Information Architecture

Information Architecture

Concept of the skill

Information architecture is the discipline of structuring information so that people and agents can find, understand, and move through it. It is the contract between a user's goal and the system's structure: when the IA is wrong, good content and good components still feel confusing because the path to them is unclear. Strong IA always starts from tasks — name the top user goals and entry points first — and only then chooses structure, grouping content by what the user is trying to accomplish rather than by how the system is built internally. Every item then gets a deliberate placement (nav item, page, tab, section, filter, or cross-link) decided by frequency and task-entry rather than by raw importance, a label drawn from recognizable user language, and a wayfinding layer that keeps the user oriented (where am I, what is nearby, what is next, how do I recover from a wrong turn). The whole structure is validated the same way it was conceived — against real task scenarios and no-prior-knowledge discovery — because an IA that merely looks logical to its authors but cannot be navigated by a first-time user has not done its job.

Coverage

Structure information so users and agents can find, understand, and move through it. Covers navigation, sitemaps, hierarchy, page grouping, labeling systems, docs structure, cross-links, wayfinding cues, content models, and IA validation through real user tasks.

IA decisions resolve into a small set of recurring choices. Placement: for each piece of content, decide whether it is a top-level navigation item, a page, a tab within a page, a section on a page, a filter over a list, or a cross-link from a related location — governed by how often the item is used and whether it is a genuine task entry point, not by how important the team feels it is. Grouping: cluster content by user goal first and implementation detail second, so a user looking to "get set up" finds setup material together rather than scattered across product-feature silos. Labeling: name each group and destination in stable, recognizable user nouns, avoiding internal jargon and ensuring the same concept is named the same way everywhere. Wayfinding: every location communicates the current position, the sibling options at that level, the likely next action, and an escape path back to a known anchor. Hierarchy depth: navigation levels are kept limited and predictable, because deep or inconsistent nesting hides content as effectively as omitting it.

A recurring tension is the single-canonical-home rule: similar content should have one authoritative location plus cross-links from related places, rather than being duplicated into several homes (which fragments maintenance and confuses users about which copy is current). Empty states and low-data states are part of the IA, not an afterthought — the structure must still read clearly when a list is empty or a section has no data yet.

Method

1. Name the top user tasks and entry points.
2. Inventory current content or screens.
3. Group by user goal first, implementation detail second.
4. Decide structure: nav item, page, tab, section, filter, or cross-link.
5. Apply label discipline: recognizable user language, stable nouns, no internal jargon.
6. Add wayfinding: current location, sibling options, next action, and escape path.
7. Test against task scenarios and no-prior-knowledge discovery.

Philosophy of the skill

Information architecture is not decoration. It is the contract between a user's goal and the system's structure. If the IA is wrong, good content and good components still feel confusing because the path to them is unclear.

Good IA starts from tasks, then chooses structure. Do not promote every important thing to top-level navigation. Do not bury frequently used workflows under technically accurate but user-invisible categories.

The credibility of an IA comes from being tested, not asserted. A structure that reads logically to the team that built it routinely fails a first-time user, because the team's mental model is shaped by the system's internals while the user's is shaped by their goal. Restraint at the top level is part of the discipline: a navigation that promotes everything communicates nothing, so the practice is to earn each top-level slot against frequency and task-entry, push the rest down into sections, tabs, filters, and cross-links, and verify the result against real task prompts.

Verification

• Top tasks have obvious entry points
• Labels use user language rather than implementation terms
• Navigation levels are limited and predictable
• Similar content has one canonical home plus cross-links where needed
• Users can recover from a wrong turn without restarting
• Empty states and low-data states still preserve structure
• IA was tested against real task prompts

Do NOT Use When