Gerry McGovern recently wrote a post talking about when content audits are not a good idea. As as someone who has overseen the rationalisation of a site from 5000 to 500 pages I have to say starting from zero and working up really is the only practical way to do it.
Working through your site to justify why each page should be removed is painful at best, and if you also need to get approval from departments and management for each action then you’re going to get nowhere fast. Instead of starting from a point of the current site and removing, you need to start with nothing and justify the creation of each page.
Justifying creation gives you a much more streamlined process, because the evidence has to exist before you start, and while it might seem like it is more involved, it is quicker too. Starting with 5000 pages and trying to get to 500 by removal means you need to make 4500 decisions along the way. Some will be easy, but some will be very hard. If you start from the bottom, you only need make 500 decisions to create 500 pages, and yes some of those decisions will be hard, but many will also be easy.
What to create ?
It might seem that working out what to create is a harder task - but if you build yourself a body of evidence you can make it much easier.
First, start with user needs - work out what your users want. This is where the Top Tasks methodology can help a lot. There is more than one way to get these results, but it’s hard to beat asking people why they are using your site.
If you work for a service based organisation, you should be able to get hold of a list of services that your organisation provides (if you can’t, then welcome to the public sector - while your organisation may not know, there are service lists that can tell you what they are suppose to be). The list of user tasks and organisational services will give you a solid base on which to work out what you need to create - but it will also throw up a couple of issues.
Users want it but we don’t do it
This is surprisingly common, especially in the public sector. Many people struggle to know which public body provides which service, and when the district council is responsible for collecting your rubbish but the county council is responsible for getting rid of it, you can’t really blame them.
The important thing to realise is that if there are a large number of people arriving at your site looking for something then they think it’s your responsibility to provide that service. Denying the existence of the service isn’t going to solve anyone’s problem, and it will just cause the user to contact your organisation in another (probably more expensive) way.
You need to acknowledge that this happens, and try to help the user as well as possible on their way to their answer. This is more than redirecting them to the homepage of another website - try to deep link, and if this is a high volume query for your site, try to get access to the data so you can answer the questions yourself.
You do it but the users don’t want it
Unless you work in the perfect organisation, you will find items on your list of services that do not feature at all in the list of tasks that users come to your site for. These are the ‘Bottom Tasks’ - and again, unless you work in the perfect organisation, these are not going away.
Work to minimise the impact they have on the majority of users. Develop a demand-driven Information Architecture, and if it makes sense, consider segmenting the content.
When it comes to the creation of the content for these ‘Bottom Tasks’, the real decisions you need to make are about the amount of effort you are prepared to invest. You only have a finite amount of time and resources to spend on your content, and you want that time to be used in the best way. Use the user needs as your guide - look to spend more time on the high volume content, and less on the low. I know it sounds overly simplistic, but if you adopt an agile approach to content creation, and create the content that you must have first, this way of prioritising will evolve as a natural part of your project’s development.