What this article is about
Customer personas done usefully — what a working persona actually contains, why most personas fail, the difference between demographic information and the things that actually matter, how to build one from real evidence rather than imagination, how to use it in actual decisions, and how to keep it alive over time. Written for owners who want a working tool, not a workshop output.
Customer personas are one of those marketing artefacts that everybody is told to make and almost nobody uses. The pattern is consistent. A workshop is held. A document is produced — usually with a stock photograph of a smiling person, a fictional name, a list of hobbies, and a few demographic facts. The document gets printed, pinned to a wall, referenced in one or two early meetings, and then quietly forgotten. A year later, the team is making decisions exactly as they did before. The persona exists; nothing has changed.
This is not a failure of the concept. A real customer persona — one built from evidence, written to be used, and revisited as the business learns — is one of the most useful thinking tools a small business can build. It makes copy sharper, channel choices clearer, product priorities more honest, and team alignment easier. The failure is in the kind of persona most businesses end up with, which is theatre. Distinguishing the working persona from the decorative one is most of the work, and it is what this piece is about.
What a Customer Persona Actually Is
A customer persona is a written summary of a specific kind of customer the business serves, designed to inform decisions about how to serve them. It is not a fictional biography. It is not a demographic profile. It is not a list of preferences gleaned from a survey. It is a document that captures the meaningful things about a customer’s situation, motivations, and decision-making in enough detail that a team can answer practical questions by reference back to it.
A useful persona is built around one question: when someone on the team is making a marketing or product decision, can they ask “what would [persona] think about this?” and get a real answer? If the persona is too vague or too generic, the answer is “I do not know” — which is the same place the team would have been without the persona. If the persona is detailed in the wrong ways — full of hobbies and demographic facts that do not bear on the decision — the answer is “the persona says nothing useful about this.” Only when the persona captures what actually matters does it function as a working tool.
The shift in mindset is from persona-as-deliverable to persona-as-thinking-aid. Once it is the latter, every choice in how to build it gets clearer.
Why Personas Done Badly Are Worse Than No Personas
A surprising thing about poor personas: they actively harm the businesses that use them. Better to have no persona than a bad one, and here is why.
A bad persona creates a false sense of clarity. The team believes it knows its customer because it has a document. The document, on inspection, says very little useful — it lists demographic facts, fictional preferences, and aspirational descriptors. But the existence of the document allows the team to stop asking the harder questions about who the customer actually is. The discipline of inquiry is replaced by the comfort of having an answer, even when the answer is empty.
A bad persona also tends to skew decisions in the direction of the persona’s most superficial features. If the persona is described as a “32-year-old urban professional,” the team begins to optimise for things that 32-year-old urban professionals like — even when the actual customer base spans a much wider age range and most of them are not particularly urban. The persona is now actively distorting the business’s understanding of who it serves.
The fix is not to abandon personas but to build better ones. The path from bad persona to good persona is the path from imagined detail to observed evidence, and from demographic categorisation to genuine understanding of context and motivation.
What a Useful Persona Actually Contains
A working persona contains a small number of things, all chosen because they bear on actual decisions.
Context. The situation the customer is in when they encounter the business. Are they researching for a future need, in active buying mode, dealing with a recent problem, planning a long-term project? Context shapes how the customer reads the marketing and what they need from the experience.
Motivations. What the customer is actually trying to accomplish, beyond the surface transaction. The person buying running shoes may be trying to start a habit, recover from injury, prepare for a race, or replace worn-out gear. Each of these motivations implies a different conversation. A persona that does not capture motivation is a persona that cannot guide messaging.
Frictions. The specific things that get in the way of the customer choosing this business — concerns about price, distrust of the category, past bad experiences with similar businesses, decision-makers they need to convince, internal politics, lack of confidence in their own ability to evaluate. Frictions are what the marketing has to address; a persona that ignores them is a persona that cannot help.
Decision criteria. What the customer is actually weighing when they decide. Not what they would say in a survey — what they are observably weighing in the real conversations. Price. Reputation. Specific features. Ease of return. The opinion of someone they trust. Different customers weigh these differently, and a persona should capture the weighting that matters.
Where they currently look. The places this customer is already going to find information, recommendations, or solutions in this category. Specific channels, publications, communities, advisers, search behaviour. This is the most practical input to channel selection — the persona tells you where the marketing needs to be.
Language they use. The actual phrases, terms, and frames the customer uses to describe their own situation. This is the closest input to copywriting. A persona that captures real customer language enables marketing that sounds like the customer’s own internal monologue rather than the company’s internal jargon.
A persona with these six elements, written briefly, is a working tool. A persona that has only demographic facts and a stock photograph is decoration.
Why Demographics Are the Least Useful Part
The persona tradition tends to lead with demographics. Age. Income. Location. Job title. These categories are easy to capture and they look like data, which is why they show up first in most personas. They are also, for most businesses, the least decision-useful part of a customer’s profile.
Consider any decision the team is likely to make. Should the headline emphasise speed or trust? Should the next product include this feature or that one? Should the marketing focus on this channel or that one? In almost every case, the answer depends not on the customer’s age but on their context, motivation, and frictions. A 25-year-old and a 55-year-old buying the same product for the same reason behave more similarly than two 35-year-olds buying it for different reasons.
This is not to say demographics are useless. They are useful as filters — for channel targeting, for content tailoring, for sizing the addressable market. But they should not be the substance of the persona. The substance is the customer’s situation and what they are trying to do. Demographics are context for that substance, not a substitute for it.
The honest test: if you removed the demographic information from your persona and showed the rest to a stranger, would they recognise a specific kind of customer? If yes, the persona has substance. If no, the demographics were doing all the work, which means the persona was not actually telling you anything about the customer’s mind.
One or Two Personas, Not Five or Seven
Most businesses, given the chance, will create more personas than they need. The instinct comes from a real observation — customers vary — and produces a wrong conclusion: that each variation should have its own persona. The result is a set of five or seven personas, each thinly described, each rarely used, each diluting the team’s mental model of who the business serves.
A useful persona set is small. One persona, sometimes two, occasionally three for businesses with genuinely distinct customer types. Beyond that, the team cannot hold the personas in mind clearly enough to use them, and the personas become a reference document that nobody references.
The discipline is to identify the one or two customer types that account for most of the business’s actual customers, and to build deep personas of those — accepting that some customers will fall outside the persona definitions, and that is acceptable. Trying to capture every variation produces a set of vague descriptions; committing to a few specific ones produces tools that actually guide decisions.
The test is whether the team can summarise each persona in two or three sentences from memory. If yes, the personas are usable. If team members have to look the personas up every time they reference them, the set is too large.
Building a Persona From Real Evidence
The single most important difference between a working persona and a useless one is the source of the information. A useful persona is built from real evidence about real customers. A useless persona is built from imagination, demographic assumptions, and the brand’s idealised picture of who it serves.
The evidence to draw on, in roughly increasing order of usefulness.
Sales conversations. The team talks to prospects every day. The patterns in those conversations — the questions that come up repeatedly, the concerns that recur, the language customers use — are persona material. Capturing them systematically takes discipline but produces real signal.
Support tickets and customer service notes. The friction customers experience after they buy is documented in the support inbox. Patterns there reveal what customers thought they were getting versus what they actually got, which is some of the most useful persona information available.
Customer interviews. Direct conversations with current or recent customers about their context, motivations, decision process, and frictions. Six to ten well-conducted interviews are usually enough to surface the patterns that matter. Interviews should be conducted with as much listening as possible and as little leading as the interviewer can manage.
Behavioural observation. Watching customers actually use the product or navigate the buying process. Where do they hesitate? Where do they ask for help? Where do they make assumptions that turn out to be wrong? Behaviour is more revealing than what customers say about their behaviour.
Survey data, used carefully. Surveys are useful for sizing patterns that have been identified by other means. They are not useful for discovering patterns in the first place, because they answer the questions the team has already thought of.
Building a persona from evidence is more work than building one from imagination, but it is the work that makes the difference between a document that gets used and one that gets ignored.
How a Persona Gets Used in Actual Decisions
A persona’s value is realised at the point of decision-making. The places where a good persona earns its keep.
Copywriting. The persona is the imagined reader. Every paragraph, headline, and call to action should be tested against “would [persona] read this and feel understood?” Copy written without a persona in mind tends to be generic; copy written with a clear persona is sharper and more recognisable to the right audience.
Channel selection. The persona tells you where they currently look for solutions. This is the most direct input to channel choice. If the persona reads three specific publications, listens to two specific podcasts, and asks two specific kinds of advisers, the channel strategy starts to write itself.
Product and service decisions. The persona’s frictions and motivations are inputs to what the business should offer and how. A persona who consistently struggles with one part of the experience is a signal about where the next product investment should go.
Prioritisation. When multiple opportunities compete, the persona helps. Which of these would serve our defined customer best? Which is just interesting? The persona makes this question answerable.
Onboarding and customer experience. The persona’s situation when they arrive shapes how the experience should be designed. A persona who arrives anxious needs a different welcome from one who arrives confident.
A persona that does not appear in any of these decisions is a persona that is not being used, regardless of how impressive the document looks.
The Signs That a Persona Is Not Being Used
A few honest tests.
The team cannot summarise the persona without looking it up. The persona lives in a folder, not in the team’s working memory.
Decisions get made without anyone referencing the persona. Campaigns are launched, copy is written, features are scoped, and the persona is not mentioned in any of the discussions.
The persona’s photograph is more memorable than its content. The team remembers the smiling stock photo and the fictional name; the substance of the persona has faded.
New hires are not introduced to the persona, or are introduced to it briefly in onboarding and never see it again.
The persona has not been revisited in eighteen months. The business has changed, the customer base has evolved, and the persona is now a record of who the customers used to be.
When several of these signs are true, the persona is decorative. The fix is to either rebuild it as a working tool or to accept that the document is no longer useful and to retire it without ceremony.
How to Keep Personas Alive
A working persona is a living document, not a finished one. Keeping it useful requires periodic revisiting.
Test the persona against new customers. Each new customer is data — does this customer fit the persona, partly fit it, or fall outside it entirely? The answers, accumulated over months, tell you whether the persona is still accurate.
Update the language. Customer language shifts over time, particularly as categories evolve. The phrases the persona was built on may no longer be the ones current customers use. Refreshing the language section keeps the persona resonant.
Retire personas that no longer fit. Businesses evolve. Some customer types become less important, others emerge. A persona set that has not been pruned in three years is likely carrying personas that no longer reflect the business. Retiring the wrong ones is as important as creating new ones.
Revisit the persona quarterly, lightly. A short check-in — what have we learned about this persona in the last quarter, is anything materially different, does the team still find it useful — is enough to keep the document honest. Annual deep reviews are reasonable for businesses with stable customers; quarterly is better for businesses in flux.
The discipline is to treat the persona the way the business treats its other working documents — as something that exists to be useful, and that should be updated when it stops being so.
Key Takeaways
- A customer persona is a written summary of a specific kind of customer designed to inform decisions, not a fictional biography or a demographic profile.
- Bad personas are worse than no personas because they create a false sense of clarity that crowds out genuine inquiry.
- A useful persona captures context, motivations, frictions, decision criteria, where the customer currently looks for solutions, and the language they use.
- Demographics are the least decision-useful part of a persona; context and motivation matter more.
- One or two personas is usually enough; five or seven is almost always too many.
- Useful personas are built from real evidence — sales conversations, support tickets, customer interviews, behavioural observation — not from imagination.
- Personas earn their value at the point of decision — in copywriting, channel selection, prioritisation, product, and customer experience.
- The signs a persona is unused: team cannot summarise it, decisions ignore it, the photograph is more memorable than the content.
- Keep personas alive by testing them against new customers, updating language, retiring obsolete ones, and revisiting quarterly.
A note from SWL
The simplest test for whether your current personas are working is to ask the team to describe them from memory, then ask when each persona was last referenced in an actual decision. If the answers are awkward, the personas are decorative — which is normal, and worth fixing. A working persona is one of the higher-leverage small investments a business can make in marketing clarity. If you are looking at yours and wondering whether they are doing what they should, we are happy to help you think that through.