Since GOV.UK launched we've been grappling with something: how should we treat content that doesn't affect many people but could affect anyone?
It wasn’t a huge deal at first, since we generally handed off to third-party sites for details. We knew we had to work it out at some point, but there were always more urgent things to take up our time.
In the next few months, though, a lot of those third-party sites will cease to exist. And their content will be coming onto GOV.UK. So we’ll be linking to us.
The challenge
For any 1 rule there might be 10 exceptions that apply in certain circumstances. If we have 1 page for the content that affects everyone and 10 pages of content that only affects a small number of people we end up with 11 pages of content about a very similar subject area.
How does that look in search results?
Say, for example, you're an employer. You have 5 members of staff and you find out that 1 of them is changing gender.
What do you need to tell HM Revenue & Customs? How does it affect payroll? Or their pension? And when? As an employer, you need to know this stuff.
You're not a specialist (a tax advisor or whatever), you definitely need to know the rules and GOV.UK is the obvious place for you to find those rules.
From the perspective of the user, it should be easy. You search on Google, you find the right page, you visit it and find out what you need to know.
Then you leave the site and have a nice life. Job done.
But from the perspective of the people trying to work out the best way to allow you to have this experience, this poses a significant challenge.
If we don't do it right, we've created a jungle. The general user has to machete their way through all the stuff that doesn't apply to them to find what they need to know. This is precisely the experience we don't want for users of GOV.UK.
Simpler, clearer, faster, right?
The solution
We have around 300 arm’s length body (ALB) websites transitioning onto GOV.UK this year. And the HM Revenue & Customs (HMRC) site.
That's a lot of content. A healthy portion of it will be 'niche'.
All this content, if it applies to a general audience, will sit on the Services and information section of the site. We're currently working on transitioning UK Visas and Immigration (formerly UKBA) content onto GOV.UK. We're setting up a multi-disciplinary team to look at the low-demand items in this content, see how they do, and work out how we can do it better.
In order to ensure the best possible solution we'll be looking at:
- user testing
- analytics - so we can see how the content is performing and how users are engaging with it
- design changes
- format changes
- new approaches to naming conventions
- new approaches to site architecture (we currently have a flat site, more or less)
We’ll go with whatever works best for the user. Anything is possible and if you have ideas for how we could best treat this kind of content, please leave a comment.
We'll be sharing results of our research and updating you on the progress in future posts.
6 comments
Comment by alex posted on
A book shop that only sells best-sellers wouldn't last very long. What makes you think a government web site that only serves the top tasks would be any useful? Ah, right, a government site has no competitors that provide better service.
Comment by Padma Gillen posted on
Hi Alex. Thanks for your comment.
I'm not sure why you think we only deal with 'top tasks', since this post explains how we're dealing with the 'long tail' too.
We see no point in duplicating content if the user need is met perfectly well on another site.
As a government site we most definitely aren't trying to compete for market share. That's not because we're lazy. We're trying to meet user needs not sell books (or even government policies). If handing off to another site is the best way to do that, that's what we do. If not, and it's within the scope of the GOV.UK proposition, we create the content.
Comment by Julián Rodriguez Orihuela posted on
We're gonna be facing this issue here in Buenos Aires too (I'm head of usability and creativity at the e-gov department)... At the moment we have all content related to "stuff the user needs to do with government" inside our "Tramites" section ("tramites" can be translated as "services" or something like that)... but then we also have sections for all ministries, so we put additional information in there...
But... what prevents us from creating one guide for each service (even if it is "long tail)?
The example on gender change is challenging, because maybe there should be a micro-site (or whatever, a content grouping of some sort) for employers with all possible situations that can arise (including an employee with a gender change)...
And like that we could make different micro-sites for (e.g.) people with children, people with automobiles, people with a disability, people starting a business...
Anyway... we still don't have a solution... eventually we'll probably steal from you... or maybe try something of our own and you can steal from us!
Cheers!
Comment by Padma Gillen posted on
Hi Julian. Thanks for your comment. Concerning your 'microsites' idea - we've found that it's important to look at the data before grouping content in any particular way. This seems to be especially true when trying to organise based on audience. Do people identify themselves as 'people with children' when they're searching for content? If not, they won't find it so easily if you decide to organise it that way.
Comment by André posted on
Hi, any update on this challenge with low demand content?
Comment by Ben Andrews posted on
Hi André. Thanks for following up question. If you are following the site closely you may see subtle tweaks and changes as we move towards what we hope is a better experience regarding this sort of content. Recent examples include improvements to the search experience or consolidating how we present metadata on a page. We will soon be introducing a new version of our 'mainstream' browse. This is in build and being tested at the moment and we hope it will do a much better job at delivering users to content, regardless of its relative popularity. We'll be sure to share the lessons on this blog.
Thanks, Ben