You have been asked to set up a brand tracker, and the first real task is the questionnaire. The questions you choose, and the order you ask them in, decide whether the numbers in wave six are comparable with wave one. This guide gives you a working flow for a brand tracking survey, with example wording for each block. For the wider method, start with what brand tracking is and how it works.
Key takeaways
- A brand tracking survey follows a funnel: awareness, familiarity, consideration, preference, usage, then image and advocacy.
- Ask unaided awareness before aided awareness, and show the brand list only after open recall is recorded.
- Keep wording, answer scales and order identical in every wave and market, or you cannot trust the trend.
- Protect length: long image grids and late-survey fatigue are the main causes of dropout and weak data.
- Check any wave-to-wave change against the margin of error before you report it as movement.
What questions belong in a brand tracking survey?
A good questionnaire mirrors the path a buyer takes: do they know you, do they know what you stand for, would they consider you, and have they bought. Each block below has a job, and each feeds a metric you will report. The brand health tracking metrics article explains how to read them once you have the data.
| Order | Block | Example question wording | Answer format |
|---|---|---|---|
| 1 | Screeners | ”Which of the following have you bought or used in the past 3 months?” | Multi-select, with an exclusion rule |
| 2 | Unaided awareness | ”When you think of [category], which brands come to mind?” | Open text, up to 5 boxes |
| 3 | Aided awareness | ”Which of these brands have you heard of?” | Multi-select list, randomized |
| 4 | Familiarity | ”How familiar are you with each brand?“ | 5-point scale, from never heard to know it very well |
| 5 | Consideration | ”Which of these would you consider the next time you buy [category]?” | Multi-select |
| 6 | Preference | ”Which one would be your first choice?” | Single choice |
| 7 | Usage | ”Which brands have you bought in the past 3 months, and which do you use most often?” | Multi-select, then single choice |
| 8 | Image grid | ”Which brands do you associate with each statement?” | Grid, brands by attributes |
| 9 | Net promoter score | ”How likely are you to recommend [brand] to a friend or colleague?“ | 0 to 10 scale |
| 10 | Ad recall | ”Have you seen advertising for any of these brands recently?” | Multi-select, then where seen |
| 11 | Demographics | Age, gender, region, income or firmographics | Single choice |
Put the screeners first so ineligible people leave early. Put demographics last, or draw them from the panel profile, because they are quick and respondents will answer them even when tired.
Why does question order matter?
Earlier questions shape later answers. Show a list of brands and every respondent now knows the competitive set, which lifts awareness, consideration and image scores for anyone on the list.
That is why unaided awareness comes before aided awareness, and why both come before any question that names brands. Randomize the order of the brand list so no name always sits first. If your tracker covers several markets, the same order must hold in all of them, a point we cover in global brand tracking.
How do you keep wording identical across waves and markets?
Treat the questionnaire as a locked instrument. Change a word in the awareness question and a rise in awareness may reflect the new wording, not the market.
- Freeze the master questionnaire and keep a change log with the date and reason for each edit.
- Keep the same scale, the same anchors and the same direction of scale every wave.
- Rotate brand lists and attribute rows, but never add or drop a brand silently. Record when a new competitor enters the list.
- Translate with a second person checking back-translation, and review any culture-specific attribute wording before launch.
- If you must change a question, run old and new versions in parallel for one wave to measure the break.
How do you design the image grid?
The image grid asks which brands fit which attributes, for example “good value,” “innovative” or “trustworthy.” It is the longest block and the first place respondents start to click through without reading.
Choose 8 to 12 attributes tied to your brand strategy and your competitors’ claims, and write each as a short, single-idea statement. “Good quality and affordable” asks two things at once, so split it. Rotate rows, and consider asking each respondent only about the brands they know, which shortens the grid and removes guesses from people who have never heard of a brand.
How long should a brand tracking survey be?
Length trades against completion. The longer the survey, the more people quit and the more the remaining respondents rush, and the people who finish may differ from those who leave. Watch the dropout point in each wave: if it clusters at the image grid, shorten the grid before touching the funnel.
Cut in this order: optional ad-recall detail, extra attribute rows, secondary usage questions. Keep the funnel questions intact, because they carry the trend.
What does a worked example look like?
The numbers below are illustrative, not from a real study. Suppose a tracker interviews 1,000 category buyers per wave, and wave two returns this funnel for your brand.
| Funnel stage | Share of respondents | Share of the stage above |
|---|---|---|
| Unaided awareness | 18% | not applicable |
| Aided awareness | 62% | not applicable |
| Consideration | 27% | 27 / 62 = 44% of those aware |
| First choice | 16% | 16 / 27 = 59% of those considering |
Consideration was 27% in wave one and is 30% in wave two. Is that growth? At n = 1,000 per wave, the standard error of the difference is the square root of (0.27 × 0.73 / 1,000 + 0.30 × 0.70 / 1,000), which is about 2.0 points. Multiply by 1.96 and the 95% margin on the difference is about 4.0 points. A 3-point change sits inside it, so the honest report is “no reliable change yet.” Read what statistical significance means before you flag movement in a readout.
What are the common mistakes?
- Asking aided awareness first. It contaminates unaided recall. Open recall always goes before any brand list.
- Editing questions between waves. Small wording tweaks create trend breaks that look like market movement.
- Writing a long, unrotated grid. Position effects and fatigue distort the attribute scores, and dropout rises late in the survey.
- Using a different scale for each question. Mixed 5-, 7- and 10-point scales make the funnel hard to compare and hard to explain.
- Mixing consideration and preference. Consideration lets people pick several brands. Preference asks for one. Merging them blurs the funnel.
- Reporting small changes as real. Always check against the margin of error and the sample size behind each cell.
How do you put the questionnaire to work?
Once the questions are locked, the practical work is turning each wave into a readout. Tools that handle market and brand intelligence let you follow brand health, consideration and competitors across waves in one place. For wide trackers with many markets, crosstabs with weighting and significance testing keep the wave comparisons honest, and marketing teams can see brand health alongside campaign and competitor signals.
For your next step, write the questionnaire out in the order above, cut it to the length your sample will tolerate, and field a small soft launch before wave one. If you are also choosing a platform to run the tracker, how to choose brand tracking software lists what to check.