How to Build Marketing Strategy When You Don’t Know Your Target Customer A woodworker I know started selling handmade cutting boards online with a plan built on a guess: his customer was “home cooks who care about quality.” He built his product photography around rustic kitchen scenes and posted into a cooking-focused corner of Instagram for three months. Sales were slow, but not zero, which made the problem harder to diagnose than if nothing had sold at all. A trickle of orders is exactly the kind of result that lets a bad assumption survive far longer than it should, because it’s easy to tell yourself the strategy is working, just slowly.
What eventually made him question the assumption wasn’t one clean insight—it was a handful of small, individually forgettable things that took a while to add up. A customer asked if a board could be engraved with two names and a date. Another mentioned, almost in passing, that she needed it shipped before a specific Saturday. A third comment used the word “registry.” None of these on their own screamed “wedding gift.” Taken together, over a few weeks, they started to feel like more than coincidence—but he wasn’t certain, and treating a hunch as a confirmed answer would have just replaced one guess with another.
So instead of overhauling his entire marketing around a hypothesis he wasn’t sure about, he tested it cheaply first. He wrote one new product listing description specifically framing the board as a personalized wedding gift, added an engraving option to the checkout page, and ran a small, low-budget batch of ads targeting people actively planning weddings—a fraction of what he’d normally spend testing anything. If the wedding framing was wrong, he’d lose very little. It wasn’t wrong. That single listing outperformed every general “quality kitchen board” listing he’d run for three months, and only then did he commit to rebuilding the rest of his strategy around it.
Why Demographic Personas Fail When You Don’t Have Evidence
The standard advice for finding a target customer is to build a persona—age range, income bracket, interests, and pain points. That process assumes you already have enough real signal to fill those boxes in accurately. When you don’t, and you fill them in anyway, you’re not doing research. You’re writing a plausible-sounding guess and then treating it with more confidence than it’s earned, simply because it’s been formatted to look like a research output.
This matters most for small or new businesses, because a demographic persona describes a category of people, not a reason any specific person buys. Age and income can vary enormously among people who are all buying your product for the exact same underlying reason, and a persona built around demographics alone will miss that shared reason entirely. It’s not that demographic data is worthless—it’s that it answers a different question than the one that usually matters most early on, which is, what specific situation is bringing someone to consider this purchase at all?
A Framework for Finding Your Real Customer: Buyer → Occasion → Intent → Evidence → Hypothesis → Test
Instead of starting with demographics, this framework starts with behavior and works backward toward a testable idea.
Buyer. Who is actually completing the purchase—not who you imagine using the product, but who is clicking “buy” and entering payment details. These are not always the same person, and conflating them is one of the most common mistakes in this entire process.
Occasion. What specific situation is prompting the purchase right now? Not a general need, but a triggering moment—a wedding, a move, a deadline, a gift-giving occasion.
Intent. What is the buyer actually trying to accomplish with this purchase? A wedding gift buyer isn’t trying to acquire “a quality kitchen tool.” They’re trying to give something that feels meaningful and personal. Intent shapes what actually matters to them far more than the product’s own features do.
Evidence. What specific, concrete signals—messages, comments, requests, questions—actually support a theory about buyer, occasion, and intent? Not a feeling. An actual thing someone said or asked for.
Hypothesis. Once several pieces of evidence point in a similar direction, state the theory plainly: “Our buyer is X, purchasing for occasion Y, with intent Z.” Write it down. A hypothesis that isn’t written down tends to quietly reshape itself to match whatever you already believed.
Test. Before rebuilding an entire strategy, test the hypothesis cheaply—one listing, one small ad set, one direct outreach message—and see whether real behavior confirms it before committing further resources.
The woodworker’s version: buyer (often not the cook), occasion (wedding or similar milestone), intent (a meaningful, personalized gift), evidence (engraving requests, shipping deadline mentions, the word “registry”), hypothesis (“our buyer is someone shopping for wedding or milestone gifts, not home cooks buying for themselves”), test (one reframed listing and a small ad set before a full strategy rebuild).
Telling a Real Segment Apart from an Interesting Coincidence
Here’s where this gets genuinely difficult: what if one customer buys for a wedding, another for a housewarming, and a third for no stated occasion at all? It’s tempting to either declare the first pattern you notice to be “the” target market or to conclude the data is too messy to mean anything. Both reactions are premature.
A single data point is an anecdote. A pattern worth acting on needs to show up independently, across multiple unconnected customers, without you having to stretch to see it. If three out of your last five customers, none of whom know each other, all mention a gift-giving occasion unprompted, that’s a real signal. If it’s one customer out of fifteen, it’s worth noting but not worth rebuilding a strategy around yet.
It’s also worth checking whether different occasions actually share the same underlying intent, even if the specific event differs. A wedding gift and a housewarming gift might both reflect the same intent—a meaningful, personalized gift for someone else’s milestone—even though the occasion itself varies. In that case, the real segment isn’t “wedding shoppers” narrowly; it’s “milestone gift-givers” more broadly, and narrowing the hypothesis too early, to just one specific occasion, can cause you to miss the actual, larger pattern underneath it.
Who Buys Is Not Always Who Uses
This is easy to overlook, and it matters more than almost anything else in this process: the person who uses your product and the person who decides to buy it are frequently not the same person, and marketing has to be built around the buyer’s decision, not the eventual user’s experience.
The woodworker’s cutting boards are used, day to day, by home cooks. But a meaningful share of his actual customers were never planning to use the board themselves—they were choosing a gift for someone else’s kitchen. If he keeps marketing to the cook’s priorities (durability, grain pattern, maintenance) while his real customer is a gift-giver whose priorities are entirely different (does it feel meaningful, can it be personalized, will it arrive on time?), the marketing will keep missing the actual decision-maker even if the product itself is exactly right for its end user.
This distinction is worth checking for in almost any product or service: ask directly whether the person paying and the person benefiting are reliably the same person, and if they’re not, build the marketing around the paying decision, not the end-use experience.
How to Actually Apply This
Look for small, easy-to-dismiss signals before you have a full pattern. A single odd request or offhand comment isn’t proof of anything on its own, but it’s worth writing down rather than ignoring, because a pattern often only becomes visible in hindsight, once enough small signals accumulate.
Separate who’s paying from who’s using, explicitly. Ask yourself directly whether your product’s end user and your actual buyer are the same person, because marketing built for the wrong one of those two roles will systematically underperform no matter how good the product is.
Require independence before trusting a pattern. A theory should be supported by multiple customers who don’t know each other and aren’t influencing one another’s behavior, not by a handful of comments in the same thread or the same social circle.
Test the hypothesis cheaply before committing fully. A single reframed listing, a small ad set, or one direct outreach message can validate or disprove a theory far more cheaply than rebuilding an entire brand identity around a guess that turns out to be wrong.
Be willing to widen or narrow the hypothesis as evidence accumulates. A pattern that looks like “wedding shoppers” might really be “milestone gift-givers” once more evidence comes in, and staying too attached to the first specific version of the theory can cause you to miss the broader, more accurate one.
Conclusion
Not knowing your target customer isn’t fixed by writing a more detailed demographic persona—it’s fixed by paying attention to specific, real behavior: who’s actually buying, what occasion is driving the purchase, what they’re really trying to accomplish, and whether the evidence for that theory holds up across more than one person before you act on it. The woodworker didn’t stumble onto a single perfect insight. He noticed a few small, ambiguous signals, tested a cheap hypothesis instead of committing blind, and only rebuilt his strategy once real behavior confirmed what a handful of scattered comments had only suggested.
FAQ
What if I only have one or two data points suggesting a pattern? Treat it as worth investigating, not worth acting on fully yet. Look for the same signal to show up independently from customers who don’t know each other before committing significant time or budget to a new direction.
How do I test a customer hypothesis without much budget? Change one variable at a time and keep it small—one product listing reframed around the new theory, a modest ad set targeted at the hypothesized buyer, or a handful of direct outreach messages. The goal is a cheap, fast signal, not a full campaign.
What if my buyer and my end user really are different people? Build your marketing language, imagery, and offer around the buyer’s decision-making priorities, not the end user’s day-to-day experience with the product, since the buyer is the one actually completing the purchase.
How do I know if two different occasions actually represent the same underlying customer segment? Look past the specific event to the underlying intent. If a wedding gift and a housewarming gift both reflect the same motivation—a meaningful gift for someone else’s milestone—they may belong to the same broader segment even though the occasions differ.
Is demographic data useless in this process? No, but it usually answers a narrower question than occasion and intent do. It’s most useful after you already have a working hypothesis to help you find more people who fit the pattern, rather than as the starting point for discovering the pattern itself.
Where can I find more structured guidance on identifying a target market? The U.S. Small Business Administration’s guide on market research and competitive analysis covers both direct and demographic research methods for small businesses working through this from scratch.
About the author: This piece walks through a realistic small-business marketing scenario built to illustrate a specific method—Buyer, Occasion, Intent, Evidence, Hypothesis, Test—rather than reporting on a documented real business. No affiliate links, no sponsored placements.
