What questions go in a brand tracking survey?
TL;DR
- Ask unaided awareness first, before showing any brand list. Once people see names, they can't unsee them.
- Core funnel: unaided awareness, aided awareness, consideration, preference or last purchase, image, advocacy.
- The recommend question with a 0 to 10 scale comes from Frederick Reichheld's 2003 Harvard Business Review article; keep it if you already track it.
- Neon Blue can field the same question set as a simulated wave between live waves, kept as a separate series. Run your first study free
Neon Blue builds simulations of human behavior. A brand tracking survey asks the same core questions in the same order every wave: unaided awareness, aided awareness, familiarity, consideration, preference or last purchase, brand image, and advocacy. Order matters because each question can prime the next, so unaided questions always come before any brand names are shown. The set should be short enough to field every wave without fatigue, with room for a few rotating questions tied to a live campaign. Change the core wording and you break the trend line.
What is the core brand tracking question set?
Ask in this order. Replace [category] with how buyers describe it, not the industry term.
- When you think of [category], which brands come to mind? [Open, record first mention separately]
- Which of these brands have you heard of? [Aided list, randomized]
- How familiar are you with each brand you've heard of? [Only the name / A little / Quite well / Very well]
- Which of these brands would you consider the next time you buy [category]? [Multi-select]
- Which one brand would you most likely choose? [Single]
- Which brand did you buy the last time you bought [category]? [Single]
- Which of these brands, if any, do you associate with each statement? [Brand-by-attribute grid]
- How likely are you to recommend [brand] to a friend or family member? [0 to 10, asked of users]
Questions 1 and 2 track awareness, 3 to 6 the funnel, 7 brand image, and 8 advocacy. The brand tracking study guide covers wave timing, sample sizes and how to read small moves.
Where does the recommend question come from?
In December 2003, Frederick Reichheld published "The One Number You Need to Grow" in Harvard Business Review. He proposed asking customers how likely they were to recommend a company on a 0 to 10 scale, classing 9 and 10 as promoters, 7 and 8 as passives, and 0 to 6 as detractors, and subtracting the share of detractors from the share of promoters. The question spread quickly. For a brand tracker it has two limits: it only means much among people who have used the brand, and it says nothing about the much larger group who never considered you. Ask it, but don't let it replace the funnel. Consideration and choice say more about growth, and the purchase intent guide covers how to read stated choice.
How do you write the brand image questions?
| Approach | How it works | Use when |
|---|---|---|
| Pick-any grid | Show statements; people tick every brand that fits | You track many brands and want association counts |
| Rating per brand | Rate each known brand on each statement | You track few brands and want depth |
| Category entry points | Statements describe buying situations ("a quick lunch at work") | You want to know when buyers think of you |
Pick-any grids are the easiest to keep stable across waves. Situation-based statements, an idea associated with Byron Sharp's How Brands Grow (2010) and Jenni Romaniuk's later work, tell you in which buying moments the brand comes to mind, which is often more useful to a media planner than "is innovative". The channel planning guide shows how those moments feed media choices.
What does a tracking wave look like on a simulated audience?
Hypothetical: a national coffee chain tracks awareness and consideration quarterly but launches a breakfast campaign between waves and wants a read in the gap.
- Audience: US adults who bought coffee away from home in the past month, split by daily and occasional buyers, by morning versus afternoon visits, and by age band; matched to the live tracker's sample definition.
- Design: the eight core questions plus two campaign questions ("Have you seen or heard anything from [brand] about breakfast in the past four weeks?" and a recall prompt), fielded on a simulated audience.
- Readout: the funnel by segment, image associations for the breakfast statements, and campaign recall, reported as a simulated series beside the live waves.
Example output layout (illustrative, not a real result)
| Measure | Last live wave | Simulated mid-wave read | Next live wave |
|---|---|---|---|
| Unaided awareness (first mention) | __% | __% | __% |
| Consideration | __% | __% | __% |
| Most likely to choose | __% | __% | __% |
| "Good for breakfast" association | __% | __% | __% |
| Recommend score (users) | __ | __ | __ |
Keep simulated and live in separate columns. The live waves stay the record.
What mistakes break a brand tracker?
- Changing the wording. Even small edits shift answers; the forbid-or-allow example in the message testing questionnaire shows how much.
- Changing the brand list. Adding a brand to the aided list changes the context for every other brand. Add new brands at the end and flag the wave.
- Changing the sample. If the audience definition drifts, the trend reflects the sample, not the brand. Write the definition down once, as in building a synthetic audience.
- Asking too much. A long tracker tires people and late answers get thin. Keep the core under ten minutes and rotate extras.
Brand managers using the tracker to judge a launch can pair it with the pre-launch research calendar.
Questions about brand tracking surveys
What questions should be in a brand awareness survey?
At minimum, unaided awareness ("which brands come to mind?") and aided awareness ("which of these have you heard of?"), in that order, with first mention recorded separately.
How long should a brand tracking survey be?
Short enough to answer in under ten minutes, which usually means the core funnel, one image grid and a few rotating questions.
How often should you run a brand tracker?
Most teams run quarterly or monthly waves, more often around big campaigns. A simulated read can fill gaps between live waves.
Can Neon Blue run a brand tracking survey?
Yes. Neon Blue fields your exact tracker questions on a simulated audience matched to your live sample and reports a separate simulated series by segment. Start a free first study.