How to Create Customer Personas That Are Actually Useful
How to create customer personas grounded in real research, not guesswork — the fields that matter, where to source them, and how to keep personas actionable.
Key Takeaways
- Useful personas are built from real research; invented personas are just the team's assumptions with a stock photo.
- The fields that drive decisions are goals, pain points, objections, and language — not age and favorite coffee.
- Ground each persona in evidence: real quotes and behaviors from communities, reviews, and conversations.
- A persona should make a decision easier — if it never changes a choice, it is decoration.
- Pair personas (who) with jobs to be done (why) for a complete picture.
Most customer personas are useless — not because personas are a bad idea, but because of how they're made. A team sits in a room, invents "Marketing Mary, 34, loves yoga and productivity apps," slaps on a stock photo, and never opens the slide again. That's not research; it's the team's assumptions wearing a costume.
A useful persona is different: built from real evidence, focused on the fields that actually change decisions, and specific enough that someone could write on-target copy from it alone. Here's how to make that kind.
The Fields That Actually Matter
Skip the demographic trivia unless it genuinely drives behavior. The fields that change product and marketing decisions are:
- Goals — what this person is trying to achieve (their desired progress).
- Pain points — what blocks them, ranked by frequency and intensity.
- Objections — what makes them hesitate or churn.
- Buying triggers — the moment or event that prompts them to seek a solution.
- Channels — where they spend attention and where you can reach them.
- Language — verbatim phrases they use to describe the problem.
Age, name, and favorite coffee order rarely change a single decision. Include a demographic field only when it actually affects how you build or market — otherwise it's decoration that makes the persona feel researched without being useful.
Ground Every Field in Evidence
The rule that separates useful personas from fiction: every field must trace to real data. Not "we think they're frustrated by X" but "here's a real quote where three different people said X."
That means personas come after research, not before. Sources, roughly by depth:
- Interviews and sales calls — the deepest signal on goals, objections, and triggers.
- Analytics — actual behavior (what they do, where they drop off).
- Reviews — candid opinions with buyers attached.
- Community discussions — unprompted goals, pains, and language on Reddit and forums.
For the pain-point and language fields specifically, community research is the most efficient source. PainPointMap scans the subreddits where your audience gathers and returns ranked pain points in the audience's own words, each linked to a source thread — those drop directly into the persona's "pain points" and "language" fields as evidence-backed entries rather than guesses.
The Build Process
- Gather the raw research across the sources above.
- Find the clusters. Look for groups of people who share goals, pains, and behaviors. These clusters — not demographic segments — become your personas.
- Document the decision-driving fields for each cluster, grounding each in a real quote or behavior.
- Name them by behavior, not demographics. "The Anxious First-Time Founder" beats "Mary, 34" because the name itself signals the job and the emotion.
- Pressure-test: would this persona actually change a decision? If not, cut or merge it.
How Many Personas?
Fewer than you'll be tempted to make — usually two to four. Each should represent a genuinely distinct set of goals and behaviors that would lead to different decisions. If two personas would get the same message and use the product the same way, they're one persona. A pile of personas is a sign you sliced by demographics instead of behavior.
Personas + Jobs to Be Done
Personas answer who; jobs to be done answer why they act. They're complementary. A persona tells you who you're talking to and what they care about; the job tells you the progress they're trying to make in a given situation. One person (persona) hires different products for different jobs. Use personas for targeting and empathy, JTBD for what to build and how to position it — together they're far stronger than either alone.
Keep Them Alive
A persona built once and never revisited goes stale as your audience and market shift. Revisit them when results change or every few months, refreshing the pain-point and language fields from current community research. A living persona stays useful; a laminated one becomes a relic.
Related Reading
- How to Do Audience Research — the research that feeds personas
- Jobs to Be Done Framework — the "why" that complements the "who"
- Find Your Target Market — narrowing before you profile
- Voice of Customer Research — sourcing the language field
Frequently Asked Questions
How do you create a customer persona?
Start with real research, not imagination. Gather data from customer conversations, community discussions, reviews, and analytics. Identify patterns — clusters of people with shared goals, pain points, and behaviors. For each cluster, document the fields that drive decisions: goals, pain points, objections, buying triggers, channels, and the language they use. Ground every field in evidence (a real quote or behavior). The result is a research-backed persona, not a guess with a name.
What should a customer persona include?
The decision-driving fields: goals (what they are trying to achieve), pain points (what blocks them), objections (what makes them hesitate), buying triggers (what prompts action), preferred channels (where to reach them), and verbatim language (how they describe their problem). Demographics matter only if they actually affect behavior. Skip the filler — name, age, and hobbies rarely change a product or marketing decision.
How many customer personas should you have?
Fewer than you think — usually two to four for most businesses. Each persona should represent a genuinely distinct set of goals and behaviors that would lead to different product or marketing decisions. If two personas would receive the same message and use the product the same way, they are one persona. Too many personas dilutes focus and signals that they are demographic slices rather than behavioral ones.
What is the difference between a persona and an ideal customer profile (ICP)?
An ICP describes the type of company or account worth targeting (industry, size, budget, needs) — it is a firmographic filter used mostly in B2B. A persona describes an individual person within that account or market (their goals, pains, and behavior). ICP answers "which companies should we sell to"; persona answers "who is the human we are speaking to and what do they care about." B2B teams often use both together.
How do you research data for customer personas?
Combine sources: interviews and sales calls for depth, analytics for behavior, reviews for candid opinions, and community discussions (Reddit, forums) for unprompted goals and language. Community research is especially efficient for the pain-point and language fields — PainPointMap scans your audience's subreddits and returns ranked pain points in their own words, which drop straight into the persona. Ground each field in a real quote so the persona stays evidence-based.
Find what your customers are actually complaining about.
Scan your target audience's subreddit and see real pain points ranked by severity — no surveys, no guesswork.
Scan My Target MarketRuns the original data and analysis pieces on the blog, scanning Reddit communities at scale to surface patterns in what founders and operators actually struggle with.