About Evaluation

Arranging a user test with actual users can be costly in terms of both time and money. Therefore, it is not always necessary to test every small change with real users. In many cases, one or two design experts can identify issues that would otherwise require extensive user testing to uncover. For example, when designing a user interface for a game, established principles help ensure clarity and usability. A skilled designer can review these principles and verify their application before conducting a larger user test.

It is important to determine the appropriate type of test to conduct. Similarly, when testing a prototype with real users, a large sample size is not always necessary to gain valuable insights. In most cases, 5-6 testers can help identify 80%-90% of the issues in a prototype—after the sixth tester, most newly identified problems tend to be repetitions of previously discovered ones. Therefore, a smaller evaluation with 5-6 users is well-suited for a low-fidelity (lo-fi) or mid-fidelity (mid-fi) prototype. Once the prototype reaches a high-fidelity (hi-fi) stage, a larger evaluation phase with more users becomes more appropriate.

Test quality often improves when qualitative and quantitative methods are combined. For example, user tests conducted through surveys or questionnaires primarily yield quantitative data. However, pairing these with interviews can provide valuable qualitative insights. This combination creates a strong foundation for design decisions in the next iteration. That said, the choice of testing methods depends on the stage of the design and development process—conducting a full-scale questionnaire and interview study may be excessive in the early stages of product development.