In the past couple of months, we’ve been making it easier to browse the site. I’m going to cover a bit of what we’ve done to improve things and tell you a little bit about why it's changed. Since early on in our beta, we’ve had a breadcrumb-like navigation indicating what “category” the content you are viewing is in. We created this to help people landing deep into GOV.UK from, say, Google, to understand the full scope of the site. But this wasn’t a breadcrumb navigation in the way it is widely understood on the web; it was there just as an indicator, rather than being meant as navigation.
Learn from observing users
Once we started user testing the navigation, we realised we were probably being too radical, pulling out ahead of users’ understanding a little too far. We needed to turn our “breadcrumb trail” into a more traditional design pattern - revealing the category and sub-category that a piece of content lives within. When either of those links are followed, a user lands on a page listing everything that also lives within that category or sub-category. This means they can effectively “zoom in” and “zoom out” of the site, helping to orientate themselves and find alternatives to the content they are already on if they land somewhere in GOV.UK that doesn’t answer their immediate need. We learnt about the dangers of not supporting a browse experience from the alpha prototype of GOV.UK, where number 3 of our Top 10 Problems via user feedback was “No browse navigation”. Once we tested our assumptions about search and the homepage against users it was clear we had to iterate towards something better with browse too. I think we have achieved that, although we will of course be reviewing how successful they are and making further improvements when we see they are needed.
The 'related' box
We’ve had the 'related links' box almost since day one. We find in user testing that participants are continually drawn to it, which is a good sign, as it is also there to assist their orientation and journeys around the site. But its greatest strength is also its weakness - it has to support several types of user journey:
- A user has landed on a piece of content, and from scanning the title and the content, they realise they are in the wrong place - the related box provides links to similar content, one of these is likely to be what they need
- A user has finished with a piece of content, but they have a further need to access more content in a similar subject area. For example, a tool to help them find out if they are entitled to the benefit they have just researched
- A user has finished with a piece of content, and they now understand that their need is more sophisticated than that particular content provided for. They now need to “dig deeper” into government information on their topic. For example, they are an industry professional who needs regulatory information
These could all be classed as onward journeys, and therefore it lives over to the right of the content, implying this moving on from where a user currently is. It's also why we've changed its name. 'Other relevant links' feels like a more accurate description of what users are seeing in the box (although if you've seen any good alternatives elsewhere on the web, let us know). We’ve found users see, understand, and use the links - and it appears to be genuinely helpful - but we are keen to improve it further in the future to support these varied user journeys. In fact we’ve tried continually to reinvent the box over the past few months but so far none of our alternative ideas have convinced us as much as the current solution to take them as far as user testing.
We’ve done a lot of testing with users with all sorts of requirements and backgrounds and they’ve informed a lot of iteration of features, but until we have data from millions of people using what we’ve built we won’t truly know what works and what needs improving. All that starts on 17 October. Everyone in the UK has a differing range of requirements from the government, a vast range of confidence and ability when using the web, and all access the web across an ever-growing range of devices. We have a duty to support all of this and so we will continue to iterate as and when we identify how we can do better. Onwards!