The template galleries hand you questions without the method, and the method guides never show a question. Here are the six survey types that do the real work of market research — each with sample questions, what it is actually for, and the sampling and weighting rigor that makes any of them count.
By the Wavefield Research team · Published Aug 19, 2026 · Updated Aug 22, 2026
The short answer
A market research survey is a structured questionnaire fielded to a defined sample of a market to measure awareness, usage, attitudes, or reactions. The six working types are brand tracking, concept testing, usage and attitude, pricing, segmentation, and ad testing. What separates a survey from a form is rigor: sampling, quotas, weighting, and honest margins of error.
Every commercial study is one of these or a combination. The sample questions show the shape each type is built from; adapt the wording, keep the structure.
Measures awareness, consideration, and image over repeated waves
Which of these brands have you heard of? (Select all that apply)
Which brand comes to mind first for [category]?
The tracker is the workhorse of brand research: the identical questionnaire fielded quarterly, so movement between waves is real change rather than wording noise. The funnel logic — awareness, familiarity, consideration, usage, preference — turns into trend lines with significance markers. The discipline that matters most: never change question wording between waves.
Puts an idea, pack, or message in front of the market before you commit
In your own words, what is this product for?
How likely would you be to buy this at $29? (Definitely would not – Definitely would)
Show the stimulus, measure comprehension, appeal, uniqueness, and purchase intent — and rotate concepts so position does not bias the scores. Split-sample designs do the honest version of message testing: half see version A, half see B, and the crosstab says which one wins with significance attached.
Maps how a category actually gets used and what people believe about it
How often do you use [category product]?
Which of these statements best describes how you feel about [category]?
The foundational study for entering or repositioning in a category: who uses what, how often, in which occasions, and with which frustrations. U&A studies run long, so quotas and question routing matter more here than anywhere — screeners send non-users past usage batteries instead of through them.
Finds what the market will actually pay
At what price would this product be so expensive you would not consider it?
At what price would it be so cheap you would question its quality?
Those two questions are the start of a Van Westendorp price sensitivity meter — four willingness thresholds that map an acceptable price range. Simpler designs test one price per split-sample arm and compare purchase intent across arms. Either beats asking people directly what something should cost, which invites bargaining rather than truth.
Divides a market into groups worth treating differently
Which of these best describes your approach to [category]?
How much do you agree: I am usually the first among my friends to try new products.
Attitude and behavior batteries feed a clustering analysis that produces the segments; crosstabs then profile each one by demographics, usage, and media habits. The honest note: segmentation is only as good as the follow-through — segments nobody activates against are a very expensive slide deck.
Measures whether communication moves anyone before the budget ships
What was the main idea this ad was trying to get across?
After seeing this, how has your interest in [brand] changed?
Recall, comprehension, appeal, and persuasion — measured against a control cell that saw nothing, or a competing execution. Rotation and split samples are the mechanics; the metric that matters is the lift between cells, not the absolute score of any one execution.
The wording rules behind every sample above — balanced scales, one idea per question, randomized options — are covered in survey question types, and a free bank of fifty open-ended questions lives in 50 open-ended survey questions.
Every type above produces numbers, but who answered decides whether the numbers describe the market or just the sample. Define the population first, screen for it, quota the mix during fielding, and weight the result to known targets. A concept test on 200 self-selected volunteers is a focus group wearing a percentage sign.
Template galleries hand you questions and stop there. The decisions live in the numbers underneath: how many completes each subgroup needs before its result means anything, what the margin of error is on the effective base after weighting, and whether the difference between two waves clears the noise floor. Reports should print those numbers next to every chart.
A full-service agency market research survey commonly starts around $15,000 to $30,000. The same instrument self-served on a modern platform runs $99 to $299 with programming, fielding infrastructure, weighting, and significance testing included — plus panel sample at a few dollars per complete if you do not have a customer list to field to. The analysis rigor is the same math either way; what changed is who has to do it by hand.
The rigor layer has its own guides: how respondents get selected in quota sampling vs. random sampling, how skews get corrected in survey weighting, how many completes you need in the sample size calculator, and the full workflow in how to analyze survey results. For where surveys sit among the other research methods entirely, see market research methods.
The sequence is the same for every type. Define the decision the study serves and the population that can answer it. Draft the instrument — screener first, then the measurement battery for your study type, demographics last, one or two open-ends placed late. Set quotas on the sample mix and weighting targets for the analysis. Field through your list, a shareable link, or panel sample you source, in batches so you can watch incidence. Then read the results in order: quality flags, weights, topline, crosstabs, significance.
On Wavefield that sequence is compressed: the agent programs the instrument from a plain-language brief, quotas and quality flags run during fielding, and the weighted topline, significance-tested crosstabs, and SPSS exports are ready when fielding closes. Trackers re-field the identical instrument as $99 waves with trend reports built in — the full tracker story is on brand tracking software. Building is free; a study is $99 to $299.
A structured questionnaire fielded to a defined sample of a market — sometimes called a customer research survey or consumer survey when the sample is your own customers or the buying public — to measure awareness, usage, attitudes, or reactions to concepts and prices. It differs from a feedback form in its rigor: a defined population, screening and quotas to control who answers, weighting to correct sample skews, and margins of error reported honestly.
Six cover most commercial work: brand tracking (awareness and image over waves), concept testing (reactions to ideas before launch), usage and attitude studies (how a category actually behaves), pricing research (willingness to pay), segmentation (dividing the market into actionable groups), and ad testing (whether communication persuades). Most real programs combine them — a U&A to map the ground, concept tests to develop, a tracker to monitor.
The method is legitimate and underpins most business and political decision-making — but the term is also borrowed by two other things: paid-survey sites that pay pennies to click through questionnaires, and sales calls disguised as research (the industry calls it sugging). A legitimate research survey never asks you to pay, never sells to you mid-survey, and states who is collecting the data and why. From the researcher's side, legitimacy means real sampling, honest weighting, and stated margins of error.
Good questions are specific, single-purpose, and matched to the study type: awareness lists for trackers, intent scales for concept tests, threshold questions for pricing. The wording rules matter more than the templates — balanced scales, no leading language, one idea per question, randomized option order. For the open-ended side, the highest-value question in most studies is the follow-up probe: asking why right after a rating you already captured.
Measuring your own customers specifically? That is its own discipline: customer feedback surveys covers CSAT, NPS, and CES programs.
Trackers, concept tests, U&A, pricing — the agent programs the instrument and the analysis comes back done. Free to build, from $99 to field.
Start a project