What is a Concept Testing? Definition, Examples & Best Practices
Concept testing helps teams understand whether a product, advert, name, message, packaging design or service idea is strong enough to develop further. A concept test survey usually presents a written description, image, mock-up, advert, prototype or package design, then asks respondents to rate it on measures such as appeal, believability, usefulness, differentiation, price fit and likelihood to buy.
In market research, concept testing sits between idea screening and product development. It is not the same as usability testing, which evaluates whether people can use something, or product testing, which evaluates the actual product experience. Concept testing examines the promise of the idea before full build, launch or media spend.
Common concept testing methods include monadic concept tests, sequential monadic concept tests, paired comparisons, MaxDiff concept testing, ad concept testing, name concept testing and packaging concept tests. The right method depends on whether the team wants absolute evaluation, direct comparison, preference ranking or diagnostic feedback.
Where it Comes From
Concept testing grew out of formal new product development practice. Urban and Hauser (1993) positioned concept testing as part of the wider process of identifying, designing, testing and launching new products. Green and Srinivasan (1978) helped establish conjoint analysis as a consumer research method for testing trade-offs between product attributes. Louviere, Flynn and Marley (2015) later formalised best-worst scaling, often called MaxDiff, as a way to measure preference strength when simple rating scales are not enough.
There is no single universal concept testing formula. A common score is purchase intent top-2-box: Concept purchase intent = respondents choosing “definitely would buy” or “probably would buy” divided by total respondents, multiplied by 100. For example, if 164 out of 400 target customers select the top two purchase intent options, purchase intent is 164 divided by 400, multiplied by 100, which equals 41%.
The case for early concept testing is practical. Harvard Business Review reported in 2011 that about 75% of consumer packaged goods and retail products fail to earn even $7.5 million in their first year. NIQ BASES stated in 2024 that its concept testing database covers 200,000+ initiatives globally and that its restage work has been directionally correct over 95% of the time across more than 5,000 global studies.
The weakness is that stated intent is not behaviour. Respondents may overstate interest because the concept is free to evaluate and carries no real purchase trade-off. Concept tests work best when paired with pricing, competitive context, behavioural data and later product validation.
Real-World Example
A UK direct-to-consumer skincare brand is preparing to launch a refillable moisturiser. The team tests three concepts with 900 women aged 25 to 44 who bought skincare online in the last six months. Each concept receives 300 monadic responses.
Concept A focuses on sustainability and scores 38% top-2-box purchase intent. Concept B focuses on sensitive skin and scores 51%. Concept C focuses on premium ingredients and scores 44%. Open-text feedback shows Concept B is seen as more personally relevant, but respondents say the refill mechanism is unclear. The team keeps the sensitive-skin positioning, rewrites the packaging copy, adds a visual explanation of the refill pod and removes a claim that respondents found vague. The decision was to refine one concept rather than average the three or run all of them into creative development.
Sources Cited
Green, P. E. and Srinivasan, V. (1978). “Conjoint Analysis in Consumer Research: Issues and Outlook.” Journal of Consumer Research.
Harvard Business Review (2011). “Why Most Product Launches Fail.” Harvard Business Review.
Louviere, J. J., Flynn, T. N. and Marley, A. A. J. (2015). Best-Worst Scaling: Theory, Methods and Applications. Cambridge University Press.
NIQ BASES (2024). “BASES Idea, Concept and Claims Testing.” NielsenIQ.
Urban, G. L. and Hauser, J. R. (1993). Design and Marketing of New Products. Prentice Hall.
Frequently Asked Questions
A concept test survey is a questionnaire that shows a target audience an idea before launch and asks them to evaluate it. The idea may be a product, advert, name, claim, packaging design or service proposition. The survey usually measures appeal, clarity, relevance, uniqueness and purchase intent.
The main concept testing methods are monadic testing, sequential monadic testing, paired comparison, ranking, MaxDiff and conjoint-style trade-off testing. Monadic testing gives each respondent one concept. Sequential monadic testing shows several concepts one at a time. MaxDiff is useful when researchers need to identify the strongest claims, features or messages.
A monadic concept test shows each respondent only one concept and asks them to evaluate it independently. This method is useful when researchers want a clean read on absolute appeal, purchase intent and clarity without forcing respondents to compare options. It usually requires a larger sample than sequential monadic testing.
A sequential monadic concept test shows each respondent several concepts, one after another, with the same questions repeated after each concept. It is more sample-efficient because each person evaluates multiple ideas. The trade-off is that results can be affected by order effects, respondent fatigue and comparison bias.
Product concept testing evaluates the idea before the product is fully built. It measures whether the proposition is appealing, clear and worth developing. Product testing evaluates the actual product experience, such as taste, usability, performance, quality or packaging interaction. Many teams use both before launch.
Concept test analysis usually compares top-2-box purchase intent, appeal, uniqueness, clarity and relevance across concepts. Researchers also examine subgroup differences, open-text comments, reasons for rejection and trade-offs. A strong analysis does not only identify the winning concept. It explains why it won and what needs improving.
MaxDiff concept testing asks respondents to choose the best and worst option from repeated sets of features, claims, benefits, names or messages. It produces clearer preference separation than ordinary ratings because respondents must make trade-offs. It is useful when many options look similarly attractive on standard scale questions.
Concept testing can be used for ad concepts, name options and packaging ideas as well as product concepts. For ads, it can test message appeal and brand fit. For names, it can test memorability and meaning. For packaging, it can test shelf appeal, clarity, perceived quality and purchase interest.