What Is Information Architecture?

Nielsen Norman Group distinguishes two components. The first is identifying and defining the site's content and functionality. The second is the organisation, structure and naming that define how they relate to each other. According to NN/g, information architecture does not appear on screen but informs the interface, and it is documented in spreadsheets and diagrams such as a site map.

Navigation is something else: the interface elements that lead to that information. NN/g lists global navigation, local navigation, breadcrumbs, filters, related links and footers. Both are part of the user experience, which has its own entry in this glossary.

The work starts with knowing what content exists and how users group it. For the second part, NN/g describes card sorting: participants sort cards with page names into groups that make sense to them. In an open card sort they create the groups themselves. NN/g calls this the most common type and recommends it, because it shows users' mental models.

Once a hierarchy has been proposed, it is checked with a tree test. Participants get the category tree and look for where a specific piece of content or feature would be. NN/g uses it to measure whether key resources can be found and recommends it over closed card sorting for evaluating categories.

On the website, the structure becomes URLs, menus and internal links. Google advises organising a site logically, because it helps search engines and users understand how each page relates to the rest, and using descriptive URLs. If a site has more than a few thousand URLs, grouping topically similar pages in directories can help Google learn how often the URLs in each directory change. Google also says every important page should be linked from at least one other page on the site, and it can only reliably crawl HTML links with an href attribute.

A structure that does not match how users think hides content that exists. NN/g puts it this way: "if you're not checking, you're guessing". The architecture has to be tested to know whether users find the key resources.

For search engines, the structure decides which pages are discovered through links and how their relationship is understood. Google limits the claim: there is no need to reorganise a site right away, because search engines will likely understand the pages as they are, however the site is organised. The benefits are long term and matter most on larger sites.

Internal linking and heading structure put the architecture into practice, between pages and within each page. Both have their own entries.

• Take an inventory of the content before designing the menu.
• Run an open card sort to learn how users group content.
• Test the hierarchy with a tree test before building it.
• Name categories in users' words rather than internal jargon.
• Link every important page from at least one other page on the site, as Google asks.

• Confusing information architecture with the menu.
• Assuming a structure works without testing it with users.
• Creating overlapping categories, so that a piece of content fits in two places.
• Leaving pages with no internal link pointing to them.
• Changing every URL at once in the hope of a boost in Google, when Google says search engines likely understand the site already.