courses, webinars, workshops, certifications, industry publications, dev/design/marketing resources)
Contractors:
- What skills do they outsource? (Design, marketing, copywriting, SEO specialists, social media managers, brand consultants)
Example mind map structure for “Dev agency owners”:
(Note, I used MindNode Classic for this.)
This visualization helps you identify:
- Where to find more people like your customer
- Who has access to your customer already (partnership opportunities)
- What alternatives they’re comparing you against
- What ecosystem you’re competing in
Hot tip
Thinking WAY into the future, this is also a great way to partner, get strategic investment, and/or get acquired. Read more about that here (but not too long, you have work to do to get there!).
Map this out by hand or digitally. Update it as you learn more. Reference it when making marketing and distribution decisions.