Pat’s Point of View

Hot off the press is my latest article The Many Faces of Information Architecture. Here’s a taste:

A lot of confusion and misunderstanding surrounds the term ‘information architecture’. The multitude of activities that can be labelled with these two words span a vast variety of people, skills and situations. [...] A precise definition of each term would be difficult, as there is still much debate even among the professionals who practice in this field, and that is not the objective here.

What is needed is a primer to help ordinary humans weave their way through the various labels that might be used to describe work related to information architecture.

To engage consultants or contractors, it is useful to understand the variety of terminology and be able to identify the area (or areas) in which you need assistance.

I wrote this article because I found myself explaining to my clients, the different flavours of work that come under the banner of IA. Most people have either a highly specialised view—whereby I couldn’t do XYZ because my business card said ABC—or they took a very wide view that ignored any specialisation, assuming that an IA-ish person could do everything.

It was a fine line to tread, because on one hand I didn’t want to split hairs, for example if I design the way in which users interact with a web form it might be best described as Interaction Design but is it really a crime to call it Information Architecture instead? On the other hand, people do get into all sorts of trouble by using one term when they really mean something else. Which is why the article includes a bit of a cop out:

While this paper attempts to explain the different labels being used for areas of speciality in information architecture, the key to success will always involve explaining exactly what it is you expect to get from a potential information architecture professional.

It really is the best solution; spell out exactly what you want done (or what you do if you are a practitioner) so there is no misunderstanding. By all means you can use a label such as Information Architecture but make sure you back this up with a plain English description.

The other thing I wanted to explore was the overlapping nature of these terms and the activity they represent. Even if you could distinctly define Information Architecture, Information Design and User Experience Design, there would definitely be overlap. My attempt to illustrate this is the lovely circular orange diagram you see in the article :)

I am expecting this will be a fairly controversial, which really serves to highlight the ambiguity surrounding these terms, and the difficulty people have when they try to use these terms. But that’s OK. If you disagree with anything I’ve said in the article, please write a comment below. Hopefully this discussion will help educate non-practitioners, if nothing else.

On more than a few occasions recently, I’ve been quite surprised by the level of knowledge—and apathy—surrounding tracking website usage.

Then I read Karl Groves’ recent article The Limitations of Server Log Files for Usability Analysis on Boxes and Arrows. Here’s a snippet:

Server log files are inappropriate for gathering usability data. They are meant to provide server administrators with data about the behavior of the server, not the behavior of the user. The log file is a flat file containing technical information about requests for files on the server. Log file analysis tools merely assemble them in a conjecture-based format aimed at providing insight into user behavior.

When speaking with the owners and managers of websites and intranets, I’m often told that they have ‘tracking’ on their website, but probing deeper I find they are referring to simple web statistics (read “hits”). Looking at the kind of information they are getting from their reports, I am usually disappointed. This old style web stats is just not sufficient these days, and I thought we had got past this.

I’m no expert in web analytics (might I suggest Hurol Inan for someone who is) but I don’t need to be, since most people I’m working with are still struggling with the basics. Like getting any sort of reporting at all. If you don’t know how people are using your website, you’re flying blind.

Don’t get me wrong, no matter how good your web analytics are, it’s still at most, only half the picture. To understand why your audience is doing what they’re doing you’ll need to brush off those ‘soft skills’ and meet your users; user testing, observation and other ethnographic techniques can give you all the data you’d ever need in this regard. But, as with all things, you need a balance. Especially since they can be so easy to gather, web analytics are a good nominee for ‘quick win’ status. It should be a “no brainer” to know what users are doing on your site, even if you don’t yet know why.

The main culprit for this situation, beyond ignorance of anything better, seems to be IT administrators refusing to play ball. I’m never that surprised to hear that no better information about site usage is possible, because some guy in IT said so. However, there is a simple solution to this: build a business case for better analytics and use business to trump technology! It’s not that hard, look for ways in which good web analytics can empower decisions regarding the website. No other part of the organisation (I hope) is operating without some meaningful form of measurement, so if your website is a key channel for your organisation—be it sales, marketing or another objectives—you can’t keep flying blind.

I have started to become quite an advocate for Google Analytics. Not because I think it’s the best thing out there, but because it works well and the value for money on offer is phenomenal. Particularly if you’re generating uninformative ‘reports’ using Analog or something similar, that honestly tell you nothing useful. So I’m amazed that some people haven’t heard of Google Analytics. Like I said, I’m not expert in this field, but if I was shopping around for a web analytics package, I might not choose Google for this purpose (there are good reasons why one might not) but it would surely be my benchmark.

I particularly like the ability to establish goals and track conversion rates along paths that feed that goal. So if you have a website which sells products, you setup the sale of each product as a goal, then identify the paths people might take to get to the point where they actually buy the product. The reports will then tell you not only how many people achieve that goal, but where they fall off the path along the way. This would be very handy for evaluating potential usability problems with shopping cart processes etc. This is the most obvious use; the more complex your website is the more goals and paths you could monitor.

Hopefully, this will help some of you gain better intelligence, in that CIA kinda way, regarding what is happening with your website (or intranet for that matter). And this is just the tip of the iceberg, there are many more sources of data which you can use to find out what’s going on, and why. Web analytics just happen to be a good place to start.

I’m very happy indeed to be speaking at OzCHI this year. My short paper on mentoring and collaboration in information architecture was accepted after the blind peer review process, to my surprise. It’s the same topic I presented at OZ-IA, but thanks for the feedback I received from that experience I hope to improve it substantially.

All this week I’ve been working hard to become nationally accredited in workplace training and assessment, which will no doubt also allow me to improve my presentation. Although I’ll only have about 15 minutes, so I don’t think there will be any audience activities :)

So I get another trip to Adelaide, which I visited a few times for work last year. I’m looking forward to it, although I’m not keen on being away from home (felt my baby kick for the first time yesterday!). However, it’s a lot better than if I had to go overseas.

See what people are blogging about: OzCHI 2007, mentoring, collaborative IA

After an embarrassingly long time, I have finally finished a case study of an intranet redesign project I did for Canon more than a year ago.

This project highlighted that intranets do not need to be structured in the ‘traditional’ way (ie like public websites).

Instead of a single home page and a rigid view of the site, a fresh approach was taken, and the information architecture for [this intranet] makes use of personalisation to efficiently meet the needs of Canon staff.

Lastly, it is worth noting that the information architecture techniques devised for websites apply equally to intranets, sometimes more so. This is a key lesson for experienced information architecture practitioners.

Continuing my ‘exclusive tips’ for those who read my blog, what you won’t read much about in the article is the intranet alignment workshop we ran.

In any organisation, there are many stakeholders who are involved in the design or management of the intranet. The first challenge confronting many intranet projects therefore becomes to create a common vision for the intranet, to align the many stakeholders, and to define the role that each stakeholder will play.

Ask anyone who has worked in a large organisation on a website or intranet project, and they will tell you that getting management buy-in can be very difficult. Political battles over ownership of the site, and thus over its redesign, are commonplace. And without some early alignment and consensus, trying to get approval for design ideas later on can be heartbreaking.

Just a half-day workshop, held after some initial needs analysis has been performed, can work wonders. Key stakeholders get their chance to give their input, they can see progress is being made, and as a group can discuss those things that are often assumed (like what is the intranet for?).

Another day, another eight sessions! There were some good discussions today, both in and out of the sessions.

I received a heap of feedback on my presentation, all positive. The use of collaboration in web design and information architecture struck a chord with the audience and it was universally acknowledged as best practice. Almost as equally the topic of mentoring was intriguing to most people I talked to.

Some of the comments thrown my way, and discussions which ensued, included:

“Very open and honest”

“How much of the mentor’s time is taken up?”

“Just how much longer did the project take?”

“I liked that it was specific, not just theory”

“Delays between sessions could be beneficial, gets them to get on with it”

“I realised I’ve been doing that for ages but I didn’t know it!”

“What is the long-term future of UX teams within government agencies?”

“How sustainable is it? Have the mentorees become mentors and champions of UCD?”

“How do you get clients to go along with it [mentoring]?”

“How comfortable was the team with collaboration?”

“It’s unusual that the organisation would want to build up the skills of the team”

“Yeah but the problem is that once you train them up, they will leave and get an IA job”

I’ll roll these ideas into the next time I present on this topic, or indeed the next round of mentoring I do.

(Thanks to JJ for the photo.)

See what people are blogging about: ozia2007

The first day of the second OZ-IA conference went pretty smoothly. There were some great presentations, really giving us something to think about. And the Mercure hotel is a pretty good venue, although the room layout could have been better (those damn columns got in the way!). I’ll write a proper review shortly.

And of course this afternoon I delivered my Collaborative Information Architecture presentation. It was quite daunting presenting to your peers, much more so than speaking to clients or people in fields other than your own. So I was a tad anxious, but I managed to cover the material I wanted…even if there were a few stumbles.

My slides are now available on SlideShare.

I decided to have an early night and spend some time at home, but I’m looking forward to tomorrow. There’s still plenty of people to catch up with!

(Thanks to James for the photo)

See what people are blogging about: ozia2007

It’s official, I’ll be one of the happy campers presenting at this year’s OZ-IA conference in September (22nd and 23rd to be exact).

My presentation is called “There’s no I in team” – a case study in collaborative information architecture and will, as the name suggests, focus on doing IA in a collaborative team environment.

It’s great to have an IA conference of our very own, without having to travel to foreign lands. And we have such a great line-up of local experts, I think it’s shaping up to be a great two days.

Now I just need to work out what to talk about (shhh, don’t tell Eric).

See what people are blogging about: OZ-IA

Today I ran the first of my series of Information Architecture Fundamentals workshops. It was a good day, with a nice small group so we had lots of good discussion.

And it was a really great group of people too, with similar situations and levels of experience. Everybody worked in public sector, either health or education. And because these weren’t absolute beginners, I could dispense with some of the material in the course and discuss stuff that was a bit more advanced.

The feedback from attendees was really positive, with high praise for the lunch (oh and they thought I was pretty good too). The most frequent comments were that the group discussion was great, giving them a chance to hear other people’s perspective. The scope and level of detail was also appreciated. This is good since I deliberately tried to keep the focus on ‘big picture’ and properly defining the problem space, rather than jumping into the details of specific techniques, which IA training tends to do.

Next step Canberra, and there are plenty of places so I might see you there.

This afternoon I presented my topic at Open Publish 2007, in Sydney. It was quite a good turn out, but still a nice small audience.

My topic covered two central topics:

• Ethnographic techniques for ‘needs analysis’ and user research, and
• The application of these techniques for ‘Whatever 2.0′ solutions

This was well received and I enjoyed talking on this subject to this audience. Hopefully, being the last presentation of the day, I didn’t bore folks too much. There were also lots of other interesting presentations and altogether it created a great little conference. There’s definitely some good thinking in the publishing (and content management) space; not at all matching the external perception of the publishing industry.

My slides are up on SlideShare.

See what people are blogging about: open publish, user research

Over the next few months I’ll be running two public workshops on information architecture fundamentals. Here’s the blurb:

One-day workshop run by Patrick Kennedy of Step Two Designs. This workshop will present an introduction to information architecture (IA), the core methodology for organising and designing information-rich websites and intranets. Fundamental principles will be covered, along with a hands-on exploration of practical techniques. A must for all web teams.

The first workshop will be in Sydney on August 24th followed by Canberra on 26th October.

For further information and to register, visit www.steptwo.com.au/seminars/070824/ [Sydney] and www.steptwo.com.au/seminars/071026/ [Canberra].

This should be a good opportunity for me to brush up on my presentation skills and hopefully send my audience away with some knowledge and skills to help them produce better websites, intranets, software or whatever their particular thing is.