Brand perception interview template for B2B SaaS founders

A brand perception interview shows how customers categorise your product, what they compare it with, when it comes to mind, and which experiences support those impressions. It does not measure market awareness or prove how common an opinion is.

Use the seven-question guide below to collect the language people would use when you are not in the room. Start with an actual buying or usage situation, capture unprompted associations before showing any words of your own, and end by connecting the strongest association to its practical consequence.

The seven-question template

Treat this as a discussion guide, not a questionnaire. Keep the topics stable across a research round, but follow the participant’s language and skip probes they have already answered.

What brand perception interviews can and cannot answer

Nielsen Norman Group distinguishes a company’s intended brand from the way customers interpret it. The company can define a promise and a set of attributes, but customers form meaning from the value they receive and the experiences they have across touchpoints. Those experiences include more than a logo: product behaviour, content, support, sales, and other company actions all contribute.

That makes an interview useful when you need to decide:

• which category or problem your home page should lead with;
• which alternatives belong on a comparison page;
• whether your intended attributes appear in customers’ own language;
• which product or service experiences strengthen or contradict the promise;
• how different groups, such as new customers and recent lost prospects, interpret you.

An interview is not a substitute for unaided-awareness tracking, message testing with a representative sample, conversion data, or a visual-design study. It can reveal interpretations and possible mechanisms. It cannot tell you that “37% of the market sees us as innovative” unless you run an appropriate quantitative study.

Set up the research before writing questions

Choose one decision

“Understand our brand” is too broad. Write the decision that the research will inform:

• Do we need to change the category language in the hero section?
• Are prospects comparing us with the competitors we assume?
• Does the onboarding experience support our promise of simplicity?
• Do customers in two roles describe the product differently?

The decision determines whom you recruit and which parts of the guide matter. If the question is about sales positioning, include recent evaluators. If it is about the experience of using the product, recruit people with enough recent experience to discuss it. Someone who has never heard of you can help with category language, but not with how your product behaves.

Recruit for variation, not praise

A useful round may include current customers, new customers, former customers, won prospects, and lost prospects. Do not place all of them in one undifferentiated sample. Record the participant’s relationship to the product and compare like with like.

There is no evidence-based universal number of brand-perception interviews for a seed-stage SaaS company. Start with the smallest set that covers the important participant types, review what remains uncertain, and add interviews deliberately. The guide to how many users to interview explains why the breadth of the question and the diversity of the audience matter more than a magic number.

Be honest about sponsorship and recording

Tell participants who is conducting or sponsoring the study, what the conversation is for, what will be recorded, and how the data will be used. Do not call the researcher independent or the interview anonymous unless that is true. If revealing the sponsor in advance would undermine a legitimate blinded design, use an approved disclosure and debrief process rather than improvising concealment.

The GOV.UK Service Manual recommends putting the introduction, recording information, topics, and follow-up prompts in a discussion guide, then piloting the guide before the study. A pilot will expose confusing placeholders, compound questions, and a session that is longer than promised.

The brand perception interview questions

1. Reconstruct the last relevant situation

“Tell me about the last time you needed to [category job]. What was happening?”

Useful follow-ups include:

• “What happened just before that?”
• “What did you try first?”
• “Who else was involved?”
• “When, if at all, did [product] enter the picture?”

This question collects context before asking for an opinion. Product Talk’s guidance on story-based interviews recommends returning participants to a specific past event when they drift into “usually” or “in general”. Specific stories still depend on memory and self-report, but they give you details you can examine rather than a free-floating summary.

Do not force every participant into the same category label. Their description of the job may be the first sign that your positioning language does not match their frame.

2. Capture the unprompted association

“When you think of [product], what comes to mind first?”

Then use neutral probes:

• “What makes you say that?”
• “Can you give me a recent example?”
• “What else comes to mind?”

Record the participant’s words verbatim. Do not assume the first phrase is a category, the second an attribute, or the third a hidden truth. The sequence is not a validated hierarchy. Each answer is an association whose meaning you still need to understand.

Avoid praising a desirable answer or looking disappointed by an awkward one. The guide to avoiding leading questions shows how an interviewer’s wording and interpretation can supply the answer a participant then repeats.

3. Ask for the one-sentence handoff

“A colleague asks what [product] is. What do you say?”

Follow with:

“What would you compare it with so they understood?”

This is not a test with one correct answer. It gives you a candidate for the sentence that travels between colleagues. Ask who the imagined colleague is and what they already know; a technical buyer, finance approver, and daily user may need different explanations.

Compare the answer with your current hero copy after the interview, not during it. If the participant uses a category you dislike, do not correct them. The mismatch is the data.

4. Find the real alternatives

“Before choosing [product], what else did you consider?”

“What would you be using or doing if [product] did not exist?”

The second question catches alternatives that are not software: a spreadsheet, a shared inbox, a consultant, an internal process, or leaving the job undone. Ask for the sequence of the evaluation, who introduced each option, and what mattered at the decision point.

Do not turn this into a preference ranking detached from a real decision. If positioning is the main objective, combine this guide with a win/loss interview so you can reconstruct the choice rather than collect retrospective slogans.

5. Add a balanced word exercise

Only now should you introduce your vocabulary.

“Choose five words that best fit your experience of [product]. What made you choose each one?”

Here is a 24-word starting list:

Adapt the list to the decision, keep positive, neutral, and negative options, and randomise the order for each participant when practical. Ask what each selected word means to that person. A shared word can hide different interpretations.

This exercise adapts the Microsoft reaction-card method, which was designed to evaluate reactions to products and interfaces. NN/g recommends a shorter list of about 25 words or fewer for survey use, a mix that is not limited to flattering choices, and predefining any words that represent the intended direction. Using it here can structure a conversation; it does not turn an interview into a validated overall brand score.

If you are testing a new concept or page rather than an experienced product, use the separate concept-testing interview script and label the stimulus clearly.

6. Use personification as a hypothesis generator

“If [product] were a person at an industry event, what would they be like?”

Then ground the answer:

• “What in your experience gave you that impression?”
• “When did the product or company behave that way?”
• “Where does that description not fit?”

Personification may help a participant express an impression they find hard to name directly. It may also produce a playful answer with no decision value. Treat it as a projective prompt, not a personality test. Do not map one metaphor to a formal brand dimension or strategy without corroborating evidence. If the participant finds the exercise awkward, skip it.

7. Connect the attribute to its consequence

Choose a phrase the participant used, preferably one supported by a concrete example.

“You described it as [attribute]. What does that mean in practice?”

“What does that let you do?”

“Why is that important in that situation?”

This is a lightweight use of laddering. Reynolds and Gutman describe laddering as a way to explore links between product attributes, consequences, and personally relevant values. For example, “quick” might lead to “I finish the report before Monday’s meeting” and then “I can walk in prepared”.

Do not assume every phrase hides a profound personal value. Repeated “why?” questions can feel mechanical and encourage a tidy explanation. Use the guide to asking why without asking why, stop when the participant has answered the research question, and record incomplete chains as incomplete.

Close with: “What should I have asked but did not?” Then summarise the main points in the participant’s language and invite correction.

How to analyse brand perception interviews

Build an evidence table with one row per participant:

Keep unprompted language separate from words you supplied. Preserve verbatim excerpts and a route back to the transcript. Then compare the evidence with the promise and attributes you intended.

For a small qualitative study, write “5 of 8 participants in this customer group used survey language” rather than “63% of customers see us as a survey”. Even the fraction is a description of this sample, not a population estimate. NN/g’s discussion of numbers in qualitative studies explains why small samples and variable protocols create large measurement error.

Your summary can use four columns:

This row is illustrative, not Maren customer research. In a real report, include contradictory evidence and segment labels. The user research report template and guide to synthesising interviews into themes provide fuller structures.

Five mistakes that invalidate the result

1. Showing your positioning words too early. Ask for associations and comparisons before the word list.
2. Defending an inaccurate description. A misconception is evidence about interpretation, even when it is factually wrong.
3. Recruiting only enthusiastic customers. Include the participant types relevant to the decision, especially lost or former customers when studying comparison and rejection.
4. Mixing methods without naming the limit. An interview, a reaction-card exercise, a logo test, and an awareness survey answer different questions.
5. Treating a pattern as prevalence. Qualitative repetition can justify a hypothesis or design change; it does not estimate the market on its own.

Brand gaps are not automatically copy problems. NN/g’s work on brand experience treats visuals, tone, and behaviour as parts of the experience. If customers say “unreliable” because imports fail, a new tagline will not repair the evidence that created the association.

Run the template in Maren

Maren includes a Brand Perception interview style for spontaneous associations, competitive comparisons, laddering, and follow-up questions. You can use the template above as the project guide, invite participants with a link or email, and review full transcripts before using individual summaries or cross-interview synthesis.

Keep human judgement in the loop. Review the guide and participant-facing introduction before launch. Make sure the sponsor description is accurate, remove questions that do not support the decision, and do not claim independence, anonymity, or confidentiality unless each claim is true. After the interviews, check every synthesis theme against participant evidence and preserve counterexamples.

The result you want is not a prettier list of adjectives. It is a clearer account of when your product comes to mind, what people think it is, what they compare it with, and which experiences made them believe it.