New Product Development Testing in Short:
New product development testing is the process of gathering feedback from real consumers at each state of a product’s journey. This means from early concept all through post-launch. The most common methods to get these insights include concept testing, in-home usage tests (IHUTs), central location tests (CLTs), online product surveys, and focus groups. The method you choose is dependent on the development stage you’re in and the decision the data needs to validate.
Getting a new product to market is expensive. Getting one to market only to discover consumers do not want it is even worse.
By that point, your team may have already invested significant time and money into product development, and making meaningful changes means losing more resources before you’ve gained capital back. But that’s what new product development testing works to solve.
New product testing development helps companies reduce the risk of getting things wrong before their product even hits shelves by giving them direct feedback from the consumers they are trying to reach at every stage.
Instead of relying solely on internal opinions, assumptions, or their own research, they are able to get reliable, robust outside feedback on which of their product ideas are worth pursuing, where improvements are needed, and when it’s time for them to leave a concept behind.
In this guide, our product testing research company walks through the five stages of new product development, what to test at each stage, and the specific methods, metrics, and outputs your team can use to make confident go/no-go decisions.
New product development testing (sometimes shortened to NPD testing) is the structured process of collecting consumer feedback at each stage of a product’s development lifecycle. The goal is to validate ideas, identify problems, and refine the product before your team makes its largest investments in production, marketing, and launch.
NPD testing sits within a larger stage-gate process that most brands are already familiar with: ideation, screening, development, market testing, and launch. Research runs at every one of those gates, and the questions being asked change as the product matures.
By the time a product is nearing launch, research is focused less on whether the idea works and more on whether every element of the customer experience is ready for market. NPD validates this and keeps businesses confident in their decisions.
For example, a company developing a new beverage may be deciding between several flavors, packaging designs, or positioning statements. Rather than making assumptions through every stage and not finding out much later they had made missteps, new product testing helps set up the development process for success with reliable findings, and will work to do so throughout the development journey.
Key Takeaway: The goal of NPD is ultimately identifying problems before they’re expensive to fix. Refining a concept or rethinking pricing is easier (and much less stressful!) to do before manufacturing has begun or once a product reaches shelves.
Most product development processes move through five phases. Each one calls for different research, and each one has a specific window where testing pays off the most. The best NPD programs treat this as an ongoing research process across the full product lifecycle, not simply a single pre-launch check.
Here’s the full lifecycle at a glance:
| Stage | Primary goal | Key questions | Best-fit research |
| Ideation | Generate directions worth pursuing | Where are the unmet needs? What language do consumers use? | Focus groups, IDIs, online communities |
| Concept screening | Fund the winners, kill the losers | Which concepts should we develop? Which should we shelve? | Concept testing survey (monadic or sequential monadic) |
| Development | Validate real-world product performance | Does the product actually work in real use? | IHUTs, CLTs, prototype focus groups |
| Market testing | Confirm launch readiness | What’s the real-world purchase intent at the real price? | Large-sample survey, simulated shelf, test market |
| Post-launch | Optimize for growth | How is the product performing? What needs to change? | Brand tracker, satisfaction survey, review analysis |
This is the messy front end. Your team is generating ideas, identifying whitespace in the category, and figuring out which directions are worth developing further.
Research at this stage is exploratory. Focus groups, in-depth interviews, and online communities help surface unmet needs and consumer language that may otherwise go unnoticed. The output is a shorter list of directions worth testing more formally.
Once you have a list of ideas, you need to know which ones deserve investment. A well-designed concept test measures appeal, purchase intent, uniqueness, and relevance across each concept. Some sail through while others are shelved. The ones in the middle get refined and tested again.
In our experience, this is the stage where the most value can be created or lost in the entire NPD process. Identifying an underperforming concept early can save months of development time and financial investment that could’ve been lost if you moved forward with the wrong concept.
A recent example: A regional bank came to us to measure demand for a new digital banking product before committing to build it.
Our team ran an online survey with 1,000 respondents and paired it with a competitor analysis.
Within two weeks, the bank had clear evidence of which features customers actually valued, what gaps existed in the market, and how to position the offering to stand out. Leadership moved forward on the features that scored highest and shelved the ones that didn’t.
Now you have a real product being built. Research at this stage focuses on functional performance.
In-home usage tests are our top recommendation here. They put real prototypes into real homes for days or weeks, and they surface the practical issues no lab test will catch, allowing for more refinements that improve the customer experience before launch.
Market testing is where you evaluate the full launch package: product, price, packaging, and positioning. Sometimes this looks like a limited regional launch. More often now, it looks like a large-sample quantitative study that measures purchase intent against real competitors at the real price point.
Test marketing has been a mainstay of NPD since at least the 1970s, when Harvard Business Review published foundational work on the practice. The core logic still holds. Understanding how consumers respond before a full-scale rollout is key for reducing uncertainty.
Testing doesn’t stop at launch. Post-launch research measures how the product is actually performing against expectations, what customers love, and what needs to be adjusted for the next iteration.
This is often where brand tracking, customer satisfaction surveys, and repeat-purchase analysis take over. The goal shifts from “should we launch this?” to “how do we grow this?”
Useful Resource: Explore this case study to see how we helped a financial institution measure demand and reduce risk before launching a new banking solution.
New product failure rates are startling.
The truth is somewhere in that range depending on how “failure” is defined, but the direction is unambiguous. Most new products don’t make it once they hit the market for a variety of reasons, but they all come back to just not meeting the needs of consumers.
The research on why is fairly consistent. Products fail because they solve a problem consumers don’t actually have, or because they solve a real problem in a way consumers don’t want, or because the packaging and pricing don’t resonate with buyers. All three of those failures can often be pointed out and addressed long before launch even happens.
Even the best-tested products sometimes flop.
Market conditions shift, competitors move, and consumer sentiment changes. Research doesn’t guarantee success, but it does help ensure your product decisions are grounded in proven feedback instead of intuition. That’s the difference between launching with confidence and being caught off guard by the results.
The other reason research pays off at every stage is that decisions compound.
A weak concept doesn’t get better with a great package. A great product with the wrong price positioning still misses. Each phase of development builds on the decisions that came before it, making early research one of the most valuable investments a product team can make in terms of cost and efficiency.
Concept testing is the single most important research method in NPD. It’s the process of putting an early idea (a product, a service, a claim, or a package) in front of your target audience and measuring their reaction before you commit to development.
A concept can be as simple as a written description or as detailed as a product mockup with pricing and packaging.
What matters is that it clearly captures the core promise so consumers can respond in a meaningful way.
The output of a concept test is a set of actionable insights your team can use to make real, people-informed decisions.
Those questions have answers when the concept test is designed correctly.
There are two main ways to run a concept test, and picking the right one changes what your data can tell you. Here’s how they compare:
| Type | How it works | Best for | Trade-off |
| Monadic | Each respondent sees one concept in isolation | Final go/no-go decisions, real-world purchase modeling | Needs a larger overall sample |
| Comparative | Every respondent sees every concept, rates or ranks them | Fast, cheap screening of a longer list | Produces artificially clean rankings |
| Sequential monadic | All respondents see all concepts, one at a time, randomized order | Getting both monadic and comparative reads in one study | Longer survey, higher drop-off risk |
| Proto-monadic | Evaluate each concept independently first, then compare | Comprehensive read on standalone appeal and preference | Most complex design of the four |
Monadic is the cleanest design for measuring how a concept would perform on its own, because that’s how consumers actually encounter products. Nobody sees three side-by-side options on the shelf. They see one.
Pro Tip: For final go/no-go decisions, monadic is the safer choice. For early screening of a longer list, comparative gets you there faster and cheaper. Sequential monadic is a good compromise when the study budget allows.
Most concept tests measure four core metrics, and each one answers a different business question:
| Metric | What it measures | What weak scores mean |
| Appeal | How much consumers like the idea overall (usually a 5- or 7-point scale) | A positioning problem, a product problem, or both |
| Purchase intent | How likely they’d actually buy it (top-two-box combines “definitely” and “probably”) | Real-world sales are unlikely to materialize |
| Uniqueness | Whether the concept feels meaningfully different from what’s already out there | Differentiation isn’t landing, even if appeal is strong |
| Relevance | Whether the concept solves a problem your audience actually recognizes | The idea is clever but doesn’t fit anyone’s routine |
Generally, a concept that performs well across all four metrics moves forward in the testing process. Those that fall short on a single metric can usually be reworked or salvaged. Concepts that struggle across two or more metrics are less likely to move forward without major changes.
Concept testing is as much about rejecting unviable ideas as it is about refining good ones. In our experience, teams that view kill decisions as wins tend to launch fewer products, but the ones they do launch perform far better than those developed from teams that develop everything.
In-home usage tests are one of our favorite methods for new product testing, especially for consumer goods. IHUTs place a real product in a real home for a set period and observe how it performs under the conditions in which it’ll actually be used.
IHUTs catch the practical stuff that lab tests miss.
A lid that leaks when you set the bottle down. A cleaning spray that works great but smells too strong for a studio apartment. A user manual nobody actually reads.
Those details decide whether a product delights or disappoints at scale. These may seem minor, but they’re often the details that can make or break a product at scale.
We recently ran an IHUT for a home fragrance manufacturer developing a new nebulizer, ultrasonic diffuser, and wax warmers.
The client shipped us prototypes. We recruited 40 participants who matched our target consumer profile, with 30 to 35 participants completing a four-week test with daily usage logs.
The feedback surfaced leaking around the siphon, unexpected noise levels, cycle time issues, and a setup process consumers found frustrating. The client used those findings to make functional changes before moving to a market-ready product with branded packaging in phase two.
Central location tests bring consumers into a controlled location (a mall intercept, a research facility, a client office) to interact with a product under observed conditions.
Where IHUTs win on realism, CLTs win on control.
In our experience, CLTs are most often needed when the product experience is sensory, and the client needs every respondent to evaluate the same thing. Beverage taste tests. Fragrance comparisons. Texture and mouthfeel work on personal care products.
The moment you introduce variability (different serving temperatures, different environments, different times of day), you compromise your ability to say formula A actually beat formula B accurately.
Online surveys scale, move quickly, and deliver statistically projectable numbers on the metrics that matter most for NPD: purchase intent, price acceptance, feature preference, and competitive positioning.
For concept testing specifically, online surveys with a properly sized sample are the standard quantitative method. Well-designed surveys pair scaled ratings with open-ended follow-ups so you get both the numbers and the reasoning behind them.
Focus groups earn their place in NPD during the early exploratory phase, before your team has committed to a product direction. A well-moderated group of 6 to 10 target consumers can surface the language they actually use to describe the category, the unmet needs they already recognize, and how they react in real time to concept boards or product mockups.
Later in the process, focus groups add depth to quantitative findings. If a concept test shows one variation scoring well without a clear reason, a follow-up focus group can show why.
Useful Resource: Use our guide, Product Testing Research for Consumer Goods, to learn more about the most effective testing methods for consumer goods and better meeting their expectations.
The most common question we get from teams considering NPD research isn’t “should we test?” It’s “which method should we use?”
The answer depends on two things:
Different research methods are designed to answer different questions, which is why choosing the right approach is just as important as running the study itself.
The stages table earlier in this guide covers most of the decisions.
But there’s a second consideration worth thinking about: realism versus control.
Some methods prioritize understanding how consumers naturally interact with a product while others provide a more controlled environment.
IHUTs deliver the highest realism (real product, real home, real time) but the lowest control, since every household is different. CLTs flip that. They offer a standardized setting that makes it easier to compare consumer reactions under the same conditions.
The Key Takeaway: Strong NPD programs use a multi-method approach. IHUTs help teams understand real-world fit and performance, while CLTs narrow in on specific product attributes before launch.
A few patterns recur in NPD projects that go sideways. The good news is that most are avoidable. Below are the five most common ones we see:
What is concept testing in new product development?
Concept testing is the research process of showing an early product or service idea to your target audience and measuring their reaction. It typically measures appeal, purchase intent, uniqueness, and relevance, and helps product teams decide which concepts to develop further, refine, or shelve before making larger investments in development and production.
When should you start testing during the product development process?
As early as possible. The most common and most expensive mistake in NPD is testing after development is nearly complete, leaving no room to act on what the research reveals. Rather than treating research as a single checkpoint, the strongest product development programs get feedback throughout the entire lifecycle. Ideally, you’re testing at ideation, again at concept screening, again during development, again pre-launch, and again post-launch.
What is the difference between concept testing and product testing?
Concept testing evaluates an idea. Product testing evaluates a physical product. Concept testing is typically conducted using descriptions, mockups, or storyboards, and it assesses whether the overall concept resonates. Product testing places the actual product in consumers’ hands (through IHUTs, CLTs, or sensory testing) and measures its performance in use. Most NPD programs run at different stages.
How many concepts should be tested at once?
Two to five is generally a workable range for most concept tests. You need enough for a comparison, but too many options can lead to respondent fatigue and require larger sample sizes. If you have a longer list, consider using an initial screening study to narrow your concepts before moving the top two or three into a more rigorous monadic test.
What sample size is needed for new product testing?
Sample sizes depend heavily on the method, target audience, and the level of confidence your team needs in the results. Concept testing surveys often include a few hundred respondents per concept to obtain reliable data. IHUTs and CLTs work with smaller samples of around 30 to 1oo participants, while focus groups run 6-10 participants per session, with multiple groups conducted for each key audience segment when needed.
Guiding smarter business decisions through new product development starts with early concepting, and goes beyond post-launch. At Drive Research, we run studies at every stage of the new product development lifecycle and provide clients with honest, reliable feedback when a concept is ready to develop, when it needs work, and when it needs to be shelved.
Our goal isn’t to simply collect research, but to help you make better product decisions before your large investments can become negatively impacted. Ready to validate your new product with confidence? Contact Drive Research today to get started!
