Prototyping, Testing and Iterative Improvement
Why Designers Build Prototypes
A prototype is a test vehicle. Designers build prototypes not because they expect them to be perfect, but because physical and digital models reveal problems that drawings and calculations cannot. When a design exists only as a sketch or a CAD file, it is possible to miss details that become immediately obvious the moment someone holds or interacts with the product.
Prototypes range widely in fidelity. A low-fidelity prototype might be a rough card model made in an afternoon to check the proportions of a product or whether a handle fits a hand. A high-fidelity prototype is a near-production unit used in the final stages of testing — made from the correct materials, with working mechanisms, and used to gather final approval before tooling for manufacture. Both types serve a purpose; choosing the wrong fidelity wastes time and money.
Prototypes range from low-fidelity (rough models to test form or scale) to high-fidelity (near-production models to test function and finish). Lower fidelity is used earlier in the process when many ideas are still in play; higher fidelity is used later when a design is being finalised.
What Prototypes Are Used to Test
A well-designed test programme evaluates a prototype against the criteria in the design specification. Typical tests cover:
- Fitness for purpose — does the product do what the brief requires? A waterproof case that lets water in after 30 minutes is not fit for purpose regardless of its appearance.
- Functionality — do all mechanisms, controls, and components work correctly together? A folding mechanism that catches and jams fails a functionality test even if each individual part is dimensionally correct.
- Marketability — does the product appeal to the intended user? User panels may assess aesthetic appeal, perceived quality, and willingness to purchase.
- Safety — does the product meet relevant safety standards? Products for children must be tested against requirements such as choking-hazard rules under the EU Toy Safety Directive and EN 71-1 (the small parts test uses a cylinder 57.1 mm tall by 31.7 mm in diameter — any component that fits entirely inside this cylinder is classified as a choking hazard and must not appear in toys intended for children under 3).
Ergonomics
Ergonomics is the science of making products fit people rather than expecting people to adapt to products. An ergonomically designed tool is shaped to match the natural movements of the human hand, wrist, and arm, reducing the force required and the risk of repetitive strain. Ergonomics considers not just physical dimensions but also cognitive factors: a control panel with too many buttons, inconsistently labelled, creates cognitive overload and is ergonomically poor even if it is physically comfortable to touch.
When testing ergonomics, designers observe users interacting with a prototype in realistic conditions. A torque wrench that is comfortable to hold during light use may cause wrist fatigue after 30 minutes of heavy use — only testing under realistic conditions reveals this.
Anthropometrics
Anthropometrics is the collection and use of human body measurement data. Designers use percentile tables to design for the range of users they expect to encounter. A product designed for the 50th percentile fits the average user but excludes half the population at each extreme. Designing for the 5th to 95th percentile range is standard practice — this ensures the product works for 90% of the expected user population.
Common anthropometric measurements used in product design include hand grip span, finger length, shoulder width, eye height (standing and seated), reach distance, and foot length. Seat heights, door handle positions, keyboard dimensions, and steering wheel diameters are all derived from anthropometric data.
Designing for the 5th to 95th percentile means the product works for 90% of the expected user population. Designing only for the average (50th percentile) user excludes 50% of users at the taller, shorter, larger, or smaller extremes of the population range.
Testing and Feedback
User testing generates feedback that designers use to make informed decisions about changes. Effective user testing uses structured methods: task-based testing (asking a user to perform a specific action and observing where they struggle), think-aloud protocols (asking the user to describe what they are thinking while using the prototype), and rating scales (asking users to score specific attributes such as comfort, clarity, and visual appeal).
Feedback from user testing informs design decisions throughout the process, not just at the final stage. A designer who tests only a finished prototype and then modifies it is doing less effective design than one who tests a low-fidelity model early, identifies a structural problem, and revises the specification before any expensive materials are committed.
Iterative design follows a cycle: build → test → evaluate findings → modify → build again. Each cycle produces a design that is better informed than the previous version. The cycle continues until the product meets all specification criteria or until time and budget constraints require a decision to proceed with the best available version.
The key distinction is whether feedback genuinely informs decisions or merely documents what happened. A strong response identifies a specific finding from testing (e.g. “users with small hands found the grip diameter of 45 mm caused fatigue after 10 minutes”), the decision it triggered (“the grip diameter was reduced to 32 mm and a softer thermoplastic elastomer grip was added”), and how the next test confirmed the change worked. Feedback that does not change anything is not iterative design — it is observation.
Critical Analysis and Evaluation
Critical analysis of a design means systematically assessing how well it meets each specification criterion, identifying where it falls short, and explaining what evidence supports that judgement. A good evaluation is specific and based on evidence. “The handle was comfortable” is not an evaluation; “seven out of eight test users rated handle comfort above 4 out of 5, but the one user with smaller hands rated it 2, suggesting the lower grip limit should be reduced” is an evaluation.
Good evaluation also considers fitness for purpose holistically — a product may pass every individual specification test but still fail to feel right in use, because real use reveals interactions between criteria that individual tests do not capture.
Students often describe user testing as useful only because it “improves the final product.” This misses half the point. Feedback from user testing informs design decisions throughout the iterative process — changing the specification, modifying the approach, and redirecting resources before expensive commitments are made. A response that shows testing changing an early prototype specification is stronger than one that shows testing polishing a nearly-finished product.