The part that makes a concept test trustworthy — randomized monadic cells, purchase intent measured the standard way, significance testing between concepts — is exactly the part most concept testing tools leave to you. Here it is the default, and a test costs $99 to $299 instead of a five-figure engagement.
Free to build · $99–$299 per test · Panel sample sourced by you · No subscription
Show someone three ideas in a row and their reaction to the third is contaminated by the first two. The fix has been standard for decades: monadic cells — each respondent sees one concept, assignment is randomized, and cells are compared statistically. It is also the part that used to require a survey programmer, which is why cheap concept testing software quietly skips it.
Here the agent wires it from the brief: display-only concept blocks, uniform cell assignment, per-cell quotas, and cell-vs-cell significance in the crosstabs. The wider landscape of study types is mapped in market research surveys, type by type.
How likely would you be to buy this product? (top-2-box share)
Each respondent saw one concept — randomized assignment, no order bias. The ▲ marks a difference that clears 95% significance on the effective bases.
Describe the concepts and the audience, and the agent programs the study: screener, concept exposure, comprehension check, appeal, uniqueness, purchase intent, and an open-ended reaction — with the concept shown as a clean display-only block before the questions.
Each respondent is assigned one concept cell with uniform randomization — true monadic design, the standard that keeps reactions uncontaminated. Crosstabs compare cells with significance testing, so the winning concept wins on evidence.
For sequential designs, concept order rotates per respondent so position effects cancel. The platform records which order each person saw — the audit trail serious testing requires.
Purchase intent on a labeled five-point scale reported as top-2-box, uniqueness and believability items, and the open-ended reaction coded into themes by AI — the standard battery, measured the standard way, comparable across every test you run.
Each cell's base size and margin of error are printed. A 6-point gap between concepts on n=200 cells is a finding; a 3-point gap is noise, and the report says so instead of crowning a false winner.
Weighted toplines per cell, cell-vs-cell crosstabs with significance letters, exports to SPSS, Excel, and Word. Field in the morning, decide in the afternoon.
Tests field through your lists, shareable links, or panel sample you source — vendor redirects built in. We do not sell sample. For concept tests, panel sample is the common choice, at a few dollars per complete from the vendor of your choosing.
Showing each person one concept needs n per cell — two concepts at n=200 means 400 completes. Sequential designs halve the sample but contaminate later reactions with earlier ones. We support both; we default to recommending monadic because clean beats cheap when a launch decision hangs on it.
Intent scales overstate real purchase behavior — everyone in research knows it, few say it. Concept tests reliably rank concepts and kill weak ones; they do not forecast year-one sales. Use them to choose what to launch, not to promise volumes.
Enterprise platforms compare your scores to category norms from their archives. We are young and do not have one, so your benchmark is your own concepts against each other and against your past tests. For a rank-and-kill decision, that is what matters anyway.
Showing a product, pack, ad, or feature idea to a sample of the target market before launch and measuring comprehension, appeal, uniqueness, and purchase intent — so the decision of what to build or launch rests on evidence rather than internal enthusiasm. The standard design is monadic: each respondent evaluates one concept, cells are compared statistically, and the strongest concept advances.
The working convention is 150 to 300 per concept cell. At n=200 per cell, differences of roughly 10 points in top-2-box intent clear significance; smaller gaps need bigger cells. Total sample is cells times n — two concepts at 200 each means 400 completes, which panel sample typically fills in a day or two.
Monadic shows each respondent exactly one concept — clean reactions, no comparison bias, but more total sample. Sequential (comparative) shows several concepts to each person in rotated order — half the sample, but each reaction after the first is influenced by what came before. Monadic is the standard for launch decisions; sequential suits early screening of many rough ideas.
Full-service concept tests from agencies commonly run $10,000 to $30,000. On Wavefield the study itself is $99 to $299 (the split-sample design needs no extra tier), plus panel sample at a few dollars per complete if you source respondents — so a two-cell test at n=200 per cell typically lands under $2,000 all-in, most of it the panel's fee, not ours.
Related: survey analysis software · brand tracking software · sample size calculator
Brief the agent, review the cells, field to your panel. Significance-tested answers by the afternoon — free to build, from $99 to field.
Start a test