Your customers rely on your website structure to help them find the information they need. It’s a fundamental part of their experience on your website, so you need to make sure you get it right.
The term ‘information architecture’ was not coined by accident. It shows that your categories, menus and links are as important for your website as bricks and mortar are for your office.
When building a great site, there are a number of aspects of structure that you need to consider carefully.
Deep vs Shallow
A deep structure is one with a lot of layers of content. Categories and sub-categories only have a few items in each, and they are all sub-divided.
A shallow structure has a few broad layers of content. Categories have lots of items of content, an often there are no sub-categories at all.
Either kind of structure can work well on your website, as long as you do it properly. Start by looking at the top level headings you think you’ll use. How many are there in relation to the number of pages you have? This will help guide you in the right direction.
User research is absolutely vital to creating a structure that works well on your site. Once you’ve come up with your top categories, put them in front of your target audience and see what they think. If you give them a list of pages, do they know what belongs in each category? If they struggle this could mean either you’ve tried to put too many things into one category, or your labels aren’t clear. Try again with more top level categories (which should mean a shallower structure), and see if things improve. We’ll talk more about labelling later on.
Audiences vs Topics
The majority of websites have a structure created by grouping content together into different topics or themes. Apple’s website, for example, lists the different types of product they sell as their main menu.
If your website has a number of different audiences with really separate needs and roles, you may find it helpful to include items in your structure that target user groups. If these target groups can clearly identify themselves, then this works really well as a basis for a structure.
Clothes shops often take this approach, where they have ‘Women’ and ‘Men’ labels, alongside others.
This mixed approach is often used to great effect on university websites, because they have such a range of different audiences. Groups outside of the university – applicants, parents, researchers, media and business – are all interested in things like programmes, research and news, so these exist as topic headings. Their internal audiences – staff, existing students, and alumni – all have their own specific needs, and so menu items take these users to their own targeted landing pages.
Audience-based categories do have a disadvantage. Some content may be relevant to more than one of your chosen user groups, which can leave you with a dilemma about where to place it. Similarly, users may waste time looking in various places only to discover that the same information is available across each category. As always, our advice is to test your proposed structure with members of each audience, and see how they respond to the categories and identify with the intended label.
Any medium-to-large website will need to provide search facilities to its users. It can be tempting, therefore, to spend less time on menus and structure than you ought to, assuming that most users will navigate by search. Search engines as well can often bring users directly to a content page without using any of your navigation at all – so maybe that means we can do away with categories and menus anyway?
Unfortunately, it’s never that simple. Even when users do arrive at their first content page through search, it’s rarely the case that their journey on your website should end there. Your website structure will determine what comes up in ‘related items’ widgets, local menus, and breadcrumb links. The customer’s onward journey from that first content page is just as crucial as getting them there in the first place. Your structure is key to providing a really excellent onward journey.
Labelling and navigation
Menus are where your structure and navigation overlap. They are what your customer uses, most of the time, to learn about and navigate your site structure, so the words you put in them are of the utmost importance.
As we said before, one of the main reasons your menus can struggle in testing is your choice of words. Even if your content groupings are optimal, your users won’t find anything if they can’t tell what’s in each one from its label. Listen to your customers to find out which words they use themselves – in search terms, in person, on social media, or wherever you can connect.
User research and card sorting
As we’ve said all the way thorough this article, user research will help you to make good decisions about your website structure. But what does it actually look like?
Card sorting exercises are a really cheap and easy way to test your structure with your users. There are lots of online tools you can use, or you can use index cards or sticky notes for in-person testing.
Make up a set of cards with all your content titles on each one. In a different colour, make a second set with all your proposed category labels – you may need more than one of these if you want to try out a few different ideas. You’ll also want some blank cards or sticky notes and a pen for your tester, and you’ll need to make your own notes as well. You can conduct one or both of these exercises with every user. Make sure to do them in the order given, however, as the second exercise will strongly influence the outcome of the first, if you flip them around.
1. Take only the content cards and ask the user to arrange them in groups, then make up a label for each group on the blank cards. You can set an upper or lower limit on the number of groups, if you like. It’s a good idea not to allow any ‘don’t know’, ‘misc’ or ‘other’ categories.
2. Spread out one full set of category cards. Give the content cards to the user and ask them to sort them by your labels. Make sure you take note of where they struggle or where different users come up with conflicting sorts. This will give you an idea of which categories might need re-labelled, or if the content item’s own name is unclear.
The first test will help you to understand the words your users will use and understand in your menus, and may give you ideas for groups that you hadn’t thought of at first. The second will help you determine if you have come up with any groups that are confusing, or poorly labelled. It’s often worth doing just the first exercise with number of testers first, before coming back with your own updated structure to do the second.
If you’ve been working on your website recently, how did you get on in deciding your structure? Did you do any user testing before you finalised it? And if so, how helpful did you find it? As always, share your comments and stories below.