ia-evaluationlisted

# Information architecture evaluation Chooses and runs the study that answers a structure question, and turns the numbers into a grouping and labelling decision. Structure definition itself (entities, relationships, content models) belongs to `silver-structure`; service-level mapping to `service-blueprint`; the words on the labels to `better-writing` once chosen. ## Inputs - The question: "how should these N items be grouped", "can users find X in this tree", or "which of two navigations works better". - The content inventory: the items or the proposed tree, with any constraints (items that must stay together, regulatory sections). - Access to participants, or the instruction to plan for later recruitment. ## Workflow ### 1. Match the question to the method Read [references/methods.md](references/methods.md) for the decision table. In short: - **Open card sort** when the grouping and the labels are unknown: users make and name groups. Generative. - **Closed card sort** when categories exist and the question is whether items fit them. Evaluative. - **Hybrid sort** when categories exist but may be incomplete. - **Tree test** when a hierarchy exists and the question is findability: participants locate tasks in a text-only tree, so labels and structure are tested without visual design. - **First-click test** when a page design exists and the question is whether the first choice on that page is right. Do not sort more than about 60 cards; split by area instea