Deliver me the content I need and nothing else or Information Architecture (IA)

Author: 
Jacinta - Web Guide Team
Category: 
The Department of Finance Archive

The content on this page and other Finance archive pages is provided to assist research and may contain references to activities or policies that have no current application. See the full archive disclaimer.

 

One of the issues raised at the focus group we held a while back was that our users want information which is relevant to them. They want to find and make use of it quickly, without barriers. Our stats show us users tend to come with a purpose and want to fulfil it quickly:

“This guide is aiming at a whole bunch of people and one way you could address that is to say ok from a management point of view these are the important chunks, if you’re designing, these are the things that you should be looking at more seriously and if you’re the bunch of people who might be writing these pages then these are the sorts of things you should be across..” – Focus group response

In thinking about this a bit further, we quickly realised that the ability to slice and dice the content on the site in a number of different ways would be very useful. It would be useful to be able to answer questions like:

  • As a project manager, what are the mandatory issues I need to address when building a website?
  • What are the policy issues surrounding accessibility? What are the technical issues?
  • How do I (as a front end developer) implement videos on my site?

....and so on.

The current site’s navigation doesn’t allow our audience to easily and quickly answer these questions. You can be left to process several articles and figure out which bits are relevant to you.

That’s why we’re implementing a faceted classification system. This should give us the ability to assign a bunch of classifications to the many pieces of content and then use those classifications as the basis for a faceted navigation system.

To plain English that – we’re dividing up the content based on a number of different category groups. The category groups we currently think would be the most valuable to you are:

  • Compliance level
  • Role
  • Topic area
  • Website lifecycle

(We’re going to take this to user testing soon, but your thoughts now on whether these category groups are the ones you need would be good).

A piece of content would get classified (possibly with more than one classification in a group).One piece of content might be classified in the following way:

  • Compliance level: Mandatory
  • Role: Web Managers, Project Managers
  • Topic area: Domain Names
  • Website lifecycle: Planning, Building

We then are going to build a set of navigation paths through which you can find that piece of content. Each navigation path has the ability to filter by the other classification groups. Say you are a Web Manager interested in applying for a new domain name, but don’t know how. You click on Domain Names and up comes a list of all the content on domain names. Instead of selecting something at this point, you could filter the results by choosing the relevant Role (in this case, ‘Web Manager’). The end result would be a more manageable and relevant subset of content to review.

Now, that process gives you the ability to navigate to a piece of content, which is fine as far as it goes. BUT, we still needed a way to do what was asked in the focus groups:

“Overwhelmingly, participants were emphatic in their feedback regarding the need for clear, practical examples within the Guide. Suggestions included scenarios, a pattern library, and checklists.” – Focus group report

Combining this with the faceted classification system, it made sense to us to have one further classification group:

  • Document type

Which means we can add in classifications of “overview”, “business considerations”, “how to” and “template”. As the site grows over time we can add further classifications to the groups, meaning we have scalability as we need it.

The difficulty with that is we might start to overwhelm the navigation pages with content pieces. A comparatively narrow topic of setting up a domain name might well blow out into four different pages as we cover general information and the business and the technical and provide a template for it. Then stability issues also arise. What happens if we choose to implement more content types in the future?

The other concern is that sometimes we write our own content, but if there is existing SME content available, we link to the source. For example, the National Archives of Australia have some very nice guidelines on archiving web content.

So we link to their content. But a search on our site would show up our checklists, but not the Archives (as the later is only a link on a page, rather than a page of its own). Which means searching for checklists is going to return less useful results. So we are making a business call – two actually. The first is that only the overview is the compulsory content type for each topic. The second is that only overviews will turn up in the navigation and search results. The overview page will then contain links to the other content types. This makes the document type classification run slightly differently from the other four.

This gets us around the problems of too many content items and when the content item you need is linked from our site to another. But it does add an extra click to get to some types of information. Everything has swings and roundabouts –user input through testing and this blog will let us know if we made the right call...

Comments (5)

I like the idea but do wonder a bit how the 'role' concept will help people like me who are one-man shows. I'm the manager, coder, information architect, designer, user experience officer and probably several other people too. Does this mean I will need to read several articles on the same topic to make sure I've got all the info?

What HCI research has gone into the IA plan/Architecture adopted by gov, also having a blog can complicate IA as there becomes no hierarhial links and it becomes a data driven keyword search architecture, curious your thoughts and plans to address convulted architectures that exisit for blog tools?

The question of a blog IA is an interesting one. When the Web Guide was a stand alone blog, we used a combination of categories and subcategories to provide a hierarchical IA. We also used author generated tag words as an alternate navigation structure.
When that blog was enfolded into this one however, the hierarchal structure would simply no longer work. We talk about project management, so we had included it in our hierarchy. If another project based blog also wanted to discuss the topic, including it in their theme made no sense as a user would more likely be interested in all project management related material.
As you can see, we have put in place a very flat structure, with each blog theme categorised. Our main navigation option now comes from author tags attached to each post, on the assumption that a user is interested in a topic. The downside with this method is of course, that it is left up to authors to attach tags, with each author having varying interpretations as to what goes with each tag.
For the moment, we don’t see it as a major issue as we have a small author pool and have manual processes in place for them to use a controlled vocabulary of tags. When our author pool grows, we may need to revisit.

Enterprise architecture in the APS - CoPs

Could anyone give me some guidance on how to find out if there is an APS CoP or discussion forum for government enterprise architects?

Thanks

Paulette

Comments on this post are now closed. Please let us know if you would like to discuss this post.

Last updated: 27 July 2016