Information architecture is the part of a website nobody notices when it works. When it fails, customers call to ask where the price list is.
What is information architecture?
Information architecture is the plan for how content is organized, labeled, and connected so people find things without thinking about it. It answers four questions for every page: what group it belongs to, what it is called, where it sits, and what links to it. The term is old and a little stiff. The job is plain: put things where customers expect them, and call them what customers call them.
Is a sitemap the same thing?
No. A sitemap is one picture of the information architecture: the list of pages and how they nest under each other. It is the most visible piece, and the one most people mean when they say "IA."
A full information architecture also covers:
- Taxonomy. The categories and subcategories, and the rules for which content goes where. On a store, the product categories. On a manufacturer's site, the product lines, industries, and document types.
- Navigation labels. The words in the menu. "Solutions" and "Resources" mean something inside the company and very little to a buyer.
- Cross-links. How a product page points to its manual, its replacement parts, and the industry page that uses it.
- Search and filters. Which words search understands, and which choices the filters offer.
There is also an XML sitemap, a file for search engines listing every address on the site. Same word, different job.
Why does information architecture matter to my business?
Because every page a customer cannot find is a sale, a support call, or a lead you lost. The usual symptoms are a menu that grew one item at a time, categories named after departments, the same product described three different ways on three pages, and a search box that returns nothing for the words customers use. A new design on top of the same structure gives you a nicer-looking version of the same problem.
Does it matter for Google and AI assistants too?
Yes. Search engines and AI assistants read the structure the same way people do: the headings, the categories, the labels, and the links between pages. A clear structure tells them what each page is about and which page answers which question. A page buried four clicks deep under a label like "Resources" is hard for a buyer to find and hard for ChatGPT or Google to cite. The same work that helps people find things helps you get found.
What does a big information architecture project look like?
At Honeywell we consolidated about 300 websites into 13. The work ran in order: content studies to see what existed, transformation rubrics to decide what to keep and merge, and a large taxonomy project so the merged sites had one set of categories. For HGTV Home by Sherwin-Williams I tested the category structure with real users and cut it from twelve to seven, and user confidence went up 13%. That one is written up in the HGTV Home by Sherwin-Williams case study.
How do I fix my site's information architecture?
Start with what exists, then test with customers before anyone designs a page.
- Inventory the content. Every page, document, and tool, with who it serves.
- Card sort. Customers group the content and name the groups. See what card sorting is.
- Draft the structure and the labels in the customers' words.
- Tree test it. People find things in a text-only menu, so you see where they turn wrong.
- Redirect every old address to its new home on launch day.
What does it cost?
A one-hour second opinion on your current structure is $150. I walk your menu and categories with you and tell you what I would test first. A full information architecture project is quoted as one fixed price on UX research and leadership, and a new website built on it starts at $6,000. Anything over $3,000 can be paid by the month.
Answered by Jared Lukes, Minneapolis. Twenty-five years in brand, UX, and technology, currently holding fractional AI and technology seats at three companies. About Jared.