Concept Testing in New Product Development: What Most Teams Get Wrong

Concept testing in new product development is the process of exposing a product idea to a representative sample of your target market before committing to full development, to measure whether the concept resonates, solves a real problem, and warrants investment. Done well, it reduces the risk of launching something nobody wants. Done poorly, it gives you false confidence and costs you more than skipping it altogether.

Most teams treat concept testing as a checkbox. They run a survey, collect some positive sentiment, and call it validated. What they miss is that concept testing is a decision-making tool, not a confidence-building exercise. The goal is to surface reasons not to proceed, not to gather ammunition for a launch you have already decided on.

Early in my career I was handed a whiteboard marker mid-brainstorm and told to run the session while the agency founder left for a client meeting. The brief was for Guinness. My first instinct was that I did not have enough information, that I needed more context before I could lead the room. I was wrong. The constraint forced me to work with what was in front of me, to ask better questions rather than wait for better answers. Concept testing has the same dynamic. The teams that get the most from it are the ones who start before they feel ready, not the ones who wait until the brief is perfect.

What Is Concept Testing Actually Testing?

This sounds obvious, but it is where most teams go wrong before they have even started. Concept testing is not testing whether people like your product. It is testing whether a defined group of people, with a specific problem, would change their behaviour because of it. Those are materially different questions.

A concept test should answer four things: Does the target audience understand what this product does? Do they believe it solves a problem they have? Do they believe it solves that problem better than what they currently use? And is the value proposition strong enough to prompt action? If your test does not structure around these four questions, you will get data that feels useful but cannot drive a decision.

I spent a period of my career overweighting lower-funnel signals. Conversion rates, cost per acquisition, return on ad spend. These numbers feel concrete. They are measurable, attributable, and they make a good slide. What I eventually understood is that much of what performance marketing gets credited for was already going to happen. You were capturing intent that existed, not creating it. Concept testing has the same trap. If you only test with people who are already aware of the problem category, you are measuring something narrower than you think. The more interesting question is whether the concept creates intent where none existed before, whether it can pull someone off the fence rather than just accelerate someone who was already moving.

This connects to a broader point about go-to-market thinking. If you are building a growth strategy, the articles in the Go-To-Market and Growth Strategy hub cover the full arc from early positioning decisions through to scaling, and concept testing sits squarely at the front of that arc.

How to Structure a Concept Test That Produces Useful Data

There is no single right methodology. The right approach depends on what stage of development you are at, how much time you have, and what decision you are actually trying to make. But there are structural principles that apply regardless of method.

Start with the problem, not the solution. Before you show anyone your concept, establish whether they experience the problem your product addresses. If they do not, their response to your concept is noise. This is the most common structural error in concept testing: showing the solution to people who do not have the problem, then treating their lukewarm response as market feedback.

Separate comprehension from appeal. Ask people to describe the concept back to you in their own words before you ask them whether they like it. If they cannot accurately describe what it does, any appeal rating is meaningless. You are measuring how well you wrote the stimulus, not whether the concept has merit.

Use purchase intent carefully. Stated intent consistently overstates actual behaviour. Someone saying they would “definitely buy” a product in a survey context is not the same as someone handing over a credit card. The gap varies by category and price point, but it is always there. Use intent scores as a relative measure across concepts, not as an absolute predictor of conversion. The qualitative reasons behind the intent scores are almost always more valuable than the scores themselves.

Test multiple concepts where possible. A single concept test tells you whether concept A passes a threshold. A comparative test tells you whether concept A is better or worse than concept B, which is the question that actually drives product and positioning decisions. If resource constraints force you to test one concept, at least test it against the current alternative your target audience uses.

Tools like Hotjar’s feedback and user research capabilities have made it significantly easier to run lightweight concept tests with real users at speed. The barrier to getting directional feedback early in the development process has dropped considerably. There is no excuse for launching blind.

Qualitative vs. Quantitative: Which One to Run First

The standard advice is to run qualitative first to generate hypotheses, then quantitative to validate them at scale. This is generally right, but the order matters less than the sequencing of decisions. If you have a clear hypothesis and a specific go or no-go decision to make, a quantitative test can come first. If you are still trying to understand the problem space, qualitative is non-negotiable.

Qualitative concept testing, whether through focus groups, depth interviews, or moderated usability sessions, gives you the language your target audience uses to describe their problem. This is underrated. The words people use to describe a problem are often the most effective words to use in positioning. I have seen this pattern repeatedly across categories: the most resonant product messaging comes directly from customer language, not from internal briefs or agency copywriting. The concept test is where you collect that language.

Quantitative testing gives you confidence in scale and allows you to segment. You can see whether the concept resonates more strongly with a particular demographic, firmographic, or behavioural segment. This is often where the most commercially useful insight sits. A concept that scores average overall might score exceptionally well with a specific segment, and that segment might be exactly the beachhead you need for launch.

If you are operating in B2B financial services, the stakes of getting this segmentation wrong are particularly high. The sales cycles are long, the relationships are complex, and a misaligned product concept can waste months of business development effort. The principles of B2B financial services marketing apply directly here: understanding which decision-makers hold the problem, not just which organisations sit in the target market, shapes how you design the concept test and who you recruit for it.

The Positioning Problem Inside the Concept Test

Here is something that rarely gets discussed in concept testing frameworks: you are always testing the concept and the positioning simultaneously, even when you think you are only testing the concept. The way you describe the product, the problem framing you use, the language in the stimulus material, all of this shapes the response. A strong concept described badly will fail a concept test. A weak concept described compellingly will pass one.

This is not an argument against concept testing. It is an argument for being rigorous about stimulus design. The stimulus should describe the concept in language that is representative of how you would actually communicate it at launch, not in sanitised research language. If your go-to-market plan involves a specific channel or format, the stimulus should reflect that context. A concept tested in a clinical survey environment will perform differently than the same concept encountered as a targeted ad or a sales conversation.

When I have been involved in reviewing marketing due diligence for acquisitions and investment decisions, one of the things I look for is whether the company has tested its positioning as rigorously as its product. Most have not. They have built a product, validated that the product works, and assumed the positioning will follow. It does not. Positioning is a separate decision that requires its own testing. The digital marketing due diligence process is a useful lens here: what evidence exists that the market wants this, and what evidence exists that this positioning can reach them?

Where Concept Testing Fits in the Broader Go-To-Market Process

Concept testing does not exist in isolation. It is one input into a go-to-market process that includes market sizing, channel strategy, pricing, and launch sequencing. The mistake is treating it as the only validation gate, or treating a passed concept test as a green light for everything downstream.

Think of it this way. A concept test tells you whether there is a market for the idea. It does not tell you whether you can reach that market efficiently, whether your pricing model is right, or whether your sales motion is aligned with how the buyer actually makes decisions. These are separate questions that require separate answers.

For B2B products in particular, the concept test result needs to be stress-tested against the commercial model. A concept that resonates strongly with end users but not with budget holders will stall in the sales process. If your go-to-market relies on demand generation through paid channels, understanding the cost of acquiring a qualified prospect matters as much as knowing the concept resonates. Models like pay per appointment lead generation make the unit economics explicit early, which is a useful forcing function for testing whether concept appeal translates into commercial viability.

The channel question is also more complex than it looks. A concept that tests well in a survey environment needs to be tested in the channels where it will actually appear. Endemic advertising, placing your message in contexts where your audience already engages with relevant content, is one approach that tends to produce more authentic response data than interruption-based methods. If your product is category-specific, endemic advertising can serve as a live concept test: real creative, real audience, real engagement signals, without the cost of a full launch.

For larger organisations running multiple product lines or business units, concept testing also needs to sit within a governance framework. Who owns the decision? Who has the authority to kill a concept based on test results? Without clarity here, concept testing becomes political rather than analytical. I have seen teams run perfectly good concept tests and then watch the results get overridden by a senior stakeholder who had already decided what they wanted to build. A corporate and business unit marketing framework for B2B tech companies is one way to create the structural conditions where testing results can actually influence decisions rather than decorate them.

Common Failure Modes and How to Avoid Them

Testing with the wrong audience. Recruiting is the most underinvested part of concept testing. A test with the wrong respondents produces confidently wrong data. Spend more time on the screener than you think you need to. If your product targets a specific professional role, industry, or behavioural profile, your sample needs to reflect that precisely. General consumer panels are rarely appropriate for B2B concepts.

Asking leading questions. Survey design shapes answers. Questions that describe the benefit before asking whether it is valuable will inflate scores. Questions that prime respondents with the problem before asking whether they experience it will inflate problem recognition. This is not always intentional, but it is common. Have someone outside the project review the stimulus and questionnaire before fieldwork begins.

Over-indexing on overall scores. Aggregate scores hide the most useful information. A concept that scores 6.5 out of 10 on average might score 8.5 among a specific segment and 4.0 among another. The overall score tells you nothing useful. The segmentation tells you everything about where to launch and who to target first.

Ignoring the competitive context. People do not evaluate products in a vacuum. They compare them to what they currently use. If your concept test does not establish the competitive baseline, you cannot interpret the results accurately. A concept that scores well in isolation might score poorly when respondents are reminded of the alternatives they already have.

The reasons go-to-market feels harder now than it did five years ago are partly structural: more noise, more channels, higher buyer sophistication. But a significant part of it is that teams are launching products without adequate front-end validation. Concept testing is not a guarantee of success, but it is a systematic way to reduce the cost of being wrong.

What Good Concept Test Output Looks Like

The deliverable from a concept test should be a decision memo, not a research report. A research report describes what happened. A decision memo says what it means and what you should do next. These are different documents and most concept test outputs are the former when they should be the latter.

A good decision memo from a concept test covers: which segment the concept resonates with most strongly and why, what objections or barriers to adoption surfaced, how the concept performs relative to the current alternative, what changes to the concept or positioning would improve performance, and a clear recommendation on whether to proceed, iterate, or stop.

The recommendation to stop is the hardest one to make and the most valuable one the process can produce. I have judged the Effie Awards, which recognise marketing effectiveness. The campaigns that win are almost always built on a clear-eyed understanding of what the audience actually responds to, not on what the brand team assumed they would respond to. That clarity comes from rigorous front-end work. Concept testing, done honestly, is part of that work.

Before committing to a launch plan, it is worth running a structured audit of your existing digital presence to ensure it can support the product you are about to bring to market. A checklist for analysing your company website for sales and marketing strategy is a useful starting point. A concept that tests well needs a website that can convert the interest it generates.

Concept testing is one of several disciplines that sit at the intersection of market intelligence and commercial decision-making. If you are building or refining your approach to growth strategy, the full Go-To-Market and Growth Strategy hub covers the frameworks and tools that connect concept validation to revenue outcomes.

The intelligent growth model frameworks that analysts like Forrester have developed over the years point consistently in the same direction: growth that compounds comes from understanding your market deeply before you commit resources to it, not from moving fast and hoping the market catches up. Concept testing is not slow. A well-designed lean test can be completed in two weeks. What is slow is recovering from a launch that the market did not want.

The BCG research on go-to-market strategy in B2B markets highlights how pricing and positioning decisions made early in the product development process are among the hardest to reverse once a product is in market. Concept testing is your best opportunity to stress-test those decisions before they become structural.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions