The Elements of User Experience: User-Centered Design for the Web and Beyond
Source: https://www.jjg.net/elements/ ↗
Garrett's five-layer diagram — strategy, scope, structure, skeleton, surface — became the canonical way UX is taught and discussed, and it did so because it solved a real communication problem: designers, developers, and product managers were talking past each other by conflating decisions that belong at different levels of abstraction.
The book gave the field a shared vocabulary for distinguishing business objectives from content requirements from information architecture from visual design.
Its influence is so pervasive that many practitioners use the framework without knowing its origin.
The limitations are real — the layers suggest a cleaner separation than practice allows, and the model is better at describing than prescribing — but as a mental map for organizing the chaos of product decisions, nothing simpler has replaced it.
Read it to understand the scaffolding beneath every UX conversation you have ever had.
Central argument
Garrett argues that the failures of early web design stemmed from conflating decisions that operate at fundamentally different levels of abstraction, and that a shared vocabulary could prevent this. He proposes a five-layer model — strategy, scope, structure, skeleton, surface — where each plane depends on decisions made in the one below it, forcing teams to resolve business objectives and user needs before committing to interaction patterns or visual form. The book's central claim is that disciplined separation of these concerns is not just a design methodology but a communication infrastructure for cross-functional teams.
Critique
The layered model implies a sequential, top-down dependency that rarely survives contact with how product decisions actually get made — strategy is frequently revised by surface-level user feedback, and structure emerges iteratively rather than being specified in advance. This descriptive clarity becomes prescriptive rigidity in practice: teams that internalize the framework too literally can use it to police process rather than improve outcomes, treating the diagram as a workflow rather than a conceptual lens. The model also predates the dominance of continuous delivery and rapid experimentation, so it has limited guidance for contexts where scope and strategy are in constant negotiation.
Why it matters for product
For a product leader, the framework's real leverage is organizational: it names the layer at which a given disagreement is actually happening, which lets you identify when a debate framed as a design argument is really an unresolved strategy question. When engineers push back on scope or designers contest requirements, Garrett's vocabulary surfaces whether alignment has broken down at the objective level or the structural level — a distinction that changes who needs to be in the room and what decision needs to be made. It also provides a diagnostic for why handoffs fail: misalignment between layers is often the hidden cause of rework that gets attributed to execution problems.