Market Research for Customer Identification: Stop Guessing

Market research for identifying customers is the process of gathering and analysing data to determine who is most likely to buy your product, why they would buy it, and what they actually need from it. Done properly, it replaces assumptions with evidence and gives product marketing a foundation that holds up under pressure.

Most companies skip this step, or they do a version of it that confirms what they already believe. That is a different thing entirely.

I have been in rooms where the entire go-to-market strategy was built on a buyer persona that one person in product wrote on a Friday afternoon. No interviews, no data, no external validation. Just a confident internal narrative dressed up with a stock photo and a name like “Enterprise Emma.” Those launches tend to be expensive lessons.

Why Most Customer Identification Starts in the Wrong Place

The instinct is to start with the product and work outward. You have built something, so you describe it, then you ask who would want it. That sequence feels logical, but it produces a distorted picture.

When you start with the product, every research question is unconsciously framed to justify it. You ask “who would find this useful?” rather than “what problem are people actually trying to solve, and how are they solving it today?” The first question leads you to people who can tolerate your product. The second leads you to people who need it.

Early in my agency career I worked with a software company that had spent eighteen months building a project management tool for creative teams. Their assumption was that the buyer was the Creative Director. Every message, every sales deck, every piece of positioning was aimed at that person. When we actually interviewed buyers across a sample of their existing customers, we found that in most cases the purchase decision was made by an Operations Manager or a Head of Studio, not the Creative Director. The Creative Director was an influencer, not the economic buyer. That distinction changed everything: the channel strategy, the sales conversation, the pricing framing, the content. They had been marketing to the wrong person for over a year.

This is not an unusual story. It is a common one. And it is almost always the result of skipping or shortcutting the research that would have surfaced the answer before the money was spent.

If you are building out a product marketing function or sharpening an existing one, the Product Marketing hub at The Marketing Juice covers the full range of disciplines involved, from positioning and messaging to launch strategy and competitive intelligence.

What Types of Research Actually Identify Customers?

There are four research types that do real work in customer identification. They are not interchangeable. Each answers a different question, and the mistake most teams make is using only one.

Qualitative interviews

Interviews with current customers, lost customers, and non-buyers give you the texture that no dataset can. You are listening for the language people use to describe their problem, the alternatives they considered, and the moment they decided to buy or not to buy. That language is not just insight, it is copy. When someone tells you “I was drowning in spreadsheets and nothing talked to each other,” that phrase is worth more than any tagline a creative team could invent.

The critical discipline here is talking to people who did not buy. Most teams only interview customers, which creates a selection bias. The people who chose a competitor, or who decided to do nothing, are carrying the most important information you do not have.

Behavioural data analysis

If you have existing customers, your CRM, your product analytics, and your support data contain patterns that tell you who actually gets value from what you sell. Look at which customer segments have the lowest churn, the highest expansion revenue, and the shortest time to first value. Those are your real customers, not the ones you assumed you were selling to.

I spent several years managing large performance marketing accounts, and one consistent finding was that the highest-converting audience segments were rarely the ones the client had originally briefed us to target. The data had a different view. The clients who were willing to follow the data tended to outperform the ones who insisted on their original assumptions.

Surveys at scale

Surveys give you volume and statistical weight, but they are only as good as the questions. Closed questions confirm hypotheses. Open questions generate them. A well-designed survey for customer identification should include open-ended questions about problems, current solutions, and decision criteria, not just demographic fields and satisfaction scores.

The Crazy Egg guide to buyer personas covers practical approaches to structuring the questions that surface genuine buyer motivation rather than polite responses.

Secondary and market research

Published research, industry reports, analyst data, and competitor analysis give you the market context that primary research cannot efficiently generate on its own. Sources like Forrester’s product marketing research can surface structural patterns in how buyers in specific sectors make decisions, which is useful framing before you design your own primary research.

Secondary research is best used to narrow your hypotheses before you test them, not to replace the primary work.

How to Structure a Customer Identification Research Programme

There is a sequencing that works. It is not complicated, but it requires discipline to follow when internal pressure is pushing you to skip ahead.

Step 1: Define the question you are actually trying to answer. “Who is our customer?” is too broad. “Which segment of the market has the highest unmet need for what we do, and what does the buying decision look like in that segment?” is a question you can research. Start with specificity.

Step 2: Audit what you already know. Before you commission new research, look at what your existing data already shows. CRM data, support tickets, sales call recordings, win/loss notes. Most businesses have more signal than they realise, and it is cheaper to mine what you have than to go and collect new data you will not use properly.

Step 3: Run qualitative interviews to generate hypotheses. Talk to ten to fifteen people across different segments: current customers, churned customers, and people who evaluated you but chose not to buy. You are not trying to prove anything at this stage. You are listening for patterns in language, motivation, and friction.

Step 4: Test hypotheses at scale with a survey. Once you have a set of hypotheses from the qualitative phase, design a survey that tests them across a larger sample. This is where you get statistical confidence. Keep it short, keep it focused, and include at least two open-ended questions.

Step 5: Layer in behavioural and secondary data. Cross-reference your primary findings with your internal behavioural data and any relevant secondary sources. Look for confirmation and contradiction. Contradiction is more useful.

Step 6: Build a customer profile that includes decision context, not just demographics. The output of this process should not be a persona with a job title and a stock photo. It should be a clear description of the problem context, the buying trigger, the decision criteria, the alternatives considered, and the objections to be overcome. That is what product marketing needs to do its job.

The Difference Between a Demographic Profile and a Buying Profile

Demographic profiles tell you who someone is. Buying profiles tell you whether they will buy, and under what conditions. These are not the same thing, and conflating them is one of the most common errors in customer research.

A 45-year-old Head of Marketing at a mid-sized B2B technology company is a demographic profile. It is useful for targeting. But it tells you nothing about whether that person is currently experiencing the problem your product solves, whether they have budget authority, whether they have tried alternatives and failed, or whether they are even aware that a solution like yours exists.

The buying profile answers those questions. It describes the situation a person needs to be in before they become a real buyer, not just a potential one. In my experience judging the Effie Awards, the campaigns that consistently won on effectiveness were the ones that had clearly identified the moment of purchase readiness, not just the audience category. They were talking to people in the right situation, not just people with the right job title.

This is why value proposition development is so tightly connected to customer identification. You cannot write a compelling value proposition without knowing the specific situation your buyer is in when they start looking for a solution.

Common Research Mistakes That Produce Bad Customer Data

There are several failure modes in this kind of research that I have seen repeatedly across different industries and company sizes.

Sampling only your best customers. Your best customers are not representative of your addressable market. They are the people who already understood your product well enough to get value from it, who had the right context, and who were willing to invest the effort. Researching only them tells you how to serve people who are already sold. It does not tell you how to reach people who are not.

Asking leading questions. “How valuable did you find the reporting feature?” is a leading question. “Walk me through how you measure the outcomes of this kind of tool” is not. The first produces confirmation. The second produces information.

Letting internal stakeholders define the research brief. When the sales team defines the research questions, you tend to get research that validates the sales team’s assumptions. When product defines it, you get research that validates product’s assumptions. The brief should be written by someone whose job is to find the truth, not to confirm a position.

Treating research as a one-time event. Customer identification is not a project with a start and an end date. Markets move. Buyers change roles. New competitors change the decision context. The research you did eighteen months ago may be meaningfully out of date. Build a lightweight ongoing programme rather than a periodic big exercise.

Confusing engagement with intent. Someone who reads your content, attends your webinar, and follows you on LinkedIn is not necessarily a buyer. Engagement data tells you about interest. It does not tell you about buying intent. These are different signals and they should be treated differently in your analysis.

How Customer Research Connects to Pricing and Positioning

Customer identification research does not exist in isolation. The findings should feed directly into your positioning, your messaging, and your pricing strategy. If they do not, the research has not been operationalised properly.

On pricing specifically, understanding who your customer is and what alternatives they are comparing you against is foundational. If your evidence suggests that buyers are comparing you to a manual process rather than a competing product, your pricing logic is completely different than if they are comparing you to an established market leader. HubSpot’s analysis of AI pricing strategy illustrates how deeply customer context shapes what pricing models are viable, which applies well beyond AI products.

On positioning, the language your customers use to describe their problem should appear in your positioning. Not a polished version of it. The actual language. I have seen too many positioning documents written in the language of the company rather than the language of the buyer. They sound impressive internally and mean nothing externally.

If you are working through how to build the full product marketing stack, from customer research through to launch execution, the Product Marketing section of The Marketing Juice covers each component in depth.

What Good Customer Identification Research Actually Produces

The output of a well-run customer identification research programme is not a persona document. It is a set of answers to specific questions that the business needs to make good decisions.

Those questions include: Who experiences the problem we solve, and in what context? What triggers them to start looking for a solution? What alternatives do they consider, and why do they choose between them? Who else is involved in the decision, and what do they care about? What would make them choose us over the alternative? What would make them choose the alternative over us?

When a business can answer those questions with evidence rather than assumption, product marketing becomes significantly more effective. The messaging is sharper because it is grounded in real buyer language. The channel strategy is more precise because you know where buyers go when they start looking. The sales enablement is more useful because it addresses the objections buyers actually raise, not the ones salespeople assume they raise.

There is a broader point here too. Marketing is often asked to do a lot of heavy lifting to compensate for product or positioning problems that research would have surfaced earlier. I have run agencies for long enough to know that a significant proportion of marketing spend goes into amplifying messages that do not resonate, to audiences that are not quite right, for products that have not been positioned against what buyers actually care about. Good customer identification research is not just a product marketing tool. It is a commercial efficiency tool. It reduces the amount of money you need to spend to get results.

For teams preparing a product launch and looking to connect customer research to execution, Later’s product launch checklist is a useful operational reference for translating research findings into channel activity.

And for the creator and SMB side of the market, where customer identification often gets skipped entirely in favour of instinct, Buffer’s work on creator pricing strategy shows how understanding your audience’s economic context directly shapes what you can charge and how you should position an offer.

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