Brand quality research: focus groups and in-depth interviews

Errors that turn focus groups into useless rituals

The most common mistake is testing packaging or a product without competitive context. Participants are shown one sample and begin to evaluate it as an art object: 'pretty,' 'I don't like the color.' As a shelf-selling tool next to a dozen similar products, packaging is not evaluated at all — and that is exactly what the test should aim for.

A second typical problem is related to neurophysiology of perception, not real preferences. If three design options are laid out on a table or screen, the central one is chosen significantly more often — simply because it is in the center of attention. This is a positional effect, not an assessment of idea quality, and it must be controlled by rotating the placement of samples between sessions.

A third error is taking the first reaction of rejection to a new product as a diagnosis of 'the product is bad.' The brain resists the unfamiliar by default, this is a protective mechanism, not a signal of concept failure. If participants frown at the first display of a bold rebranding — this is a normal reaction to novelty, and one should look at how their attitude changes after repeated contact, not rely on the reaction from the first second.

How to uncover what a person hides even from themselves

People systematically rationalize their behavior and give socially desirable answers — they say they choose a product 'for quality,' although the real trigger for purchase is price or packaging, which is awkward to admit as the decisive factor. The direct question 'why did you buy this' almost always gives a beautiful but inaccurate answer.

Projection techniques are used — they relieve the pressure of the 'correct answer', because the question is not asked about the participant themselves, but about an abstract 'person' or image.

• Verbal associations — what are the first words that come to mind when seeing a brand or packaging
• Incomplete sentences — 'The person who buys this product usually...'
• Product assortment matrix — how the entire product line is perceived as a whole, whether individual products are lost in the background of others
• Metaphors and role-playing — 'If this packaging were a person, who would it be and why'

These techniques provide what a direct question never would: an unedited first reaction, which is closer to real motivation than a carefully thought-out socially acceptable answer.

How many people to invite and how much it costs

The standard focus group size — 6–8 people; many practices consider the optimal size to be around 8 people. With 4–5 participants, the discussion becomes weak, replies repeat, and the group quickly exhausts itself. With 11–12 and more, manageability drops sharply: some people simply don't have time to speak up, the conversation breaks into side dialogues. There is an counterintuitive point — increasing the group from 8 to 12 people does not add new insights, but the manageability of the discussion deteriorates significantly. The quality of participant selection is more important than quantity: a well-selected group of eight will yield more than a random dozen.

A moderator distorts the results — even when silent

Gender, age, intonation, and even nods from the moderator influence what participants say — people adjust to the expectations of the facilitator, often unconsciously. It is impossible to completely eliminate this influence, but it can be minimized.

1. To understand one's own expectations and hypotheses before the session begins — to write them down, to distinguish fact from confirmation of one's own theory
2. To record the session video and then analyze one's own behavior — pauses, reactions, non-verbal signals
3. Work according to a detailed guide with pre-formulated neutral formulations of questions
4. Conduct several sessions with different moderators and compare results
5. At the end of the session, directly ask participants whether everyone was able to express their opinion — this reveals dominance of some voices over others

Data analysis: from impressions to verifiable insights

The weak point of most reports is the subjective interpretation 'at a glance' immediately after the session. A normal methodology requires several levels of processing, and skipping stages is a direct path to distorted conclusions.

• Preliminary report — compiled several hours after the session, from the moderator's memory, suitable only for quick orientation
• Descriptive report — prepared 1–2 weeks, with transcriptions, records what was said
• Full analytical report — takes two weeks, includes coding of statements, theme identification, interpretation, and recommendations

Coding — a key moment that is often done incorrectly. The same phrase often relates to several categories at once, and rigid binding of 'one phrase — one theme' impoverishes the data. It is important to record not only words but also non-verbal signals: pauses, moments of group liveliness, conflicts between participants — often these are the real tension zones around the brand, not the final agreement statement.

Focus group or in-depth interview: different tasks, not competition

Contrasting these two methods is a methodological error that many articles make. They solve different tasks and complement each other well in one research project.

Here it is worth addressing a widespread misconception: a focus group is not representative of the market sample, and it is normal — it is not created for that purpose. The task of qualitative research is not to confirm a percentage, but to understand the logic and the language people use to describe a brand, to find hypotheses that can then be verified quantitatively. To demand that a group of eight people be representative of the entire audience is to misuse the tool incorrectly, rather than to identify its shortcomings.