Concept Testing Methods: Designs and Examples
Compare four concept testing methods: monadic, sequential monadic, direct choice and qualitative research. Choose the design that answers your decision.
Concept testing methods answer different questions: how an offer works on its own, which ideas to screen, which option wins a choice and why an idea is misunderstood. Choose the method for that decision. Our concept testing guide covers the full study, from choosing an audience to deciding what to do.
Four designs for four decisions
Decision | Design to consider / Main tradeoff |
|---|---|
Evaluate an offer independently | Design: Monadic: one concept per person Tradeoff: Needs separate cells for each concept |
Screen several early directions | Design: Sequential monadic: each person evaluates several, one at a time Tradeoff: Exposure and order can affect later reactions |
Choose between concrete alternatives | Design: Direct comparison, with independent measures first if needed Tradeoff: Relative winner may still be weak in absolute terms |
Understand why an idea fails | Design: Qualitative exploration with a consistent stimulus Tradeoff: Depth does not establish population prevalence |
Independent ratings, early screening, direct choice and detailed diagnosis are all useful for different jobs. You do not need to buy all four.
Example: two $25 bottle offers
In this hypothetical example, target bottle buyers compare two $25 offers: Desk Brew and Travel Brew. The team must decide whether to judge each alone, screen both or choose between them. Each sequence below shows what a buyer sees before answering.
What each buyer sees before answering
In a monadic study, different buyers see only A or only B. In a sequential study, they see A then B or B then A. A later direct choice is a different question from those first reactions.
Person or question block | Exposure / Readout |
|---|---|
Buyer 1 | Exposure: A only Readout: Independent A reaction |
Buyer 2 | Exposure: B only Readout: Independent B reaction |
Buyer 3 | Exposure: A → B Readout: Sequential evaluations |
Buyer 4 | Exposure: B → A Readout: Same blocks, reversed order |
Later direct-choice block | Exposure: Defined pair → direct choice Readout: Preference and reasons |
Each Buyer 1–4 row is a different person. Arrows show what they see, not a change in preference. The last row is a later direct-choice question.
Qualtrics’ concept program describes single and sequential exposure. Choose the setup for the question you need answered, not just because the software offers it.
Choose what buyers see before each answer
A one-card study shows each buyer one matched bottle offer and asks the same questions. A sequential study shows both, one after another in a planned order. A direct choice shows the pair when asking for a preference.
The first card may teach someone about the category or benefit, making the second feel repetitive. Balancing order helps you check that effect. It cannot turn later answers into independent first impressions. A side-by-side choice also encourages comparison.
Match the amount of detail and visual finish. Show the price and sales channel for each offer. To test a presentation change alone, keep the other facts fixed. Otherwise, describe the complete offers being compared.
Monadic: one concept per person
For first reactions to the finished $25 bottle offers, randomly assign target buyers one card each. Ask what they think it is, when it helps, what worries them and whether they would consider buying before showing alternatives.
Show every group’s answer distribution and number of valid responses, with understanding beside appeal. For the same number of ratings per concept, one-card groups need more people than asking everyone to rate both. Both groups can come from one target recruitment pool. More ratings from one person are not more first impressions.
You can add a later preference question, but label it clearly. It shows the choice after comparison; it does not change what the earlier first response meant.
Sequential monadic: a planned series of concepts
Use a sequence to narrow early ideas before developing them further. Keep it short enough for people to pay attention, ask the same questions after each card and save the respondent, order, position and version.
For two concepts, balance AB and BA. For larger sets, plan which offers each person sees and how much work that requires. Ask for results by concept and position so a combined winner does not hide different later reactions.
Keep answers from the same person linked, including any missing blocks. Many ratings are not the same as many independent people. If order changes which idea moves forward, retest the remaining candidates with independent first exposure.
Direct choice: preference within the set shown
A direct choice with reasons can help select one of two workable pack directions to improve. Include “none” or another rejection option when buyers can reject both. Winning the pair does not guarantee that either meets the development goal.
Say what alternatives are in the comparison. Travel Brew beating Desk Brew cannot show that it beats the bottle someone already uses. An unrealistic alternative can make a preference look stronger than it would be in the market.
If appeal on its own matters, ask about it before showing alternatives and report it apart from later preference. Do not combine the two into an unnamed score. Say which action each result supports.
For courses, our higher-education research guide compares a standalone workplace-skills course with an award-pathway module. Curriculum, assessment and progression terms affect which complete offer to develop, beyond its relative preference score.
Qualitative research: understand the problem
Example: a repair-collection fee
In this hypothetical repair-collection example, people like a card but think the collection fee includes the repair. Show the same material and ask what they receive, what the fee covers and what is missing. Use the explanations to propose a specific edit.
This gives depth, not the share of a population holding that view. After fixing the card, use a broader study with the target audience to test appeal. A large survey of an unclear card would count reactions to several imagined services.
Judge speed by what the study lets you decide. A sequence may reduce recruitment but changes the context of later answers. A focused diagnosis may save a larger study by fixing confusion first.
What should the results help you decide?
Agree the outcome, exact card, when alternatives appear, which answers count and what lets an idea move forward. Ask for the report that fits: one-card groups, results by position, direct choice or a misunderstanding linked to an edit.
Concept responses help development and communication. Physical performance, service delivery and actual repeat buying need evidence suited to those questions. For simulated responses, agree the data behind the audience and ask how the method has been checked for the task.
Plan assignment, checks for order effects and which answers stay linked. The design is ready when you can name the open question and what you will do with the answer.
Across markets, our international concept-testing guide covers shared product facts, local prices, alternatives and question meaning. For sales before production, our preorder research guide distinguishes waitlists, refundable reservations and paid orders.
Our concept testing guide follows a simulated protein-drink result into a proposed paid trial.
Explore Sapien's product-development research for comparing concepts, features, packaging and claims.
Choosing between independent reactions and a faster screen? discuss that tradeoff with Sapien before planning the study.
