In kitchens and in workplaces, ethnographic research informants are asked to put up with camera-toting ethnographers. In the process, neighbours come by, children get hungry and raid the refrigerator, and small disagreements break out. Often these are sparked by questions the ethnographer has asked. If the ethnographer is lucky, they are sparked by something that just happened to take place while she was there.

So negotiation and contest about the meaning of daily practices may be played out under the nose and camera lens of the ethnographer. The natural setting is, indeed, the place to be. But this reality, while a nice enough place to visit, seems not to be a place in which most market researchers want to live. Reality is complex; reality seems perilous. Can a set of stationary cameras capture it? Nick suggests they can't and he is correct.

The camera is not enough because only a trained human being can make competent decisions about what to include and what to exclude from view, what to ask about, what to participate in. At the end of the day, only a well-grounded and holistic ethnographer will know what data to compare with which theory and with which prior research to discover the most strategic and relevant implications of the fieldwork.

A stationary camera excludes what is not in the frame. A stationary camera cannot react and shift its gaze. It excludes the past and the future, and humans, through language, live in both worlds, between, as the song says, memory and hope. A human ethnographer can spend time in all three worlds by looking at artefacts and asking about them, by feeling anger or sadness or delight along with someone she is talking with. A stationary camera cannot.

So if reality and the daily practices that constitute it are so damnably complex, why not cut back on the complexity and focus on what matters the most? Hasn't science provided a method to do this?

Ethnographers do indeed sharpen their foci. But knowing when and how to do so doesn't come easy. Cutting out (or carelessly messing about with) the context in the early stages means making the results un-interpretable. I would argue that this is what makes focus groups difficult to interpret beyond a very superficial "they said this and that" level.

When ethnographers do cut out the context, they are doing what the anthropologist Mike Agar calls testing, which they should not do until they and their sponsors are convinced they have done enough learning to know what bits matter the most.

There are plenty of tools for testing cultural patterns, cognitive maps, people's material and political realities. Stationary cameras are among these. Statistical and quantitative approaches do, after all, have an important place in ethnography during the testing moment. But asking the wrong questions produces errors that are undetectable by statistical tests.

Ethnographic teams, with their business counterparts, can see how people come not to do what they say they do. They can learn how daily practices do or do not present opportunities for new understandings and new relationships among producers and consumers. For making and exchanging goods for something of value has always been, and always will be, about human relationships, about much more than dollars and cents, pounds and pence.

We are told, these days, that brands and selling are about relationships. This makes sense. Likewise, new relationships among producers and consumers won't come about by using some sort of spy-cam and calling it ethnography. That's where the real peril is: not engaging completely with our shared, messy reality.

Observers behind a glass (or a stationary camera) are in a very peculiar, even risky relationship with those on the other side of the glass (or lens). Better to leave the camera at home than miss the opportunity to build new relationships among companies and the clients these companies aim to serve.

See also: The joys of ethnography, by Nick Jakel-Elliot