Why Information Architecture Matters

This is one of the most frequently asked questions by my clients so I thought I’d share a some insight.

In a nutshell…


Information architecture: the practice of deciding how to arrange the parts of something to be understandable.[1] This might conjour up images of libraries and the KonMari method. But this discipline goes far beyond superficial organisation. It is at the core of designing great solutions.

But why is it important?

Information architecture (IA) is the blueprint or framework that we build solutions on. It’s a “structure or map of information which allows others to find their personal paths to knowledge.”[2] Getting this structure right is vital for the success of a website or digital product as it helps users achieve their objectives in an easy and simple way.

Some background

The inventor of this discipline, Richard Saul Wurman, was refining his ideas in the 1970’s, long before the advent of UX design, the internet, or websites. He realised that the principles of architecture could be applied in a 2D sense, “the creating of systematic, structural and orderly principles to make something work.”[3] Using Wurman’s philosophy, IA now helps us organise information within complex websites.

How do designers implement IA to improve the overall experience of a product?

In order to get it right we must refer to the user to help us understand the pain points of the person using the product (such as a website or app), or things they need help with. Just as a good architect would not build a multi-story house for a wheelchair user with loads of steps  — they would incorporate lifts or slopes into the design or make it all on one level to enable the user to get around easily — the architecture of a website should factor in the needs of a user the help them accomplish what they set out to do.

Gary Woolf explains that “the presentation of information can be more important than the information itself”[4] because its all about connecting the user with the stuff they need. By doing things like card-sorting and user research, we can get a better understanding of the mental models or our users and structure our product accordingly, presenting the information in a timely, logical, and rational way.

Whats a mental model?

A mental model is the way we perceive and organise information around us. No two people are exactly the same, and their mental models will inevitably differ, a bit like a fingerprint - the way we understand and exist in the world is formed by our own set of experiences and opinions. There is no right or wrong answer, every mental model is valid in its own way.

As a designer my job is to ask questions and try to understand a number of different mental models so that I can see trends and interpret the commonality.

However. It’s really easy to pigeon hole people into oversimplified categories based on their mental models; the best designers are those who can understand nuances of the people they are creating for and make solid judgement calls on what is important and what isn’t.

Ok, sounds good. Show me how this works in practice then?

Here’s an example. I recently asked a few users to complete a task for an online grocery shopping product I am developing. I asked them to group a list of food items together in categories, and to name the categories. They had completely free rein to answer as they liked.

As expected, no two users gave the same answers. Of course there were similarities and overlaps but no two people gave EXACTLY the same response. This is good. This shows that grocery shopping is actually quite a complex system, and the way people interact with the system varies according to their need.

Categorisation is difficult because you are dealing with semantics, different contextual knowledge, cultural differences and opinions. In the above example, this card sort revealed that some people group things by meal type such as “Breakfast” or “Convenience food”. Others went for classic “grocery, bakery, butcher, fishmonger”. Others used more common labels such as “dairy, meat, vegetables”.

My job is to interpret these results and find the common ground that makes it instantly understandable to the target demographic for the product in question.

In this example, I visualised these as a Venn diagram (below):


The great thing about asking lots of questions is when you get answers you aren’t expecting. Breakfast is perhaps the most interesting category choice as it has the potential to include items from almost all of the other categories, and is something I would not have thought of by myself.

This gives me a basis for trying out a design that combines food groups in a different kind of way. Maybe it will work, maybe it will perceived as too prescriptive - breakfast is straigtforward, but lunch and dinner might not be so distinct depending on who you talk to.

For the users who struggle to remember to buy all the necessary ingredients for a meal, grouping items in terms of meal planning rather than aisles of a supermarket may save them time and energy. Plus it is a solution that cannot be offered so easily in the real world due to the physical limitations of a store and shelf space.

At some point I have to make a judgement call based on what I think it the right route. This all feeds into a site map like this one:

instashop site map.png

Based on the information architecture we can then proceed to design a solution thats hopefully rooted in a context that is easy for users to navigate.

Of course, we don’t know until we test it whether it works or not, but that’s the great thing about working in sprints - we can course-correct when we get things wrong.

Here’s a quick mockup of what the store could look like, based on the site map:


References and further reading:



‘Understanding Context: Environment, Language, and Information Architecture’ By Andrew Hinton

‘Graphic Information Architects’ by Richard Saul Wurman