Ethnography can sound like one of the more academic corners of market research. In practice, the principle is remarkably simple: if you want to understand how customers behave, you sometimes need to stop asking them and start watching them.
Most market research relies, in some form, on asking people questions. What matters to you? Why did you choose that product? How do you normally make this decision? What would make you switch? And while the answers to these questions are informative, there is one fairly substantial problem with this - what people say they do and what they actually do are not always the same thing.
This isn’t because customers are deliberately misleading us. Even our own behaviour can sometimes be difficult to recall and even harder to explain. Much of what we do is habitual, contextual or so mundane that we barely notice ourselves doing it. Self-reported research also has well-documented limitations. Recall can be imperfect and answers can be influenced by what researchers call ‘social desirability bias (SDB)’ – our tendency to give responses that present us in a more favourable or socially acceptable light. A 2026 systematic review of methods designed to reduce this bias, covering 121 experiments across 79 peer-reviewed papers, describes SDB as a serious challenge to the validity of survey research.
Ethnography helps reduce the distance between researchers and behaviour by adding direct observation and context to market research. It is a qualitative research approach centred on observing people in the environments in which behaviours actually take place, often supplemented by conversations, interviews, fieldnotes and other forms of contextual evidence. Instead of simply asking someone how they shop, work, cook, choose a supplier or use a product, researchers get closer to the real experience.
Research published in the Journal of Marketing Research found that market-oriented ethnography can generate strategically important insights with implications for market segmentation and targeting; product and service positioning; and product, service and brand management.
So where does ethnography fit in a marketing plan?
A strong marketing plan starts with diagnosis before moving into strategy, followed by the tactical elements of product, price, distribution and communications. Ethnography sits primarily within the diagnosis phase, specifically within market research (covered in Module 2 of the MiniMBA in Marketing) but the insight it produces has the potential to influence almost everything that follows.
It is particularly useful early in a research programme when marketers do not yet fully understand customer behaviour. Those observations can reveal patterns, meanings and questions that interviews, surveys, sales data and other forms of research can subsequently explore or test.
While a proper ethnographic study can become sophisticated, marketers do not need to become anthropologists to understand why the method is valuable. Simply put, the quality of everything downstream depends on the quality of the research and diagnosis upstream.
There is no single, universally accepted five-stage process for conducting ethnography. But for marketers, the following five stages provide a practical way to understand and apply its core principles within a wider market research programme.
Stage 1: Decide what you need to understand
The ethnographic process doesn’t begin with the observations that make it such an important part of the research phase. It begins with the decision the research needs to inform – drilling down into what you need to understand and, crucially, why you need to understand it.
This follows a broader principle of good market research: begin with the decision rather than simply producing a list of questions you would like answered. A MiniMBA article on backwards market research makes the same distinction – research should ultimately improve a marketing decision rather than simply produce interesting information.
This stage also means thinking about who needs to be observed, in which context and for how long, alongside practical considerations such as access and informed consent. The objective is not simply to find people to watch, but to design the research around the behaviour and decision you need to understand.
Take these common examples: customers are abandoning purchases, and it is not clear why. A product is being repurposed for a task it was not designed to perform, despite other products existing specifically for that job. Buyers keep choosing an established competitor despite claiming your proposition is better. Or people are simply uncertain about how a product fits into their daily lives while secondary research tells a different story.
Starting with a rationale for conducting the research keeps it focused. If there is no clear rationale to begin with, observing customers can produce plenty of interesting material without necessarily producing anything useful.
Be clear about what you want to study and which decision a better understanding of it will help you make.
Stage 2: Get close to the real behaviour
Depending on the market, this observation stage might mean spending time in someone's home, accompanying them while shopping, watching people use a product, observing a B2B team completing a task or following part of the customer journey. Whatever you are observing, it must be linked to what you are trying to understand and studied within the context in which the behaviour normally occurs.
You want the behaviour to be as close to natural as possible. This is one of the defining characteristics of ethnographic research: behaviour is studied in everyday settings rather than deliberately constructed experimental environments.
Observation is central, but it does not have to be the only source of evidence. Researchers can record detailed fieldnotes, talk to participants about what they are doing, conduct interviews and examine relevant objects, documents or other materials. Combining these different forms of evidence can provide a richer understanding of what is happening and why.
What we are trying to do here is uncover context that other forms of research may miss or struggle to explore deeply enough. A survey can tell you that convenience matters to a consumer, but ‘convenience’ has a number of possible meanings. Ethnography might reveal that convenience means being able to do two things simultaneously; it can also mean speed of service, proximity to a location or automation while purchasing.
Importantly, that does not make the survey wrong. It gives you a better understanding of what the answer might actually mean.
Context changes our understanding of behaviour.
Stage 3: Look for patterns, contradictions and workarounds
Observation gives you the raw material. Analysis gives it meaning. But the two do not necessarily happen one after the other. Ethnographic research is often iterative: researchers record and analyse what they are seeing while the fieldwork is still taking place, allowing early observations to influence what they investigate next.
The real work comes from asking what these observations mean within the context of your original rationale and research requirements.
There are plenty of observations to be made, and any preconceived ideas of what ‘might’ be happening should be held lightly rather than allowed to determine what the researcher looks for. Otherwise, confirmation bias risks turning observation into a search for evidence that supports what the business already believes.
Researchers might look for repeated behaviours, unusual shortcuts, frustrations, compromises, language, rituals or contradictions between what participants say and what they subsequently do. Detailed fieldnotes are particularly important here because they allow researchers to capture not only what happened, but also the setting and context surrounding it.
If customers repeatedly create their own workflows, ignore a new product feature, combine two products or develop an unofficial shortcut, there may be an unmet need hiding in plain sight. These ‘workarounds’ can be particularly revealing, although it is important to remember that a single observation or insight proves very little.
Looking for repeated behaviours, differences and patterns that participants may not even recognise themselves helps researchers build a richer interpretation of what is happening.
Stage 4: Turn observation into hypotheses
Ethnography helps us uncover possibilities that research from a distance may struggle to reveal because it places behaviour back into its wider context. That does not make surveys, interviews or focus groups weaker forms of research. They answer different questions.
Crucially, ethnography doesn't automatically prove that an observed behaviour applies across an entire market. But neither should its value be reduced simply to generating something for a survey to test.
The first job is interpretation: understanding what the observed behaviour might mean, why it is happening and how different pieces of evidence relate to one another. From those interpretations, marketers may also develop hypotheses worth exploring through further research.
Observations may reveal that several customers struggle with the same part of a buying process. That does not mean 63% of the market has that problem. It does mean you have identified something potentially important.
This is where ethnography works particularly well alongside other research methods. Interviews might explore the issue further. Quantitative research might establish how prevalent it is. Behavioural or sales data might show whether it appears in the wider customer base.
One useful research sequence is therefore to use ethnography or other qualitative techniques to explore an issue and identify the behaviours, language and variables that matter before using quantitative research to establish their scale.
This can also make subsequent research sharper. Rather than asking consumers about everything that might matter, marketers can design surveys around factors that earlier qualitative research suggests actually do.
Ethnography can reveal things worth understanding and measuring. Further qualitative research can deepen our understanding of them; quantitative research can help establish their scale.
Stage 5: Turn insight into marketing decisions
An ethnographic insight might influence a product feature, proposition, customer journey, distribution decision, communication brief or even the way a market is segmented.
As an example of how customer insight can influence decisions well beyond a research report, IDEO's work with Sealy used consumer insights to influence product design, brand principles, manufacturing decisions and sales and marketing strategy, eventually contributing to an overhaul of 85% of the company's mattress portfolio.
The output should therefore move through three simple questions:
- What did we see?
- What does it mean?
- What should we do about it?
Without that final question, ethnography risks becoming interesting observation rather than useful market research. The point is not simply to know the customer better. It is to make a better marketing decision because of what you now know.
Sometimes, just go and look
Marketers have access to more customer data than ever. But having more data does not necessarily mean being closer to customers. Ethnography adds something different: context. It puts marketers closer to the situations in which choices are actually made, compromises happen, and products are really used.
Sometimes the simplest way to understand your customer better is also the most obvious.
Go and look.
Cover: Unsplash / Richard Stachman / Roma