Buyer Personas Are Broken. Here Is How to Fix Them

A buyer persona is a structured profile of the type of person most likely to buy from you, built from real data about your customers rather than assumptions about who you wish they were. Done well, a persona tells you how someone thinks, what they are trying to solve, and what would make them choose you over an alternative. Done badly, which is most of the time, it tells you that your target customer is a 38-year-old called “Marketing Mary” who enjoys yoga and drives a mid-range SUV.

I have sat in more persona workshops than I can count, across agencies and client-side teams, and the pattern is almost always the same. Someone pulls up a PowerPoint template, the room starts filling in boxes, and within an hour you have a fictional character with a name, a stock photo, a salary range, and a list of “pain points” that nobody has actually verified. The document gets shared on Confluence, referenced in the next strategy deck, and quietly forgotten. That is not research. That is creative writing with a marketing label on it.

Why Most Buyer Personas Are Not Worth the Slide They Are Built On

The problem is not the concept. The problem is the method. Personas fail at the research stage, before a single word of the profile is written.

When I was growing an agency from around 20 people to over 100, one of the things I noticed was how often we built audience profiles based on what clients told us about their customers, rather than what we could verify independently. The client would say “our buyer is a CFO, 45 to 55, risk-averse, focused on cost reduction.” And we would take that at face value, build messaging around it, and then wonder why the campaigns underperformed. The CFO was often not the buyer. Or they were the approver, not the evaluator. Or the pain point was not cost, it was compliance, or internal politics, or fear of looking wrong in front of their board.

Demographics are a starting point, not an answer. Age, job title, and income tell you who might be in the room. They do not tell you what they are thinking when they are in the room. The work of building a useful persona is the work of understanding the thinking, not cataloguing the characteristics.

This is part of a broader pattern I have seen throughout my career: marketers defaulting to inputs that are easy to collect rather than inputs that are hard to get but actually useful. Buyer personas are a prime example. It is faster to run an internal brainstorm than to conduct 20 customer interviews. It is easier to pull demographic data from your CRM than to analyse why 40% of your pipeline went cold in the last quarter. But fast and easy produces generic and wrong.

If you want to build sharper audience intelligence, buyer personas are one piece of a larger puzzle. The Market Research and Competitive Intel hub covers the full range of tools, from customer review analysis to positioning decisions, that give you a clearer picture of who you are actually selling to and why they buy.

What Good Persona Research Actually Looks Like

There are three sources that consistently produce useful persona data, and most teams use only one of them.

The first is customer interviews. Not surveys. Not Net Promoter Score. Actual conversations with people who bought from you, ideally within the last six months, where you ask open questions about the problem they were trying to solve, the alternatives they considered, and what tipped the decision. Fifteen to twenty of these, done properly, will tell you more than any analytics dashboard.

The questions that tend to produce the most useful answers are not “what do you like about our product.” They are: what were you doing before you found us, what made you start looking, who else did you speak to, and what almost made you choose someone else. The last question is particularly valuable. The thing that almost stopped a purchase is usually the sharpest signal about what your messaging needs to address.

The second source is lost deal analysis. Most teams spend their research time talking to customers they won. The customers they lost are often more informative. When I was running agency new business, I made it a practice to debrief every lost pitch, not just to understand what went wrong tactically, but to understand whether we had misread the buyer in the first place. Were we pitching to the wrong decision-maker? Were we solving a problem they did not actually have? Lost deals are a direct line to the gap between your persona assumptions and reality.

The third source is behavioural data from your own channels. What content do your best customers engage with before they convert? What search terms do they use? What pages do they visit, in what order, and where do they drop off? This is not a replacement for qualitative research, but it is a useful cross-check. If your persona says your buyer is motivated by cost savings, but your highest-converting content is about risk reduction, that is a signal worth investigating.

The Five Components That Actually Matter in a Persona

Strip out the demographics-first template and focus on five things that will actually change how you market.

The trigger. What caused this person to start looking for a solution? This is almost never “they decided they needed our product.” It is usually an event: a failed audit, a new hire, a competitor move, a board question they could not answer, a contract renewal coming up. Knowing the trigger tells you when to reach someone and what they are feeling when you do.

The desired outcome. Not the feature they want. The outcome they are trying to achieve. There is a significant difference between “wants project management software” and “wants to stop missing deadlines without having to hire more people.” The second version gives you a messaging brief. The first gives you a category.

The barriers. What would stop this person from buying, even if they wanted to? Budget, internal approval, risk aversion, previous bad experience with a similar product, distrust of vendors, timing. Understanding barriers is what separates a persona that informs messaging from one that just describes an audience.

The decision criteria. How does this person evaluate options? What matters most to them, what is table stakes, and what would they trade off? This is particularly important in B2B, where buying committees mean different people are applying different criteria to the same decision.

The information sources. Where does this person go when they are trying to make sense of a problem or evaluate a solution? Peer recommendations, industry publications, analyst reports, LinkedIn, specific communities or forums. This is your channel brief. If your buyer trusts peer recommendations above all else, that has direct implications for your content strategy, your review generation approach, and where you invest in brand presence.

How Many Personas Do You Actually Need

The honest answer is: fewer than you think.

I have seen teams build eight or ten personas and then produce marketing that could have been written for any of them, because the personas were not distinct enough to create genuinely different briefs. If your personas are so similar that the same campaign works for all of them, they are not personas. They are audience segments with names.

A useful test: if you removed the name and photo from two of your personas, could you tell them apart from the content alone? If not, merge them or go back to the research stage.

For most businesses, two or three well-evidenced personas are more actionable than a library of vague ones. If you are in a complex B2B sale with a buying committee, you might need separate profiles for the economic buyer, the technical evaluator, and the end user, because they have genuinely different motivations and information needs. But that is a specific structural reason, not a reason to multiply personas for the sake of thoroughness.

There is also a question of where you are in the business cycle. Early-stage businesses often do not have enough customer data to build credible personas, and should be honest about that. A hypothesis-based persona, clearly labelled as such and built for testing rather than for execution, is more useful than a confident-looking profile built on guesswork. I would rather see a team say “we think our buyer looks like this, here is what we are doing to validate it” than watch them build a campaign on assumptions they have never questioned.

The Persona Validation Problem Nobody Talks About

Building a persona is not a one-time exercise. Markets shift, buyer priorities change, and the person who bought from you three years ago may not look much like the person buying from you today. Most teams treat personas as a deliverable rather than a living input, which means they gradually drift from reality without anyone noticing.

One of the more useful habits I developed over the years was building a lightweight validation loop into the marketing calendar. Every six months, someone on the team would pull five or six recent customer interviews, review the last quarter’s lost deal data, and check whether anything in the existing personas needed updating. It takes a few days, not a few months, and it keeps the personas connected to what is actually happening in the market.

The other validation issue is confirmation bias. If the person running the persona research is also responsible for the marketing strategy, they have an incentive, often unconscious, to find evidence that confirms what they already believe. This is not a character flaw. It is how human cognition works. The fix is to separate the research from the strategy, even if that just means having someone else conduct the interviews or review the findings before they are incorporated into a persona update.

I judged the Effie Awards for several years, and one of the things that distinguished the work that won from the work that did not was the quality of the audience insight underneath it. Not the creativity of the execution, not the scale of the media spend. The sharpness of the human understanding. The campaigns that worked were almost always built on something specific and true about how a particular type of person thought about a particular problem. That specificity does not come from a workshop. It comes from research.

Turning a Persona into a Marketing Decision

A persona that does not change a decision is a document. The test of a useful persona is whether it produces different choices than you would have made without it.

Those choices might be about channel: if your persona tells you your buyer is a senior operations director who reads industry newsletters but never engages with social media, that is a direct input into where you spend your media budget. Or about message: if the primary barrier to purchase is risk aversion rather than budget, your messaging should lead with proof and reassurance rather than features and price. Or about timing: if the trigger is a contract renewal cycle, your outreach calendar should be built around that, not around your own quarterly targets.

The persona should also inform what you do not do. One of the more commercially important lessons I took from running agencies is that saying no to the wrong audience is as valuable as saying yes to the right one. If your persona research tells you that a particular segment has high acquisition cost, low lifetime value, and a tendency to churn at the first contract renewal, that is a reason to deprioritise them, even if they look attractive on paper. Personas are not just a tool for finding more customers. They are a tool for finding better ones.

There is also a connection here to something I think about a lot: the tendency to over-invest in capturing existing demand rather than creating new demand. If your personas only describe people who are already in the market and actively looking, you are missing the much larger group who have the same problem but have not yet recognised it or started searching for a solution. The most commercially valuable persona work often includes a profile of the latent buyer, the person who would buy if they understood the problem better, not just the active buyer who is already comparing vendors.

Simplicity in persona structure matters too. A profile that tries to capture everything ends up informing nothing. Forrester has made the case for simplicity as a strategic advantage, and that principle applies directly to how personas are built and used. If the profile cannot be summarised in a single paragraph that everyone on the team can remember, it is too complex to be actionable.

Buyer personas sit at the intersection of research, strategy, and execution. If you want to understand the full research toolkit that feeds into this kind of audience work, the Market Research and Competitive Intel hub covers the methods, frameworks, and practical approaches that make the difference between surface-level profiling and genuinely useful audience intelligence.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions