User research in product development: the steps to a successful study

A brilliant product idea remains theoretical until you know whether the end user can and will actually use it in practice. Many innovations fail not because they are technically unfeasible, but because they solve a problem nobody experiences, or because the operation fails in the real world where the lab model still worked perfectly. User research in product development is therefore not about collecting opinions, but about systematically removing uncertainty before you invest in expensive moulds and mass production.

Manufacturable innovation only emerges when user insights directly guide technical choices, material selection and assembly sequence. Research in this process is not a separate phase you tick off before kickoff, but a continuous validation stream that moves with every step from concept to physical product. Without that link you may build a technically perfect device that nevertheless fails on the market because it does not match the user’s physical or cognitive capabilities.

Why user research is the foundation of successful product innovation

Innovation projects often get stuck on the gap between what stakeholders assume and what users actually do. Teams can spend months developing features that turn out to be irrelevant in real use. Assumptions are necessary to start a project, but they must never be the sole basis for final design decisions or investments in tooling. Validated user needs replace speculation with facts and give the engineering team a concrete framework to design within.

The difference between assumptions and validated user needs

You too often base product decisions on internal convictions instead of external evidence. A stakeholder may be convinced that a touchscreen is the right interface, while observations show that users wearing gloves need physical buttons. That insight changes the entire electronics architecture and housing design before a single PCB is ordered. Validation here does not mean asking people what they want, but testing what they can actually do under realistic conditions.

Risk reduction in the development process through early validation

The Nielsen Norman Group states that every dollar spent on early-stage user research saves on average ten to a hundred times that amount in corrections during production and after launch. That saving comes from a simple fact: errors in the concept phase cost hours, while the same error after mould production causes weeks of delay and tens of thousands of euros in tooling changes. Early validation is therefore not an expense, but insurance against failure costs that rise exponentially as the project progresses.

Integrating user research into every phase of product development

Research should progress in step with the product’s technical maturity. Each phase answers specific questions directly tied to the Technology Readiness Level scale and the associated decision points for manufacturability. PEZY integrates user validation by default into the Technology Readiness Level process, so that insights about operation and ergonomics directly influence the choice between injection moulding and 3D printing for series production. If research is detached from those technical milestones, insights linger in reports without impacting the final product.

Exploratory research during the idea phase and problem definition

In the lowest TRL phases the focus is on sharpening the problem and its context. You observe users in their own environment to discover latent needs they would not mention in an interview. This phase does not produce specifications, but a validated problem statement that prevents you solving the wrong problem. If you skip this exploration you risk developing a solution for a symptom instead of the root cause.

Formative testing during design and prototyping

Formative research is at the heart of the iterative design process and takes place while the product is still in development. Rapid prototypes serve as tangible discussion points to continuously adjust assumptions about interaction and ergonomics, using rapid prototyping for quick iteration to get feedback on shape, weight and operation within days. The aim is not perfectionism, but learning: each test result feeds directly into the next design cycle and refines the technical requirements.

Summative validation prior to mass production

Summative validation confirms that the final design meets all specified usability and safety requirements before production starts. This is the moment to statistically demonstrate that the product works as intended for the entire target group. Where formative research looks for improvements, summative research seeks evidence of suitability. Without this final check you enter production with a residual risk that can later lead to recalls or failed certification processes.

Methods that lead to manufacturable designs

Academic purity does not help an engineer. You must select methods for their ability to deliver direct input to technical specifications and manufacturability optimisation within available time and budgets, and a method is only relevant if its output can be translated into CAD parameters, material properties or assembly steps. Theoretical models that do not bridge to the drafting table remain dead letter in an innovation project.

Qualitative methods for deep insight into the usage context

Contextual inquiries and task analyses reveal the physical and environmental factors that determine design. You see how light incidence, vibration, dirt or stress affect operation in ways no laboratory test can simulate. This qualitative data translates directly into requirements for surface roughness, grip, contrast ratios or IP rating. It is precisely these ‘soft’ observations that define the ‘hard’ technical constraints.

Quantitative validation for scalability and statistical certainty

When the design is stable, quantitative tests provide the certainty needed for series production. Measurements of task times, error rates and applied force give objective limits for tolerances and performance requirements. This data prevents overdimensioning out of caution or underdimensioning out of optimism, and statistical significance gives management and certifiers the confidence that the product is robust enough for the market.

The combination of data and observation for technical translation

The true value lies in the synthesis of what people say, what they do and what the sensors measure. A user may report that a handle ‘feels nice’, but force sensors may show that the required grip force after twenty minutes leads to fatigue. Only by combining these layers do you arrive at an ergonomic design that remains functional after prolonged use and that distinguishes manufacturable innovation from cosmetic improvements.

From user insight to technical specifications and DfM

The crucial step in product development is translating subjective user experiences into objective engineering requirements that are directly applicable in the Design for Manufacturing guide and production planning. A remark like ‘the device feels cheap’ is worthless to a designer, but translated into ‘surface material must have a non-slip coating with Ra > 3.2 µm’ it becomes a specification. That translation requires researchers and engineers to speak the same language and work together in one team.

Translating user needs into measurable engineering requirements

Every user need must end up as a verifiable parameter in the requirements set. ‘Easy to clean’ becomes a requirement for gap widths, chemical resistance and surface energy. Without that concretisation, DfM remains a vague ambition rather than a measurable result, and the engineer can only optimise for production when they know exactly which user requirements must be guaranteed within the mould constraints.

Impact of user behaviour on material and process choice

During the development of a portable infusion pump, contextual research showed that nurses often had to operate the device one-handed. That led to a complete redesign of the housing and button layout before the moulding phase. If this had only been discovered after the first pilot run, the mould would have been unusable and the project would have suffered months of delay. User behaviour therefore not only dictates form but also partly determines whether you choose injection moulding, thermoforming or machining.

Feedback loops between research and iterative design

Research does not stop at the first prototype but continues until the mould is ready for production. Each iteration raises new questions that can only be answered by testing again with adjusted models. This cycle ensures manufacturability and usability evolve in parallel rather than sequentially, preventing the classic pitfall of a perfectly tested lab model that proves unproducible in the factory.

Specific challenges for regulated products and medical devices

For medical and regulated products, user research is not optional but a legal obligation intrinsically linked to patient safety and market access. According to IEC 62366, usability engineering is an integral part of the risk management process for medical devices, where use errors must be systematically analysed and mitigated. Those who ignore this process or try to fix it afterwards risk not only rejection by the notified body but also endangering patients.

Usability engineering as a mandatory part of risk management

Usability research for medical devices coincides with safety validation and must be carried out early in the project to avoid costly re‑certification. Use errors are treated as potential hazards weighted as heavily as electrical safety or biocompatibility. That means you must demonstrate from the concept phase that you have identified and controlled use errors through design measures. More on these specific requirements can be found in our explanation of product design for medical devices.

Documentation requirements for validation under IEC 62366

The standard requires a traceable dossier linking known hazards, usability requirements, validation tests and residual risks. Each test report must be traceable to a specific risk in the risk management file. If that traceability is missing, the research will not count toward certification no matter how thoroughly it was conducted. Documentation here is not an administrative burden but the evidence that safety was designed, not merely tested.

Common mistakes in user research within product teams

Even teams with good intentions make mistakes that result in attractive reports but no better product. Often this happens because teams treat research as an external service rather than a core competence of the development team. The most damaging mistake is isolating research from technical reality, so insights never land in the CAD model or production plan. Manufacturability requires research and engineering to coincide in time, space and responsibility.

Starting validation too late in the development process

Teams often wait until there is a working prototype before involving users, out of fear of negative feedback or because they want the technology to be ‘finished’ first. This delay makes validation expensive and painful, because fundamental changes then render earlier engineering efforts wasted. Prefer to validate rough concepts with low‑fi mockups rather than refined prototypes with high sunk costs. Early feedback is cheap; late feedback is a crisis.

Isolating research from the technical team

If researchers present their findings in a separate session without engineers present or involved in the analysis, nuance is lost in the handover. Engineers need to see a user struggle with a mechanism to feel the urgency of a design change. Integrate research into the daily development rhythm and ensure insights are discussed directly in technical reviews. Only then does research become a driving tool rather than a retrospective report.

Working with a product developer: research as an integral component

User research only yields real return when embedded in an end‑to‑end development trajectory in which design, engineering and production collaborate from day one. In a fragmented process insights drown in handover moments between agencies and suppliers, whereas an integrated partner like PEZY ensures every observation directly reaches the technical design and production planning. This coherence is not a luxury but a necessity for complex physical products where user experience and manufacturability are inseparable.

However, this research is not a universal solution to every business problem. It validates the interaction between human and product but does not solve strategic market positioning or business model questions. Teams stuck on assumptions about their product concept will find a structured starting point to test those assumptions in a Break-through Session before the full development trajectory begins.