Your Customer Persona Isn’t Real Until You’ve Done This First.
Most businesses build a persona before they understand their audience. Here’s why that order is backwards — and what to fix first.
I write about practical marketing for solo founders and small teams at marketedbyvivek.com this piece is part of a longer series on audience and persona research.
Ask most business owners who their customer is, and you’ll get an answer fast. Too fast.
“Women, 25 to 40.” “Small business owners.” “Working professionals.”
None of that is wrong. It’s just not a person — it’s a segment. And here’s the mistake I see constantly, including in my own early work: businesses skip straight from that vague segment to building a “customer persona,” complete with a name, a stock photo, and a neat little bio. It looks finished. It feels like progress.
It’s usually built on nothing.
Two Different Jobs, Often Confused as One.
There are really two separate steps here, and most people collapse them into one.
The targeted audience is the raw material, the broader segment of people who share something real: an age range, an income bracket, a lifestyle, a location, or a problem. It’s built from data. Google Analytics, Search Console, Meta Ads Manager, Hotjar session recordings, and competitor comment sections if you’re too new to have your own data yet. It’s research.
A customer persona is the finished character built from that raw material—one specific, semi-fictional person with a name, a job, a set of goals, and a set of pain points, standing in for the audience segment you already understand.
The audience tells you who’s out there. The persona tells you how to talk to them. And you genuinely cannot do the second one honestly without having done the first.
What Happens When You Skip the Audience Step.
Here’s what a persona looks like when someone builds it without audience research first:
“Rahul, 32, works in tech, likes fitness and good coffee.”
That’s not a persona. That’s a guess, wearing a name tag. It has no data behind it — no reason to believe this is actually who buys, no behavioral pattern, no validated pain point. It feels specific because it has a name and an age, but specificity isn’t the same as accuracy. You could write ad copy for “Rahul” all day and still be talking to nobody real.
Compare that to a persona built the right way around — audience first, persona second.
How This Actually Plays Out — Using Myself as the Example.
I run a one-person marketing practice. If I’d started by writing a persona off the top of my head, I’d have probably landed on something like “male, 30s, small business owner.” True, but useless — it wouldn’t have told me anything about how to actually reach that person.
Instead, the audience work comes first. Demographically: solo operators and small teams, usually without a marketing hire, often juggling the work themselves. Psychographically, this is where it gets useful: genuinely interested in tools and shortcuts that save time but skeptical of anything that feels like hype. Behaviorally, they research before they buy, they don’t respond to heavily produced marketing, and they engage with content that’s honest and practical over content that’s polished and empty.
None of that is invented. It comes from watching what content actually gets engagement, what questions show up in comments and DMs, and what objections come up in real conversations before someone decides to work with me.
Get Vivekhari’s stories in your inbox
Join Medium for free to get updates from this writer.
Only then does the persona get built, and it comes out specific instead of generic:
Vivek, 30, entrepreneur, running a small business on his own. No team to delegate to, so every extra task lands on him directly. Actively looking for tools that save time rather than add to his plate. Values work-life balance — he’s not chasing growth at any cost, and family time and a bit of space to cook isn’t negotiable for him. He researches before he buys, and he tunes out anything that sounds like a pitch.
Notice the difference. This version tells you exactly what subject line would get opened, what tone your customer service should use, and what kind of content would make him stop scrolling. “Male, 30, business owner” tells you none of that. The only difference between the two is whether audience research happened first.
Audience Isn’t One Thing—Know Which Angle Actually Explains Your Buyer.
Part of why people skip the audience step is that “find your audience” sounds like one vague task. It’s actually four different angles, and most products are mostly explained by just one or two of them.
Demographics are the obvious ones—age, income, occupation, and family status. It’s the easiest to find data on, and it’s usually where people stop. But on its own it only tells you who someone is, not why they buy.
Psychographics is the ""why"—interests, values, and lifestyle. Two people can share the exact same age and income and buy for completely different reasons. A ready-to-cook food brand isn’t really targeting “working professionals” as a demographic; it’s targeting a lifestyle—busy, limited time, and unwilling to trade away quality for convenience. That’s a different lens entirely, and it’s usually the one that makes your messaging land.
Geographic matters more than people give it credit for. A water filter brand solving a hard-water problem doesn’t launch everywhere at once it goes where the water is actually hard. If your product solves a location-specific problem, geography can be the highest-leverage angle you’re ignoring.
Retargeting is different from the other three; it's not about finding someone new; it’s about following up with someone who already raised a hand. They visited, they engaged, they didn’t convert. They’re not a stranger; they just haven’t finished deciding yet.
Most businesses end up using a mix, but usually one angle does the heavy lifting. For me, it’s psychographic; the interest in time-saving tools and the resistance to hype matter more than any demographic detail. Knowing which angle actually explains your buyer tells you where to point your research instead of collecting data everywhere and understanding nothing.
A Persona Isn’t Finished Until It’s Been Checked.
One more step people skip: validating the persona before treating it as final.
Once you’ve built it, take it back to your team and see if it matches what they’re actually seeing in conversations and support tickets. If you can, take it to a few real customers and ask if it sounds like them. A persona that’s never been checked against a real person is still a guess it just has a name attached now, which makes it feel more finished than it is.
This is also why a persona isn’t a one-time document. Revisit it every few months. Customer behavior shifts, and a persona built a year ago on last year’s audience data may not reflect who’s actually buying from you today. The audience moves; the persona has to move with it.
Why the Order Actually Matters.
It’s tempting to think of this as a formality do the audience work, check the box, and move to the persona. It’s not. The order changes what the persona is even capable of doing for you.
A persona built without audience research is a projection of what you assume your customer is like, usually a slightly idealized version of yourself or whoever you last had a good conversation with. It’ll feel right because you made it up, and things we make up always feel intuitively correct to us.
A persona built on audience research is a compression of a pattern that actually repeats. It holds up under pressure because it isn’t a guess; it’s a summary of real behavior. That’s the version your content, your ad copy, and your product decisions can actually be built on with confidence.
The Practical Takeaway.
If you’re about to build a persona and you haven’t touched analytics, talked to a customer, or read a competitor’s comment section yet stop. Do that first. It doesn’t need to be exhaustive. Even a rough pass at who’s actually engaging, what they’re asking, and what’s stopping them from buying is enough raw material to build a persona that means something.
Once you’ve got that raw material, a blank structure helps me put together a fill-in persona template that walks through each field (demographics, psychographics, behavior patterns, goals, and pain points) so you’re not starting from a blank page.
You also don’t need to wait for perfect data. If you’re a brand-new business with no customer base yet, study your competitors’ platforms instead — look at what they promote and how, and read the comment sections for people actively engaging. They may not be your exact audience, but they’re close enough to give you a real starting signal instead of a blank page. The persona you build from that will still be more honest than one built purely from assumption.
The persona is the payoff. The audience is the work that makes the payoff true instead of decorative. One explains who’s out there; the other turns that understanding into someone specific enough to write for, design for, and sell to without either team guessing separately.
Get the order right, and one clear persona built the honest way is enough to align your content, your product, and your sales conversations around the same real person. Get the order backwards, and you’ve just given a fictional character a name.
If you want the fuller breakdown — where to find audience data, how to validate a persona, and the full template — I’ve written it out in more depth at marketedbyvivek.com/blog.
