Monday 15 December 2008

Where to start?

OK, so I have just returned from OZCHI and I'm all fired about getting this blog up and going.  Hopefully it won't amount to shouting into an empty chasm.  Then again, if it does, at least it will be a private humiliation.

I'm working with a big government client right now who are in the early stages of requirements analysis for a large project.  There are a few ways that we can add value as User Experience Designers at this stage of a large project:

1. We can start our user research.  For me, this often takes the form of Contextual Enquiry (watching users in their habitat of use) - an applied variation on the broader ethnographic work pioneered by social scientists.

2. From observations made in the above-mentioned site visits, we may be able to articulate personas - which are richly rendered narrative and pictorial portraits of representative users of the system.  These include synthesised details about demographics, attitudes (and importantly) goals and objectives.  Goals drive users to perform tasks, and we design functions on the system to allow our users to complete tasks.  It is important that we think in terms of goals first as these will be our foundation.  Tasks may come and go as we bring the technology (or other solutions such as new business process) into play but the original goal will always remain - unless we see a change in scope.

3. Next we might want to describe some Context Scenarios.  These are "Day-in-the-Life" type descriptions of the steps that one (or a number) of our personas might undertake to achieve their goals.  Through these scenarios, we imagine how the user might do their work with the system supporting them.  Cooper suggest treating the system as though it is magic at this stage.  In other words, try not to get too bogged down in the detail of technical implementation.  On the other hand, any solutions need to be reasonably feasible within reason.  This is where we are really doing the first bits of design at a high level.  The scenarios help us test out our functional requirements (e.g. use cases) later and give an indication of major user pathways through the functional areas that are already bubbling up in everybody's minds.  In other words, they can help us move towards a picture of a conceptual model and therefore, possibly, a navigation model.

Alan Cooper's About Face 3 has much more excellent detail on theses (and more) important steps in the early stages of requirements analysis
   
Finally, if all else fails at this early stage of the project, we can act as business analysts and help the team get to a point where we have enough information to begin what we should be doing.  A good UX person will have most of the skills required to draw out business requirements as well as user requirements (these two types of requirements are quite different and will be the subject of a subsequent post).

No comments: