The descriptive ego: why customers explain their choices but don't know them

Ask a buyer why they chose your product and the answer comes back with confidence: “good camera”, “convenient”, “I trust the brand”. The answer is sincere. It is also, almost certainly, wrong. The part of the mind that produces it is called the descriptive ego. It is the narrator inside us that takes something that has already happened and wraps it in words. It does not lie. It explains what it never properly saw.

Picture someone on the bank of a muddy river. They look at the surface: ripples, foam, a floating branch. Smooth, so it must be calm. Foam, so it must be rough. The riverbed, the rocks, the deep currents are invisible to them. The narrator in our heads works the same way. It watches the top layer of the mind and stitches those fragments into a tidy story about “why I did that”.

For a business this is money, not philosophy. Every survey, every focus group, every exit interview hits the same wall: you receive a well-formed guess the customer made about themselves, and you write it down as a cause. For a company selling into Russia from abroad the wall is taller, because the guess arrives through an agency, a translator and a summary slide, and each step makes it look more like a fact.

Why an explanation is not the same as understanding

Between “I can explain what I did” and “I know the real reason I did it” lies a gap. The narrator bridges it in three ways, and each one adds distortion.

First, it sees only what reached awareness. Most of a choice is settled earlier and deeper, without words. The hand reaches for the familiar pack on the shelf automatically, in a fraction of a second. The body has already chosen. Only afterwards does the narrator switch on and compose a respectable version: “I take this brand because I trust the quality”. No. You took it because your hand followed its usual route. The narrator has no access to that level, so it honestly offers the only explanation within its reach.

Second, it cannot tolerate gaps. Where the real cause is out of reach, the narrator inserts a plausible filler. A fact that gets in the way of a coherent story is quietly rewritten or skipped. A connected account is worth more to it than an accurate one with holes.

Third, it works in words, and words are already a filter. The sense that “something is off” is rich and many-sided. The sentence “I’m not comfortable with it” is flat and thin. Part of every experience is lost on the way into language.

Stack the three layers and you get a person sincerely telling you a story that may have almost nothing in common with the process that produced their choice. And they believe it.

A man picks up an expensive coffee in a handsome tin and, asked by his wife, says “it tastes better”. Perhaps. But the tin that will look good on the kitchen shelf, or the wish to be the kind of person who buys good coffee, are just as likely to have decided it. The narrator names taste because taste is a socially approved reason.

Why surveys and focus groups mislead even when people are honest

You seat eight people in a room, show them the packaging, the new flavour, the ad mock-up, and ask: “would you buy it?” They answer “yes, interesting”, “I like the colour”, “handy bottle”. Six months later the product reaches the shelf and fails.

The participants did not lie. On a focus group it is their narrator who answers, and it constructs “I would buy it” from whatever lies on the surface: the wish to be polite, the mood in the room. A real purchase in a real shop is governed by levels the narrator cannot reach.

Hence a simple and unpleasant rule. A customer’s answer can be completely honest and completely wrong at the same time. Asking “why” is asking a person to guess about themselves. Questions about the future are the worst of all. “Would you buy this at 3,000 rubles?” The narrator will happily say yes: yes is more pleasant than admitting to being stingy. The wallet at the till answers differently. Intention and action live on different floors, and a survey only reaches the top one.

In Russia the top floor has an extra coat of paint on it. A stranger with a questionnaire is still, for many respondents, someone to be handled rather than confided in. Foreign companies that read a warm summary in English from a Moscow agency tend to underestimate how much of that warmth is courtesy, and how much of the rest was flattened by translation.

None of this makes surveys useless. It means their output has to be read differently: not “here are the reasons people buy” but “here are the stories people tell about themselves”. The story shows which words a customer leans on to justify a choice. Those words belong in your ads and on your site: for a company writing Russian copy from abroad they are the one thing a translator cannot invent. What the story cannot be taken for is the engine of behaviour.

What to watch instead of what to hear

The practical conclusion is short. Do not trust what people say. Watch what they do.

Replace “why did you buy” with “what did this person do before, during and after the purchase”. Behaviour passes through every floor of the psyche at once. That is why it is more honest than words.

Watch which screen people linger on and which they leave in two seconds. Where in a conversation the pace changes and pauses appear. What goes into the basket and what gets taken out again. These are signals from the levels the narrator cannot enter.

Here the Russian market is unusually generous. Yandex Metrica ships session replay, click maps and scroll maps free on every counter, a toolset Google Analytics users normally pay a separate vendor for. A foreign marketer can sit in another country and watch a Russian visitor’s cursor hover over the price, scroll back to the delivery terms and leave.

A good marketer behaves like a tracker rather than an interviewer. Instead of asking “do you like our advertising”, they run two variants and look at clicks, time on page and inquiries. Numbers do not know how to be polite. They show where a living person went, on their own feet and with their own money. This is why Yandex Ads tests are a research tool for me before they are a sales tool.

The same applies to product work. A recorded session of a person fumbling through an interface is worth ten surveys asking “is it convenient”. In the survey they say “convenient”. In the recording you see them miss the button three times and swear.

Take an online-store checkout. Ask ten shoppers why they did not complete the order and you hear a steady chorus: “too expensive”, “changed my mind”, “found it cheaper”. The analytics show something else: people leave at the step where the shop suddenly demands registration, or where a delivery fee nobody mentioned gets added. The narrator did not record the moment trust disappeared and substituted the handiest explanation: expensive.

For a foreign brand this substitution has a local flavour. A Russian visitor abandoning a checkout may have hit a card form that rejects a domestic card, a delivery block with no pickup point, or a phone field that refuses a +7 number. Ask them afterwards and they will say “expensive”. The recording shows the field they fought with.

Discounts are even more telling. Ask a customer whether free delivery above a threshold affects them and they wave it off: “no, I buy what I need”. The data say the average order crept up to exactly that threshold. The behaviour changed. The account of it stayed the same.

If you cannot avoid asking

Sometimes there is nothing to observe. The product does not exist yet, and a decision has to be made. Then a survey is the only instrument left. Use it, knowing that you are holding a drawing made by someone who saw only the riverbank. You can steer roughly by it. You cannot trust it in detail.

A few techniques make a survey more honest. Ask about the past rather than the future: “tell me how you last chose a product like this” instead of “would you buy”. Ask for facts rather than ratings: “what exactly did you do next” instead of “do you like it”. The more concrete the question, the less room the narrator has for fiction.

For interviews run in Russia from abroad, two more corrections. Brief the interviewer to leave silence after an answer: the second sentence a Russian respondent offers is usually closer to the truth than the first. And read the transcript, never the summary. A summary is written by another narrator.

Keep this in mind whenever you build positioning. A formulation that a customer approved in an interview may have no bearing on whether their hand reaches for your shelf. Test formulations with behaviour, not with applause in a meeting room.

The most valuable signal hides in the mismatch

Keep both channels open, what a person says and what they do, because the most valuable discovery happens exactly where the two disagree.

The customer says “the price is fine” and refuses at the till. Says “the site is convenient” and abandons the basket at step three. Says “sustainability matters to me” and buys whatever is cheaper. Each mismatch is a signpost rather than an error in the data. In the gap between words and actions hides the real motive, the one the narrator could not reach. Finding that gap is where customer research starts: what your customers tell you laid next to what they actually do.

For a company entering Russia the mismatch is the first thing worth paying for, before a single ad runs. The survey commissioned before launch says one thing. The Metrica recordings from the first month say another. The difference between them is the shortest description of the Russian market you will get: it names the place where the story you were sold and the customer you actually have part ways.

Words show who a person wants to be seen as. Behaviour shows who they are. Listen to the first to learn their language. Trust the second to learn their choice.

A customer will never lie to you with their feet.

Frequently asked questions

Sources

• Russian version of this article on belokrylovo.ru: Дескриптивное эго: объясняет, но не понимает