Operations Workflows Workflow Intermediate

AI Website Structure Workflow

Organize a site so people and crawlers find things — inventory the content, group it into a real hierarchy, design the sitemap and navigation, then document the information architecture for the build.

The problem

A website's structure is decided whether you plan it or not — and unplanned, it becomes a pile of pages bolted on as they were needed, with navigation no one can follow and content buried three clicks from where people look. Information architecture is the deliberate version: what pages exist, how they group, how they nest, and how someone (or a crawler) moves between them. It's not visual design and it's not copy — it's the structure underneath both. This workflow owns that layer: inventory the content, group it into a hierarchy that reflects real relationships, design the sitemap and navigation, and document it so the build inherits a map instead of inventing one.

Recommended workflow

Each step uses an existing NewPrompt tool, pre-filled by a matching resource. Open the resource to read it, or jump straight into the tool with the inputs ready.

  1. Anchor as an information architect

    Put the model in an architect's seat aimed at information, not code or visuals — reasoning about hierarchy, grouping, and findability. The job is the structure underneath the site, so the framing has to be structural.

    Outcome The model reasoning about hierarchy and findability, not page design.
  2. Inventory and group the content

    List everything the site needs to hold, then group it by how the content actually relates — not by who owns it internally. Good grouping is what makes a site feel obvious; bad grouping is what buries pages.

    Outcome All the site's content grouped into a logical hierarchy.
  3. Lay out the sitemap and navigation

    Turn the groups into a sitemap — the pages, how they nest, and the relationships between them — and the navigation that lets people and crawlers reach any page without guessing. This is the structure the build will follow.

    Outcome A sitemap with a navigation hierarchy and clear page relationships.
  4. Document the IA for the build

    Capture the architecture — the page tree, the navigation, the rules for where new content goes — in a spec the build and content teams share, so the structure survives past the planning session.

    Outcome An information-architecture spec the build and content teams work from.

Expected outcome

A website's information architecture designed before the build — content inventoried and grouped, a sitemap with navigation and page relationships, and an IA spec the team works from — so the site is something people and crawlers can navigate instead of a pile of pages bolted on over time.

Best for

  • Designing a website's sitemap and navigation before building
  • Organizing content into a hierarchy that reflects real relationships
  • Documenting information architecture for a build and content team

Not for

  • Deciding what content to publish and why — use the AI Content Strategy Workflow
  • Writing the conversion copy on a page — use the AI Landing Page Copywriting Workflow
  • Designing software architecture — use the AI Project Architecture Workflow

FAQ

Is this a UX or visual-design workflow?

No. This owns information architecture — what pages exist, how they group and nest, how navigation connects them. It's the structure underneath the site, decided before visual design and copy. Those layers build on this one; they don't replace it.

AI website structure vs project architecture workflow — which one do I need?

Use website structure for a site's pages, hierarchy, and navigation; use AI Project Architecture for software — services, data flow, and boundaries. This one outputs a sitemap and IA spec; that one outputs a system design. Same architect framing, completely different artifact and deliverable.

AI website structure vs content strategy workflow — what's the difference?

Content strategy decides what to publish and why; this decides how it's organized into a navigable site. Strategy hands you topics and pieces; structure inventories, groups, and nests them into a sitemap. Run strategy first, then design the architecture that holds its output.

What does the AI website structure workflow output?

Three artifacts you produce by running its prompts: a grouped content inventory, a sitemap with navigation hierarchy and page relationships, and an information-architecture spec the build and content teams work from. NewPrompt supplies the prompts and step order; you run them and own the result.

How do I run the AI website structure workflow?

Work the four steps in order in your own AI tool: anchor the model as an information architect, inventory and group the content, lay out the sitemap and navigation, then document the IA. NewPrompt links each step's prompt and resource; you run them and review every output.

What inputs does the website structure workflow need before I start?

A list of everything the site must hold — pages, sections, and content it needs to carry, ideally the output of a content strategy pass. From that raw inventory the workflow groups by real relationships, nests into a sitemap, and documents the architecture. Without the content list, step two stalls.

Part of these projects

Complete build journeys that include this workflow as a stage.

Recommended next workflow

Tip: Each step's resource opens its tool pre-filled — start at step one and carry the output forward.