Start with what people need to do.
A simple method for making design decisions traceable and easy to revise.
Observe the problem
State the intended audience, the important questions and the surrounding constraints. Collect examples of the content before deciding on a visual direction. A false premise produces polished but unhelpful work.
Build a limited system
Choose a small set of layout primitives, scale rules and typographic roles. Test them with inconvenient content: long titles, short labels, empty sections and difficult screen sizes.
Test and refine
Examine legibility, interaction states and keyboard navigation where relevant. Ask whether a visitor can identify the next useful action without being told where to click.
Document the decisions
Write down the assumptions, rules and known limitations. This helps a future editor or developer adapt the design without having to rediscover its logic.
A useful process includes at least one pass with awkward content. Long headings, dense paragraphs and narrow layouts reveal constraints hidden by a polished mock-up. The design should still have a sensible reading order without decoration. When an interaction cannot work in the published concept, replace it with a clear explanation rather than presenting a false control. Record which observations are genuine and which are proposed checks.
A practical way to use this page is to write down the behaviour described by “process”, then compare it with one observable example. Keep the actual result distinct from the assumption that motivated the test. When something differs, inspect the smallest relevant part before changing other settings.