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

Market testing in new product development is the process of exposing a product, concept, or proposition to real customers before full-scale launch, with the explicit goal of reducing commercial risk. Done well, it tells you whether demand actually exists, not just whether people say they like the idea. Done poorly, it generates a false sense of confidence that costs far more than the testing budget ever saved.

Most teams treat market testing as a formality. A box to tick before the launch plan goes live. That instinct is expensive.

Why Most Market Testing Produces Comfortable Lies

I’ve sat in more product development reviews than I can count, and the pattern is almost always the same. Someone presents concept testing results showing strong positive sentiment. The room relaxes. The launch plan accelerates. And then, six months later, the sales numbers look nothing like the research suggested they would.

The problem isn’t the research itself. It’s what the research was designed to measure. When you ask someone whether they like a product concept, you’re measuring their imagination of the product, not their willingness to exchange money for it. Those are fundamentally different things, and conflating them is one of the most common and costly mistakes in new product development.

People are generous in research settings. They want to be helpful. They respond to well-designed stimulus material with enthusiasm that rarely survives contact with a real purchase decision. I’ve seen this play out across consumer goods, B2B software, financial services, and healthcare. The category changes. The dynamic doesn’t.

Good market testing is designed to surface disconfirming evidence. It’s structured to find the reasons a product might fail, not to validate the reasons it should succeed. That requires a different mindset from most internal teams, who have already invested months of energy and political capital in the product they’re testing.

The Difference Between Concept Testing and Market Testing

These two terms are often used interchangeably. They shouldn’t be.

Concept testing happens early. It’s exploratory. You’re trying to understand whether a problem resonates, whether your proposed solution makes sense to the people you’re building it for, and whether the language you’re using lands. It’s qualitative, directional, and relatively cheap. It should inform product development, not validate a launch decision.

Market testing comes later. It’s designed to simulate, as closely as possible, the conditions of a real purchase. You’re measuring actual or proxy behaviour, not stated preference. You’re looking at conversion rates, price sensitivity, repeat purchase intent, and the friction points that kill transactions. The closer you can get to real money changing hands, the more reliable the signal.

The distinction matters because the decisions each informs are different. Concept testing should shape the product. Market testing should shape the go-to-market. Conflating the two means you’re often making launch decisions on evidence that was never designed to support them. If you’re thinking about how market testing fits into a broader go-to-market and growth strategy, there’s more context on that in the Go-To-Market and Growth Strategy hub.

What Rigorous Market Testing Actually Looks Like

The gold standard is a controlled market test: a limited geographic or demographic rollout with real pricing, real distribution, and real marketing spend. You’re measuring actual purchase behaviour against a defined baseline. It’s expensive, slow, and logistically complex. It’s also the only method that gives you genuinely reliable commercial data before you commit to full-scale launch.

Most organisations don’t have the appetite for that. So they compromise. The question is whether the compromise still gives you a useful signal or just creates the illusion of one.

Some methods that sit between pure concept testing and a full market test:

Landing page tests. You build a page that describes the product, sets a price, and invites people to register interest or pre-order. Traffic is driven through paid channels. You measure conversion rates. This tells you something real about demand at a specific price point, in a specific channel, for a specific audience. It’s not perfect, but it’s far more honest than a focus group.

Fake door tests. You present the product as if it exists, measure how many people attempt to buy it, and then explain that it’s coming soon. Ethically, this requires transparency and follow-through. Commercially, it gives you a conversion signal without building the product first. Behavioural analytics tools like Hotjar can help you understand where users drop off and what friction looks like in practice.

Pilot programmes. You sell the product to a small, defined customer group at full price, with full support. You measure everything: acquisition cost, conversion rate, satisfaction, repeat purchase, and churn. A well-run pilot is the closest thing to a real market test that most organisations can execute within a reasonable budget and timeline.

A/B testing on proposition variants. If you’re uncertain about pricing, positioning, or messaging, you can run structured tests across these variables before committing to a single approach. This works best when you already have some distribution and audience access. It’s less useful for genuinely new products entering markets where you have no existing presence.

Price Sensitivity Is the Variable Most Teams Avoid Testing

I understand why. Pricing is politically charged inside most organisations. It touches margin targets, competitive positioning, and brand strategy simultaneously. Nobody wants to surface a finding that suggests the product can’t support the margin the business needs.

But price sensitivity is one of the most commercially important variables you can test. And avoiding it doesn’t make the problem go away. It just means you discover it at launch, when the cost of being wrong is much higher.

There are structured approaches to testing price sensitivity that don’t require you to run a full market test. The Van Westendorp Price Sensitivity Meter asks respondents four questions about price perception: too cheap, cheap but acceptable, expensive but acceptable, and too expensive. The resulting data gives you a defensible price range, not a single optimal price point, which is actually more useful for decision-making.

Gabor-Granger methodology tests purchase intent at different price points and produces a demand curve. It’s a blunt instrument, but it’s better than pricing by intuition or competitive benchmarking alone.

Neither of these methods is perfect. Stated willingness to pay consistently overstates actual willingness to pay. But they give you a directional signal that’s more reliable than asking whether someone “likes” a product at a price you’ve already decided on internally.

The Segmentation Problem in New Product Testing

One of the most consistent errors I’ve seen in new product development is testing against the wrong audience. Teams recruit participants who broadly fit the target demographic, rather than people who actually have the problem the product is designed to solve. The result is data that looks clean but measures the wrong thing.

Early in my career I worked on a product launch where the concept testing had been conducted with the right age and income profile, but the sample had no particular experience with the category problem the product addressed. The concept scored well. The launch underperformed. When we went back and spoke to people who genuinely had the problem, the feedback was completely different. They had questions the research hadn’t anticipated and objections the product hadn’t been designed to handle.

The lesson is straightforward: test with people who have the problem, not just people who fit the demographic profile. For B2B products, this means testing with people who have the actual authority to buy, not just the people who are easiest to recruit. The Forrester research on healthcare go-to-market challenges illustrates this well: even in sectors with clear clinical need, the gap between end-user enthusiasm and buyer decision-making can be significant.

Getting segmentation right in testing requires more effort upfront. It usually means smaller samples, higher recruitment costs, and longer timelines. It also means your data is actually worth something when you’re making a capital allocation decision.

What Qualitative Testing Can and Cannot Tell You

Qualitative research is underrated in new product development, but not for the reasons most people think. Its value isn’t in generating quotable enthusiasm. It’s in surfacing the language customers use to describe their problem, the mental models they bring to a category, and the objections they haven’t been asked directly but will raise when they’re deciding whether to buy.

When I was running agency teams working on new product launches, the most useful qualitative sessions were always the ones where we’d stripped the stimulus material back to almost nothing. No polished creative. No brand identity. Just a plain description of the problem and a plain description of the proposed solution. The less the stimulus material looked like marketing, the more honest the response.

What qualitative testing cannot tell you is whether enough people have the problem to make the product commercially viable. That’s a quantitative question, and it requires a different methodology. Qualitative research tells you why. It doesn’t tell you how many. Teams that treat focus group enthusiasm as a proxy for market size are making a category error that costs money.

The right approach is sequential. Use qualitative research early to shape the product and the proposition. Use quantitative research to size the opportunity and stress-test the commercial assumptions. Use behavioural testing, as close to a real purchase as you can get, to validate the launch model before you commit significant spend. Continuous user feedback loops can help maintain that connection to real customer behaviour beyond the initial testing phase.

When Market Testing Gets Skipped and What Happens Next

Timeline pressure is the most common reason market testing gets compressed or cut entirely. The launch date has been set, often for reasons that have nothing to do with market readiness: a trade show, a competitor move, an internal planning cycle. The testing budget gets reallocated to launch execution. And the team proceeds on the basis of concept testing data that was never designed to answer the questions that actually matter.

I’ve seen this pattern play out across large organisations and small ones. The scale changes. The dynamic is consistent. When the launch underperforms, the post-mortem almost always surfaces signals that were present in the earlier research but weren’t acted on because the team was already committed to the launch plan.

The cost of skipping market testing isn’t just the failed launch. It’s the inventory commitment, the channel investment, the marketing spend, and the organisational energy that goes into a product that wasn’t ready. Go-to-market execution is getting harder, and the margin for error on a poorly validated product launch is narrower than it used to be.

There’s also a subtler cost. Failed launches erode internal appetite for innovation. When a product fails publicly, the instinct is often to tighten governance and slow down the next development cycle. The lesson drawn is usually “we moved too fast” when the real lesson is “we tested the wrong things.” Those are different problems with different solutions.

The Demand Creation Problem That Testing Can’t Solve

Here’s something worth saying clearly: market testing can tell you whether existing demand exists for a product. It cannot tell you how much demand you can create through marketing.

This distinction matters more than most product teams acknowledge. If you’re entering a well-established category with a differentiated product, market testing is a reliable tool. There’s existing demand to measure against. But if you’re creating a genuinely new category, or solving a problem that customers haven’t yet articulated, testing against current behaviour will understate the opportunity.

Earlier in my career I spent a lot of time focused on capturing existing demand through lower-funnel performance channels. It looked efficient. The attribution was clean. But much of what we were measuring was demand that was going to convert anyway. The harder and more valuable work is reaching people who don’t yet know they need what you’re selling. That’s where growth actually comes from, and it’s much harder to test in a controlled environment.

The implication for market testing is that your results will be conservative if your product is genuinely innovative. Build that assumption into how you interpret the data. A product that converts at a modest rate in testing, against an audience that wasn’t previously aware of the category, may convert at a much higher rate once you’ve built category awareness at scale. Understanding how market penetration strategy works alongside product testing helps frame what testing can and can’t predict.

How to Design a Market Test That Actually Informs a Decision

The starting point is being explicit about what decision the test is designed to inform. That sounds obvious. In practice, most market tests are designed to generate data rather than to answer a specific question. The result is a dataset that’s interesting but not actionable.

Before you design the test, write down the decision you’re trying to make. Is it whether to launch at all? Whether to launch in this market versus another? Whether to price at point A or point B? Whether to lead with proposition X or proposition Y? Each of these decisions requires a different test design. Trying to answer all of them with a single piece of research produces answers that are too diluted to act on.

Then define your success criteria before you run the test, not after you see the results. If you’re running a landing page test, what conversion rate would give you confidence to proceed? What would give you pause? What would stop the launch entirely? These thresholds need to be set in advance, because if you set them after seeing the data, you’ll set them at whatever the data shows.

I’ve judged the Effie Awards, where effectiveness is the explicit criterion. The entries that stand out are always the ones where the commercial objective was defined precisely before the campaign ran. The same discipline applies to market testing. Vague objectives produce vague findings. Specific questions produce specific answers.

Finally, build in a mechanism to act on what you find. A market test that surfaces a problem but doesn’t change the launch plan is a waste of money. If the organisation isn’t genuinely prepared to delay, modify, or cancel a launch based on what the testing shows, you’re not running a market test. You’re running a validation exercise, and those two things are not the same. There’s more on how testing fits into the broader commercial framework in the Go-To-Market and Growth Strategy hub, which covers the full arc from strategy to execution.

The Organisational Conditions That Make Testing Work

Market testing doesn’t fail because of methodology. It fails because of organisational dynamics. The team has already committed to the product. The launch date is in the board deck. The agency has been briefed. In that context, testing becomes something to be managed rather than something to be learned from.

The companies that do this well treat market testing as a genuine input to decision-making, not a compliance step. That requires leadership that is genuinely comfortable with a test result that changes the plan. It requires product teams that are more attached to commercial outcomes than to the product they’ve built. And it requires a culture where surfacing a problem early is rewarded rather than treated as a threat to the project.

BCG’s research on go-to-market strategy and organisational alignment makes the point that commercial outcomes depend as much on internal alignment as on external execution. That’s particularly true in new product development, where the distance between the product team and the customer is often at its greatest.

If you’re working in an organisation where testing results are routinely reinterpreted to support the decision that was already made, the problem isn’t the testing methodology. It’s the decision-making culture. No amount of rigorous research design fixes that. What it takes is someone senior enough to make the results stick, and willing to act on what they show.

The growth strategies that actually produce results share a common characteristic: they’re built on honest assessment of what the market wants, not on what the internal team believes it should want. Market testing is one of the few mechanisms that forces that honesty into the development process, if you let it.

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