You have a slate of product ideas and the budget to launch one. A concept testing survey scores each idea with your target audience, so you launch the one most likely to sell. These eight templates cover the early calls, from new products and packaging to claims and competitor positioning.
The Concept Testing Module is a guided research workflow. It runs a monadic design, showing each respondent one idea so scores reflect the concept, not the lineup. Respondents rate appeal, uniqueness, and purchase intent across up to 10 concepts at 200 respondents each. Key Driver Analysis maps every attribute against purchase intent to show which one drives the decision.
The built-in Audience panel reaches more than 1 million US respondents, with 19 million more worldwide, and returns results in about an hour. Concept tests average a 95.1% completion rate at about 12 questions. Short tests of 1-10 questions finish at 98.8%; those past 21 fall to 81.5%.
Once results are in, T2B (Top-2-Box) scoring ranks the concepts and industry benchmarks show whether a winner beats category norms. AI Auto-insights names the leading idea and any subgroup that sees it differently, given at least 50 responses and 30 per subgroup. For the full methodology, see our concept testing guide.
How to create a concept testing survey
Open the Concept Testing Module or pick one of the eight templates on this page, starting with the market research product survey.
Add your concepts as text, images (PNG, JPG, or GIF), or linked video, up to 10 per project.
Set the sample. The default is 200 respondents per concept, enough for a clean top-line score.
Choose your attributes. Appeal, uniqueness, and purchase intent are standard, and Key Driver Analysis maps each against the buy.
Field to the Audience panel by job, region, or category usage, or send to your own list by link or email.
Review results: T2B scoring ranks the concepts, AI Auto-insights names the winner, and benchmarks frame the scores.
When to use concept testing surveys
Before a product launch: Test three to five ideas with your market and let purchase intent pick the strongest start.
In packaging and design: Show visual concepts to see which earns the highest appeal before you print.
For claims and messaging: Validate which benefit statement lands hardest, tied back to purchase intent.
Early in development: Screen a slate of rough ideas to kill the weak ones before they cost budget.
For competitive positioning: Show your concept next to a reference product to read relative appeal.
Who runs concept testing surveys
Consumer insights managers: They validate concepts across the pipeline and rely on monadic scores for clean, unbiased reads, on project budgets around $15,000 to $40,000.
Product managers: They choose between feature or product directions early, before engineering spends on the wrong one.
Brand managers: They test names, packaging, and creative directions to rally the team around the idea most likely to sell.
Market research agencies: They run concept tests for clients with the built-in panel and benchmarks, often on $10,000 to $20,000 yearly budgets.
Tips for creating a concept testing survey
Test three to five, not ten: The module allows 10 concepts, but a smaller slate keeps Key Driver Analysis readable and the winner clear.
Keep monadic unless you need head-to-head: One concept per respondent prevents the order bias that skews scores; switch to sequential monadic only for direct comparison.
Size for the subgroups you will report: AI Auto-insights needs at least 50 responses and 30 per subgroup, so grow the sample beyond 200 if you plan to cut the data.
Read Key Driver Analysis, not just the average: A concept with moderate appeal that correlates strongly with purchase intent often beats a flashier one that does not.
Frequently asked questions about concept testing surveys
What is monadic concept testing?
Monadic means each respondent sees one concept and nothing else, so their scores reflect that idea on its own. With no lineup to compare against, you avoid the order and context bias that creep in when people rate several concepts in a row. The Concept Testing Module uses monadic by default. Sequential monadic, where each person sees a small defined set, is available when head-to-head comparison is the actual goal.
How many concepts should I test at once?
The module allows up to 10, but three to five usually gives the cleanest decision. A smaller slate keeps the Key Driver Analysis scatterplot readable and lets the winning concept stand out. Testing too many at once dilutes focus and rarely improves the call. If you have a long list, run an idea screen first to cut it down, then concept-test the finalists.
How many respondents does a concept test need?
The default is 200 respondents per concept, enough for a reliable top-line score and benchmark comparison. If you want AI Auto-insights to break results out by demographics or usage, plan at least 50 total responses and 30 per subgroup. Grow the sample to match the cuts you intend to present. Screening questions qualify the right respondents before they enter.
How do I know which concept wins?
T2B scoring rolls the top two positive ratings on each attribute into one percentage, so concepts rank at a glance. Key Driver Analysis then shows which attribute predicts purchase intent, separating a flashy score from a meaningful one. AI Auto-insights names the leading concept and flags any subgroup that disagrees with the overall sample, so you can position without running a second study.
What stimuli can I use in a concept test?
Text descriptions, images in PNG, JPG, or GIF, and video through a YouTube or Vimeo link. There is no native video hosting, so a pre-launch clip needs a shareable link first. Image files go up to 10 MB, and each project supports up to 10 stimuli. For broader setup and question design, see our market research solutions.
