Why a brand questionnaire can’t reach you

The document arrives as a shared link. Describe your brand in three words. Who is your ideal client. What makes you different from your competitors. What are your core values. You set aside a proper block of time and fill it in honestly, because you want to get something out of this.

The answers come out fluent. You read them back and something is off, and you cannot say what.

A brand questionnaire is a self-report instrument. It assumes the answer is sitting inside you, and that the way to get at it is to look inward and write it down. The research on introspection says inward is the least reliable place to look for that particular answer, and that the harder you look, the further you can end up from it.

In this article

• What a brand questionnaire is actually asking you to do
• People explain choices they never made
• Analysing your reasons changes your answer
• What reaches it instead
• Common questions

What a brand questionnaire is actually asking you to do

In 1977 Richard Nisbett and Timothy Wilson published a review of the evidence on whether people can report accurately on their own mental processes. Their conclusion was that there may be little or no direct introspective access to the higher-order processes that produce preferences and choices.

What people do instead, when asked why, is generate a plausible theory. The report comes from an implicit sense of what would be a reasonable cause of a response like that, rather than from any observation of the process itself.

Nobody experiences this as making something up. There is no felt difference between a theory produced this way and a memory, because from the inside there is nothing available to tell them apart.

Which is what the questionnaire collects. Ask what makes a business different and you get a plausible theory about what would make a business like that different. We are premium but approachable. Our clients value quality over price. We care more about the outcome than the invoice. None of those is false exactly. They are the sort of thing that is true of a whole category, produced by someone asked to account for themselves who has obliged. It is the borrowed voice problem arriving before a single word of copy has been written.

This lands harder on solopreneurs than on anyone else, for a reason that has nothing to do with ability. A company of forty has people who can describe the business from outside their own head. Somebody in sales knows which line makes buyers lean in. Somebody who left last year said something on their way out that stuck. When you work alone, every answer on that form comes from one source, and that source is also the subject. There is no second reading to check it against, which is a large part of why the About page is the hardest page to write.

People explain choices they never made

The demonstration of this is not subtle.

In 2005 Petter Johansson and colleagues showed 120 people pairs of photographs of faces and asked them to pick the one they found more attractive. On some trials a card trick swapped the pictures, so the face handed back was the one they had just rejected. They were then asked to explain their choice.

Across 354 manipulated trials, 46 were spotted at the time. Thirteen per cent. Counting every form of detection the researchers could find, including a post-experiment interview that walked people through exactly what had been done to them, no more than 26 per cent of manipulated trials were ever exposed. This was not a matter of the faces looking alike. Detection rates were no better for the set chosen to be obviously dissimilar.

The explanations are the part that should stop you. Independent raters scored every report for emotional engagement, level of detail and confidence. Across all three, the explanations people gave for choices they had genuinely made were indistinguishable from the explanations they gave for choices that had been swapped underneath them. Some reports referred to features only the swapped face had, which means a participant explained a preference for dark hair while holding a photograph of the blonde woman they had turned down.

Then the detail that makes it personal. Of the participants who noticed nothing whatsoever, 84 per cent said, when asked, that they were confident they would have noticed.

That last figure is the one to carry into your own answers. The machinery that produces reasons runs whenever it is asked to run. It does not sit idle when there is nothing behind the question, and it does not flag the difference. You get a fluent answer because you asked a question, and the fluency tells you nothing about whether anything was there.

Analysing your reasons changes your answer

A form that failed to reach you would be a waste of an afternoon. The finding underneath this one is worse than that, because the exercise moves you.

In 1991 Timothy Wilson and Jonathan Schooler ran two studies. In the first, 49 students tasted five strawberry jams and rated them, and their ratings were compared against the assessments of the Consumer Reports tasting panel. Students asked to analyse why they felt the way they did agreed with the expert panel less than students who simply tasted and rated.

In the second, 243 students chose university courses. Some were asked to analyse their reasons, others to evaluate every attribute of every course. Both forms of introspection produced choices that corresponded less closely with expert opinion than the control group’s did.

The mechanism the authors describe is precise. Analysing reasons pulls your attention onto the criteria that are easiest to put into words. Those criteria look like plausible causes, and once you have named them you begin deciding on the basis of them. Easy to articulate and actually important are separate properties, and the act of explaining yourself substitutes one for the other without announcing that it has.

A follow-up study made the cost concrete. Participants looked at posters and chose one to take home. Those asked to think about their reasons first chose a different kind of poster from the control group, and when they were contacted three weeks later they were less satisfied with what they had chosen.

Read that against your own site. You filled in the form. Filling it in pulled you towards the parts of your business you could most readily turn into sentences. The positioning got built from those parts, and the copy got built from the positioning, and it went live. You have not liked it since and you have never been able to name the fault, because the fault is not in the copy. It sits in what the copy was made from, which is also why redoing the design does not fix it.

There is a second stage that makes it harder to see. Whoever receives that form has no independent source. Your answers are the only material they hold, so your answers are what gets built. They are professionals, so they will make those answers sound considered. Sentences get tightened, the opening line gets a rhythm to it, the hierarchy on the page starts guiding the eye properly.

And the better that job is done, the harder the result becomes to argue with. A weak input executed badly looks like a mistake and you can point at it. A weak input executed well looks like a finished thing. You sit in front of a site that is competent in every respect you can name, feeling that it is not you, with nothing to point at. Most people conclude at that stage that the problem is their own taste, and pay somebody else to do it again.

What reaches it instead

None of this makes intake forms worthless, and it would be convenient of me to pretend otherwise given that I ask people things too.

A questionnaire is good at facts. What you charge, who your last ten clients were, which service actually pays, what you are contractually unable to talk about, where the enquiries come from. Those are matters of record. You can look them up and you are not the only witness. Ask that sort of question on a form and the form does its job.

The failure is confined to a specific kind of question, and it is the kind every brand questionnaire is mostly made of. What makes you different. What do you stand for. Who are you really for. Those ask you to report on the thing the research says you have the least direct access to, and they ask it in the format most likely to distort the answer.

Wilson spent a career on this and set out the practical version in Strangers to Ourselves. If you want to know what you are like, look at what you actually do, and at what other people see in you.

Neither of those is a form. Both need something outside your own account of yourself.

What you actually do is already documented. It is in which enquiries you answer within the hour and which sit unopened for days. It is in the projects you extended without invoicing and the ones you were relieved to finish. It is in what you turned down last year, which is usually more revealing than anything you accepted. None of that requires introspection, because it is evidence rather than recall.

That evidence has one property your answers do not. It was produced when you were not being asked to describe yourself, which means none of the distortion in this article had a chance to operate on it. Nobody performs for their own calendar.

What other people see in you needs somebody in the room. In practice that looks like being asked about one particular client rather than an ideal one. Being asked why that project and not the other one. Being asked again when the first answer turns out to be the category answer, and again after that. It looks like somebody registering the difference between how you talk in the first ten minutes and how you talk in the fortieth, when you have stopped presenting and started explaining.

The difference is in what happens to your first answer. A form records it and moves on. A conversation treats it as an opening position and keeps going until something surfaces that you had not prepared.

If your site is already built off a form, none of this means starting again. It means gathering the evidence you have been generating anyway. Go and find an email you sent to a client you liked, explaining what you were going to do for them and why, written quickly and with no thought of anyone else reading it. Put it beside your services page. The email will be more specific, more confident about what you are not going to do, and considerably more like a person. That distance is the whole problem, and it is measurable in your own sent folder.

Which is the honest answer to the obvious objection about the Brand Mirror, since it is eight questions and you answer them yourself. It does not ask you to state your positioning and it does not score you. It takes your answers and puts them next to the evidence, which is your own website, and the only instruction at the end is to go and look. The gap between the two is the finding. Your answers are not being trusted as a report. They are being used as one half of a comparison.

The questionnaire is not useless. It is an accurate record of what you can currently say about your business. If what you can say were the whole of it, you would not be sitting there with a document you filled in properly and a website that still does not sound like you.

Common questions

Why do brand questionnaires produce generic answers?

Because they ask you to report on processes you have limited introspective access to. Nisbett and Wilson established in 1977 that when people explain their own preferences they generate a plausible theory rather than observing the cause. What comes back is what would plausibly be true of any business in your category.

Does filling in a brand questionnaire change what I think?

Yes. Wilson and Schooler found that analysing your reasons pulls attention onto criteria that are easy to put into words, and that this produced choices matching expert judgement less closely than not analysing at all. A follow-up study found people who analysed their reasons before choosing a poster were less satisfied with it three weeks later.

Are brand questionnaires useless?

No. They work well for matters of record: what you charge, who your clients have been, which service is actually profitable, where enquiries come from. The failure is confined to identity questions such as what makes you different and what you stand for, which is unfortunately most of what a brand questionnaire contains.

What works better than a brand questionnaire?

Evidence and an outside observer. Timothy Wilson’s conclusion is to look at what you actually do and at what other people see in you. In practice that means examining decisions already made, including what you turned down, and being asked follow-up questions you had not prepared for.