Information architecture: the blueprint of the digital world

By Sabrina Singh, Service Designer

29 April 2024

— 9 minute read

In the vast realm of our digital world, we can liken today's digital products such as apps and websites to the towering skyscrapers we see in our cities. They are built with purpose and designed with structure and navigation in mind. Much like traditional architecture with physical buildings, information architecture is the craft behind constructing digital products.

 

 

Laying the foundations: what is information architecture?

Information architecture (IA) can be described as the foundation of digital products. It refers to the structuring, organisation and labelling of information, with the aim of providing it to users in the most effective and sustainable way possible.

IA connects three important elements:

  • the user (who is looking for information)
  • the content (the information itself)
  • the context (the place where the information lives).

The interdependent nature of these elements is referred to as the information ecology.

Figure 1 - information ecology

The main goal of IA is to help users find information and complete tasks quickly and easily. Humans are cognitive misers, where we tend to take mental shortcuts to find a solution to our problems. In the world of digital products, this means that there is a risk that our users will abandon our product when information becomes too complicated or too slow. When this happens, it becomes more difficult to bring them back.

This is where IA plays a key role. Content should not only be valuable for users, but it should also be easy to find. Well organised and well-structured content will help users better interact with a product, leading to a better user experience and increased retention rates.

 

The building blocks: information architecture systems

Just as all buildings are built with foundations, walls, floors, beams and a roof, there are core components that should be considered when building out an IA. There are four key systems:

Organisation systems:

These are the categories and subcategories used to sort and structure information. Let’s use a shopping centre as an example and liken the categories to levels. Fast food stores are typically grouped together in a food court, whereas luxury stores are typically grouped together on a separate level.

Labelling systems:

These are the actual names given to each category and subcategory. Drawing back on the shopping centre example, think about the signage for each store, bathrooms, lifts and exits.

Navigation systems:

These are the ways users browse or move through a digital product such as menus, links, and buttons. In a shopping centre you can find these in the form of elevators, escalators and even car parking ramps.

Search systems:

These are the tools users will use to look for information on your digital product, such as a search bar. Think about information desks or interactive maps in a shopping centre.

Figure 2 - IA systems and their related activities

Constructing an effective information architecture

Building effective IA is not a one-size-fits-all task . Just as the structural requirements change between constructing a high-rise office and a beachside villa, it will be shaped by the context and content of your product. It’s a process that is tailored to the needs and behaviours of your users. While you may find that your approach will vary, here are some practical steps you can take when building it out.

1. Understand your users

A common pitfall when building digital products is content overload. This happens when you have too much content on your site, making it hard for users to find the information they need. To avoid this, you can conduct research with your users through methods such as interviews and usability testing. This will help you identify your users’ needs and behaviours, which in turn will help inform IA decisions like choosing what information is most relevant for your users.

2. Define and refine your content

To identify what content you have and where it currently lives, you can develop a content inventory. This is a comprehensive catalogue of information about your content, such as title, format, metadata, dates and more (this is usually done in a list or spreadsheet). Your content inventory should be a source of truth for your site contents and should be governed through content audits.

Content audits involve reviewing your content inventory to assess how useful, accurate and effective your content is. You should mark any content that is out of date, irrelevant to users or any new content that will be added in the future. Content audits help us make informed decisions about the organisation and optimisation of our content.

3. Categorise your content

To categorise the content of your site or product, you can use a UX design method called card sorting. There are two key approaches to conduct card sorting; open and closed.

Open card sorting

In this method, participants are given a set of cards (each card containing a different piece of content derived from the content inventory) and are asked to organise them into categories that make sense to them. This helps us understand user’s mental models, which can then be used to create the first iteration of our taxonomy and metadata.

Taxonomy in IA refers to how information is grouped, classified, and labelled. Metadata is the data that describes and identifies other data. Together, taxonomy and metadata play a key role in making the information on your site more discoverable and accessible. Let’s use an online clothing store for example. Clothing may be categorised by type, fabric, colour, and occasion. Therefore, when building the site, you can tag content with the appropriate metadata e.g. blue denim jeans, so that users can easily filter or search what they are looking for.

This method is great when you’re building your IA/product from scratch.

Figure 3 – Open card sorting (from the Interaction Design Foundation, © CC BY-SA 4.0)

Closed card sorting

This method is like open card sorting, except participants are given a set of cards (each card containing a different piece of content) and a set of predefined categories. Participants are asked to organise those topics into the given categories. These categories are usually developed based on existing knowledge or assumptions.

Tip: You can look to see if there is a standard taxonomy for your industry/domain to get a head start.

This method is great when you want to validate an already built IA or want to add a new piece of content to your product.

Figure 4 – Closed card sorting (from the Interaction Design Foundation, © CC BY-SA 4.0)

4. Create a sitemap

A common challenge we see when building digital products is the inclusion of various external links or resources to information, which leads to an inconsistent user experience. Creating a sitemap will help you plan how content will be interconnected and navigable so that you can build the most seamless experience for your users. A sitemap usually displays the pages of your site or product as boxes and arrows or lines are used to demonstrate the hierarchy of content and the relationship between pages.

5. Consider the searchability of your information

Now you have confirmed the information and pages, you need to consider how users will search for and find your information. Will there be a search bar? What will the navigation look like?

Another common pitfall is thinking a search bar will automatically solve for the searchability of your information. While this is a great first step, there is still more that can be done. Your content should be aligned to the taxonomy and tagged with the metadata you identify when speaking to your users. This is so they can search based on the attributes they associate with a particular piece of content. Search results can also be filtered to only show the most relevant content for users.

6. Build, test and iterate

We should be testing our IA with our users and iterating it based on their feedback. This can be done through many different methods, the most common being tree testing or prototype click testing.

As a “lo-fi” method of testing our IA, we can use tree testing. It involves testing your categorised content, or a ‘tree’, without the influence of visual design or navigational aids. 

Figure 5 - Tree testing process

Click testing is conducted by evaluating how users interact with a prototype or product. It is a more holistic method of testing your IA as it provides more context to the user. Click testing involves identifying where a user clicks when trying to complete a task or find information, which provides insights into how intuitive and navigable a design is.

 

Principles of information architecture

There are 8 key principles to keep in mind which make information more findable, a digital product more navigable, and enhance the user experience.
 

Swipe for more
  Principle Application
Object principle


Content should be treated as a living, breathing objects with a lifecycle, behaviours, and attributes.

 
  • Keep your content inventory up to date to track your content lifecycle
  • Conduct card sorting and apply the taxonomy/metadata you discover to your content – this helps users find content based on the attributes they associate with it
Choice principle



The more choices there are, the harder it is to make a decision.
 
  • Strike a balance in the number of choices presented to a user – too many may increase cognitive load but too few may limit exploration
  • Conduct card sorting or IA testing to determine a focused set of choices that are meaningful to users
Disclosure principle


Just enough information should be shown to help users understand what kind of information they will find as they dig deeper.
  • Conduct IA testing to identify whether the labelling of your content is sufficient in helping users determine what they will find next
Exemplar principle

Showing examples of content can be more effective than describing them.
 
  • Conduct IA testing to identify any difficulties in interpreting content
  • Iterate content that is difficult to understand with an example or diagram
  • Test different variations of content with your users to find what resonates with them the most
Front door principle


Users can enter a site or an app from any page and not just the home page.

 
  • Design every page as an entry point– users should know where they are and what else they can do
  • Ensure your navigation bar is visible on every page
  • Consider using breadcrumbs to keep your user informed
  • Ensure important links or resources are available on each page
Multiple classification principle

Users like to browse content in various ways, not being limited to one predetermined path.
 
  • Organise information in various ways, to accommodate to different user’s mental models
  • Conduct and analyse your card sorting results and only use the most common categorisations
Focused navigation principle


Navigation should help users understand the content offered, not distract them.


  • Navigation bars/menus should be focused and free of extraneous links
  • Any names on your navigation bar/menu should indicate its purpose rather than its location
Growth principle

The design should anticipate and accommodate for the natural growth of the site’s content and user base over time.  
  • Speak with content owners to understand what content may be added in the future
  • Label and design your product in a way that anticipates the addition of new content
  • Conduct card sorting or IA testing if you’re looking to add a new piece of content to your site 

 

Build and maintain

In our digital world, information architecture is the blueprint for success. A well-designed IA will allow our users to find their information and complete tasks as easily as possible. By not determining an IA, there is a greater risk of our users abandoning our product for our competitors, no matter how aesthetic.

As with most design work, information architecture doesn’t end with doing a few rounds of testing and iteration before leaving it to set and forget. As your product evolves, so should your IA. Regular testing and refinement will ensure your digital structure continues to meet your users’ needs and expectations, thereby building a successful product.