Most founders know they should have a customer persona. Far fewer have one that actually changes how they market. The persona ends up as a tidy one-pager nobody refers back to, full of demographic detail that does not help anyone write a better ad or pick a better channel.
The research is consistent on why this matters. McKinsey's analysis of personalisation found that brands that get personalisation right generate 40% more revenue from those activities than average performers, and that personalisation typically lifts revenue by 10 to 15%. None of that is possible without a clear picture of who you are talking to. The Cintell B2B Buyer Benchmark Study, still one of the most-cited pieces of research on this, found that 71% of companies that exceed their revenue and lead goals have documented personas, compared to only 37% of companies that simply meet their goals.
This post is a practical, step-by-step walkthrough for building your first persona from scratch - one you will actually use.
Step 1: Decide what the persona is for
Before you collect any data, get clear on the decision the persona needs to inform. A persona that has to do everything ends up helping with nothing.
Useful framings include:
- "Who is this product for, so we can stop trying to sell to everyone?"
- "What hooks should we use in our ads?"
- "Which content topics should our blog focus on?"
- "Which channels are worth showing up on, and which are wasted effort?"
Pick one primary use. You can build more personas later, but the first one should be tied to a single, concrete marketing decision you have been struggling with.
Step 2: Pick your data sources
A persona is only as good as the inputs behind it. The most common mistake is leaning entirely on one source - usually your own gut feel or a quick scroll through analytics.
Aim for at least three of the following:
- Customer interviews. Twenty-minute calls with five to ten existing customers. Ask about the problem they were trying to solve, what they tried before, and how they made the decision to buy.
- Sales and support conversations. Even if you do not have a sales team, your inbox is a goldmine. Search for the words customers use to describe the problem and your product.
- Quantitative survey or quiz responses. Open answers tell you what people care about. Closed answers (multiple choice, scales) let you spot patterns at scale.
- Analytics behaviour. Which pages convert? Which traffic sources stick around? Which features get used and ignored?
- Existing customer reviews and competitor reviews. Both yours and competitors'. The complaints are often more useful than the praise.
This is not about volume. It is about triangulating. A pattern you can see in interviews and in survey responses and in your analytics is far more trustworthy than any single source on its own.
Step 3: Look for patterns - especially psychographic ones
This is the step most persona templates skip. Demographics ("a 34-year-old marketing manager called Sarah") are easy to write down and rarely useful for marketing decisions. What actually shifts how you write copy and pick channels is psychographic data - the values, motivations, and habits behind the buy.
When you look across your data sources, group what you find into a few categories:
- Motivations. What outcome are they trying to reach? What does success look like to them?
- Frustrations. What is broken about the current way they solve this problem?
- Discovery habits. Where do they go when they want to learn something new? Newsletters, podcasts, YouTube, Reddit, peers?
- Decision style. Do they research exhaustively before buying, or trust a recommendation and move quickly?
- Values. What do they care about beyond the product itself - sustainability, status, simplicity, community?
For more on why these categories matter more than demographics, see our post on what psychographic segmentation is and why it beats demographics.
You are looking for two or three clear segments. Most early-stage brands find that one segment dominates and one or two are interesting but smaller. That is a useful finding in itself.
Step 4: Write the persona
A good persona document is short. One page, maximum. Anything longer stops being a tool and starts being a deliverable.
Use this structure:
- Name and one-line summary. Give the persona a memorable name and describe them in one sentence. "The Quietly Ambitious Founder: solo or two-person business owner, scaling carefully, allergic to hype."
- The job they are trying to do. The outcome they want, in their language.
- What they have tried before, and why it did not work. The current alternatives and their flaws.
- What they care about. Three to five values or priorities.
- Where they spend attention. Two or three discovery channels you have evidence for.
- What would make them trust you. Proof types that matter most - case studies, peer recommendations, free tools, content depth.
- What would put them off. The anti-patterns.
Skip stock photos, fake quotes, and demographic detail that is not actionable. A persona that says "reads Hacker News, distrusts marketing language, wants to see the product before reading about it" is more useful than one that says "lives in London, age 32, married with one child".
Step 5: Apply the persona to actual marketing decisions
A persona that does not change what you do is not a persona, it is a document. Run yours through three tests:
- Copy test. Take your homepage headline. Would your persona stop scrolling for it? Rewrite it in their language - the words you saw repeated in interviews and reviews.
- Channel test. Look at where you are currently spending time and money. Is your persona actually there? If they get their information from peer recommendations and long-form content, paying for short-form social ads is probably the wrong move.
- Offer test. Does the next thing you ask them to do match how they buy? A persona that researches exhaustively wants a comparison page, not a "book a call" button.
This is where the persona earns its keep. A study published in the Salesforce State of the Connected Customer report found that 88% of customers say the experience a company provides matters as much as its products or services. The persona is your shortcut to designing that experience deliberately.
The shortcut: let a quiz do the data collection
The bottleneck in this whole process is Step 2 - collecting data from real customers. Interviews take weeks to book. Surveys get low response rates because they feel like work. Most founders run out of patience before they have enough signal to spot patterns.
A persona-building quiz solves this by making data collection feel like value, not work. The customer gets a personalised result; you get the answers behind it. Once a hundred people have completed the quiz, the patterns are already there. Group those patterns and you have the raw material for a persona, without the weeks of interviewing.
If you want a step-by-step on the quiz design itself, our guide on how to create a customer persona quiz covers the question types and result type structure that produce the best persona data. Our 5 steps to build personas with quiz data walks through what to do with the responses once they come in.
Try Profyl
Profyl is built for exactly this. You create a short, branded quiz - a few well-chosen questions - and the platform translates the responses into an Audience Snapshot: psychographic scores across five dimensions, a result distribution showing which segment dominates, and AI-generated marketing insights you can apply to copy, channels, and offers.
It turns the persona-building process from a multi-week research project into something you can run in the background of your normal marketing, with your real customers as the data source.