Information architecture is perhaps a proto-content strategy skill set. In the earlier days of the web, information architecture was the skill set that dealt with all the content on the site from a user experience point of view. It answered questions like:
What kinds of information do our users need?
How are we going to structure that content to make it easier to find?
Which information should be on each page?
Information architecture was well defined in “The Polar Bear Book” or Information Architecture for the World Wide Web. Information architecture is:
“The structural design of shared information environments
The combination of organisation, labelling, search, and navigation systems within websites and intranets.
The art and science of shaping information products and experiences to support usability and findability.”
Today, information architecture is a skill set that content strategists and UX practitioners use to structure websites, intranets, extranets, e-commerce sites, and other information products. It considers questions like:
What content do we currently have?
What terms do users use and how do we reflect this in our website?
How will the pages all be related to each other?
What metadata and taxonomy are needed?
How will navigation and browse work?
How will search work?
How will the information be laid out on the page?
The practice can be applied to many areas in corporations and organisations. This article will mainly focus on information structures within websites and intranets for content-heavy websites.
Before we dive into deliverables and principles, a few other definitions are needed:
Site map: The hierarchical layout of the pages on the site. Sometimes a site map is referred to as “the information architecture of a site” but this phrasing is ambiguous. We use site map instead.
Content model: Defines the content types and their associated metadata and taxonomy.
Content types: The distinct layouts and requirements for different kinds of content on the site. Includes the on-page content sections as well as metadata. For example, a news article has different metadata, taxonomy, and on-page content than an event posting.
Metadata: Data about data, or information about each page of content on a site. Includes things like title, description, author, dates, and any taxonomy tags.
Taxonomy: The pre-defined, controlled list of terms (or tags) that can be used on content. These controlled lists are also referred to as controlled vocabularies and are a way to standardise content tags on the site so they can be used to suggest related content, display dynamic content, and allow for content personalisation and customisation.
Wireframes: The low fidelity design for the navigation, each content type, to show where the content elements appear on the page.
When designing websites and intranets, information architecture touches a number of deliverables in the discovery and design process. While there can be many variations of the deliverables, in general information architecture produces:
Metadata and taxonomy audits
Information architecture findings and recommendations
Information architecture can also lead or have input into:
Content strategy findings and recommendations
Information architecture work can happen in small and large organisations as well as agencies. The scope of deliverables depends on each project.
Principles of information architecture
In our definition above, information architecture looks at organisation, labelling, navigation, and search to support usability and findability. This organisation system supports both users and internal staff.
Structure is easy for users to understand Sites should be organised in an easy-to-understand way. The aim is to make a site structure easy to understand or easy to learn. The organisation on the site has been created according to user needs and has been tested with users. Labels reflect user language. The navigation behaves consistently, and the search and search results function according to user expectations.
Note that you might hear people say they want an “intuitive” site or “three clicks to anywhere.” Instead of taking these as design requirements, take them as indications that the site structure isn’t currently working well. Realistically, many sites can’t have “intuitive” structures that, according to definition, are used “based on what one feels to be true even without conscious reasoning; instinctive.”
Many content-heavy sites have content that’s technical, detailed, or simply something that needs to be learned. (Think about your electricity company support site, or a support site for your smartphone. That content can’t be intuitive.) Many content-heavy sites also have too much content to be “three clicks to anywhere.” These complaints can be mitigated by doing user research and user testing.
Content is findable for staff Whether a content team is centralised or distributed, content needs findable for staff. When staff can find content, they can reuse it or reference it. This reduces content re-creation effort, content bloat, and, ultimately, costs to maintain an information-rich site. Findability is supported with organisation and labelling.
Basics of organisation The terms “organisation” and “labelling” are very generic terms. Organisation and labelling are created with page structures in site maps and with metadata and taxonomy in content models. What these systems look like depend on your user needs, business goals, and content strategy.
Site maps Site maps lay out all the pages on the site and the parent-child hierarchy of these pages. A page on a website is an HTML page or PDF file that needs to live in one spot. It’s important to lay out this structure with the appropriate user-centred labels. But you need to know more about the pages than just their physical location.
Content models The content model with its metadata and taxonomy detail what you need to know about different pages. A content model outlines the content types, their relationship to each other, the metadata on each content type, and which metadata are driven by taxonomy.
Once you know more about each page, you can use this metadata to find it outside of the site map hierarchy, to dynamically display content, or links on different parts of the site.
Most information architecture projects follow a very similar pattern of research and discovery, strategy, and design. Information architecture can also heavily contribute to content migration efforts.
Research and discovery Whether working with other UX practitioners or not, this phase typically consists of discovering user needs and goals and business needs and goals. We uncover user needs and goals with user research techniques, such as user interviews and user testing. We uncover business needs and goals with stakeholder interviews and alignment workshops.
In this phase, we also do content audits for structure. We look at the page hierarchy of the site, the metadata available in the content management system, the taxonomies used. It also includes a review of the analytics and SEO keywords.
From this research, we create user personas, user journeys, and a findings report.
Strategy Information architecture strategy is often incorporated into a larger content strategy. However, it is possible to create an information architecture strategy that maps how the site map, metadata, and taxonomy solution will fix the problems discovered in the Discovery phase.
A strategy would have elements such as a vision, guiding principles, and strategic rationale. These are used in the design to keep focused on what the information architecture deliverables are trying to accomplish.
Design There are a number of deliverables in the design process and the items you do depend on the time and scope. Whenever you create a deliverable, it should be tested with users to ensure it is meeting user needs and adhering to your information architecture strategy or content strategy.
To create a site map, the process includes creating a draft site map, reviewing it with project stakeholders, revising it, testing it, revising it, then doing a final review with stakeholders. Testing site maps can be done with card sorting or task testing.
Creating a content model involves similar steps: create a draft, review it, revise it, finalise it. Content models can be tested once the content types are made into wireframes. When reviewing the content model, this should be with the project team and web developers, ensuring they understand how to build out the content model in the CMS.
Wireframes manifest the site map as navigation and the content model as page layouts. The process for creating wireframes is very similar: identify the wireframes to create, create them, review them, revise them, test them, revise them, finalise them. Any changes to the wireframes may result in revising the content model and possibly the site map.
When you create a taxonomy, you’ll follow a process of collecting all the taxonomies available, identifying missing taxonomies, then creating a draft of the taxonomy, test it with card sorting or task testing, revising it, reviewing it with subject matter experts and the project team, then finalising it.
Who is involved?
Information architecture is a skill set that interacts with a number of other specialities.
Content strategists Provides the structure in which other content strategy techniques can build upon.
Researchers Feed questions into research, and pull information from the research results to influence the IA deliverables.
Internal stakeholders Ensure business goals are identified and met in the information architecture deliverables.
Designers Either designing wireframes, prototypes, or comps, uses the IA to inform the design.
Developers Developing the backend of the CMS with input on structure.
SEO strategists Ensures SEO needs are represented in the site map and taxonomy.
Content (re)writers Uses the site map and content model to migrate content to the new site.
Examples of great information architecture
There are many many examples of websites with good structure, metadata, and taxonomy. Here are a few examples that are publicly available from content-heavy sites.
Goodreads uses taxonomy and metadata to allow users to find books of interest. One of their taxonomies is Genres. A piece of metadata is “new release.” Combine this together with a specific genre, and you can see what’s new in Historical fiction.
On the homepage of GOV.UK you can see all the topics for which they have content. If you pick a topic, you can browse deeper into the hierarchy. They have both wayfinding pages on their site and content pages. They get people to browse for their Topic via the Topics taxonomy, make a choice, then show content. Users can also search and narrow search results via taxonomy. Many government websites are good examples of content-heavy sites.
The Usability.gov site uses two levels of navigation, then displays content tagged with different taxonomy terms. Go into How To & Tools to find all the different articles they have about Methods, Resources & Templates, and Guidance.
Many Support websites are content-heavy. Looking at Support.Apple.Com, the home page has a search box as well as their product categories. Dive into a product category to narrow down to different topics for that product. Search for term such as “password” and use the filters to narrow down your results.
Here are some pretty useful information architecture tools:
Microsoft Excel A ubiquitous format, Excel is great for content audits, site maps, content models, and taxonomies.
OptimalWorkshop An online tool used for card sorting and task testing, among other things.
Key Pointe’s founder and principal consultant Theresa Putkey applies content strategy, IA, and taxonomy to solve problems within a content management or digital asset management environment. With more than 15 years experience, Theresa brings a valuable external perspective to each consulting project. As a content strategist, she applies her skills and passion for content and structure to produce great user experiences that meet business needs. She focuses on bridging the communication chasm between business leaders and developers and acts as a liaison between stakeholder groups with divergent interests and drivers.