Pick the right design and the right sample size, then get a defensible read on your concept or ad before you spend to bring it to market.
What Concept And Ad Testing Measure
Concept testing evaluates a product or ad campaign and how consumers will receive it before it goes to market, per SurveyMonkey's concept testing guide. Ad testing scores a specific creative on appeal, believability, purchase intent, relevance, and uniqueness, per SurveyMonkey's ad testing guide.
SegmentOS ships a Concept Testing template with the questions, screeners, attention checks, and quality controls already built in. Free to start. Paid plans from $29 a month.
Trusted Scale And Transparent Pricing
A verified panel of 30M+ respondents across 127 countries, with response costs shown before you launch.
What You Learn From A Concept Test
A concept test scores five metrics on a 5-point Likert scale, then reports each as top-2-box, per SurveyMonkey's ad testing guide.
Appeal shows how enticing your imagery and message are.
Believability shows whether the message reads as credible.
Purchase intent shows whether the creative moves people to buy.
Relevance shows whether it fits what your audience wants.
Uniqueness shows how much it stands apart from other ads.
Top-2-box combines the two most positive answers into one percentage. Example: 30% "extremely believable" plus 20% "very believable" is a 50% top-2-box score.
One caveat we hold to. In our experience running these studies, stated purchase intent reads higher than real buying. We treat it as a directional signal and plan for actual behavior to land lower.
Which Concept And Ad Testing Survey Template Fits
The design decides your sample cost and how clean your read is. Three designs cover almost every concept or ad test.
Monadic: One Concept Per Respondent
Each respondent sees one concept only, per SurveyMonkey. This gives the cleanest read on absolute performance and removes order bias. It costs the most, because each respondent fills just one cell. Use it for final validation.
Sequential-Monadic: All Concepts, One At A Time
Each respondent sees every concept in rotation, one at a time, per SurveyMonkey. It costs less, since each respondent covers several cells. Randomization limits order effects. Use it to screen several concepts on a budget.
Comparative: All Concepts Side By Side
Respondents see all concepts together and rank them, per Conjointly's monadic testing guide. Results are easy to read but interdependent, so each score is relative to the others rather than a standalone measure. Use it for quick rank ordering.
Sizing The Study
Plan about 200 responses per concept, per Conjointly, with a working range of 100–250 per cell. To size the study, use this formula:
Total N = (N per concept × total concepts) / concepts seen per respondent.
A monadic test of 3 concepts at 200 each needs 600 respondents. A sequential-monadic test where each respondent sees all 3 needs 200.
Who Concept Testing Is For
You are a marketer testing ad creative before a media spend.
You are a founder validating a product concept before you build.
You are a product manager choosing between feature directions.
You are a strategist ranking messaging routes before a launch.
How To Launch Your Study
Open the Concept Testing template. Questions, screeners, attention checks, and disqualification logic are already set up.
Add your concepts or ads, then pick a design: monadic, sequential-monadic, or comparative.
Choose your audience. Use your own list or the verified panel, and see feasibility and cost before you launch.
Launch, then read top-2-box results and segment breakdowns in real time.
What's Included In The Concept Testing Template
The template ships ready to field:
Industry-standard questions for all five metrics, on 5-point Likert scales.
Screeners and disqualification logic to reach the right respondents.
Randomization for sequential-monadic and comparative designs.
Real-time charts, segmentation, and CSV, XLSX, PDF, and PNG exports.
Every study runs through six quality layers: attention questions, screener disqualification, speeding detection, device fingerprinting, logic-trap questions, and quota management. Responses that fail never reach your dataset.
Panel access is included on the $79 Pro plan.
Pricing
Free to start. No card required.
Paid plans from $29 a month for the full research-grade builder and templates.
Pro at $79 a month adds verified panel access. Panel responses start at $0.73 per B2C response, and the cost is shown before you launch. No surprise invoices.
Build a study, reach verified respondents in your market, and get answers back in days.
Everything a research team does. Without the research team.
With SegmentOS you can build the study, reach a verified audience, and get data you can actually trust. End to end, no research background required.
Data quality
Every study runs through device fingerprinting, speeding detection, attention checks, and screener disqualification.
Related Templates And Tools
Compare designs and pricing before you commit:
Frequently Asked Questions (FAQ)
How many respondents do I need per concept?
Plan about 200 per concept for Likert-scale tests, with a working range of 100–250 per cell, per Conjointly.
How do I calculate total sample size?
Total N = (N per concept × total concepts) / concepts seen per respondent. Three concepts at 200 each: 600 respondents monadic, 200 sequential-monadic.
What is top-2-box?
It combines the two most positive answer choices for a metric into a single percentage, per SurveyMonkey. It gives you one clear number per metric.
Does purchase intent predict sales?
Not on its own. In our experience, stated purchase intent reads higher than real buying. We treat it as a directional signal and plan for actual behavior to land lower.
Can I use my own audience instead of the panel?
Yes. Collect from your own list, or target the verified panel of 30M+ respondents across 127 countries.
