Designing for Usability: Key Principles and What Designers Think
Source: https://dl.acm.org/doi/10.1145/3166.3170 ↗
Full text: open-access via Unpaywall ↗
Gould and Lewis's 1985 paper is the founding document of usability engineering — three principles that remain the practical minimum for designing interactive systems: early focus on users and tasks, empirical measurement, and iterative design.
The paper is short and surprisingly modern; everything called "user-centred" or "design thinking" since carries its DNA.
For product direction, reading the original is clarifying — most contemporary frameworks are elaborations on these three principles, and seeing them stated plainly is a useful reset.
A forty-year-old paper that has aged better than most contemporary product literature.
Central argument
Gould and Lewis argue that usable systems require three non-negotiable principles: early and continuous direct contact with actual users (not proxies or documents), empirical measurement through observed use of prototypes, and iterative redesign based on those findings. Their empirical contribution is equally pointed: when 447 working designers and developers were surveyed, only 2% mentioned all four principles unprompted, and just 20% mentioned iterative design at all — demonstrating that these principles are neither intuitive nor commonly practiced, even among professionals who believe they follow them. The deeper claim is that designers routinely confuse superficially similar practices — reviewing designs with users, getting sign-off, reading user profiles — with genuine user-centered design, and this confusion is itself a primary obstacle to usable systems.
Critique
The study's empirical foundation has a significant methodological ceiling: asking designers to write down steps before a human factors talk measures recall and unprompted articulation, not actual design behavior or outcomes. A designer who cannot name a principle in a pre-talk exercise might still apply it competently in practice, meaning the gap between stated principles and real-world usability quality is asserted but not demonstrated. Furthermore, the paper treats the three principles as largely context-independent, yet the feasibility and cost of continuous user access, iterative cycles, and empirical measurement vary enormously across product types, organizational constraints, and development timelines — tensions the authors acknowledge only briefly and do not resolve.
Why it matters for product
For a CPO, the most operationally sharp insight is Gould and Lewis's distinction between user involvement that merely generates agreement — sign-offs, reviews, user representatives on analysis teams — and user involvement that actually shapes design from inception. This maps directly onto a common organizational failure mode: discovery processes that are structurally downstream of key product and technical decisions, functioning as validation theater rather than genuine input. The paper also provides a defensible rationale for investing in continuous access to real users as infrastructure, not a project-level cost, since the authors show that even professionals with human factors awareness systematically underweight iteration and empirical measurement when left to their own design instincts.