This posting is in response to Ralph Lord's C&S request and Ash
Donaldson's following discussion posting:

> However here in Australia, as Tania exemplified with her anecdote, I have noticed that it is
> not uncommon for heuristic evaluation to be performed by one practitioner, and used in
> isolation (and in this case it seems, in the final stages of production).  Of course, I
> understand that this was due to commercial constraints...

See separate posting comparing usability evaluation methods (to keep
this posting on topic).

I think an overview of my current project may provide Ralph with one
example of how to integrate UCD into an existing development process
(and possible clarify any misunderstandings generated from the above
posting at the same time).

I also refer readers to an excellent article on this exact topic:
Metzker, E. & Offergeld, M. (2001). An Interdisciplinary Approach for
Successfully Integrating Human-Centered Design Methods into Development
Engineering for Human-Computer Interaction: 8th IFIP International
Conference, Toronto, Canada.  Available at:,5,28;journal,590,1358;linkingpublicationresults,id:105633,1

Similar free article by one of the same authors:

Tier 1 IT/Management consulting firm doing large ERP implementation, a
SMALL component of which is a customer web application interface.
Waterfall development methodology being used. Consideration of user
experience / usability / interaction design from development team
perspective and experience only. No usability people in project team.  I
understand the project is already way over budget.

First it is worth noting that my engagement was conditional upon adding
value where I could but not delaying the project launch date (which I
understand given the political environment).  There were also limited
funds and they could only afford one person to help out.

when I started, I already had a high understanding of the specific user
group and their goals (from focus groups and usability testing I
conducted on another project within the organisation).  If not, I
wouldn't have proceeded without this understanding.

I advocated usability testing from day one and "test early - test often"
approach.  I tried to show how this could occur within the existing
methodology and timeline.

I conducted an expert review of functional specification including
application flow and preliminary screen shots (which really just showed
the fields).

I highlighted numerous POTENTIAL usability problems and provided
suggested solutions/interaction design recommendations. To help convince
and educate the client, I gave verbal examples from previous test
sessions to support my analysis.

I stressed to my client the drawbacks of the expert review approach and
that my findings really only identify potential problem areas and should
be used as a basis for usability testing.

I have managed to convince the client to at least undertake usability
testing prior to launch at the same time as schedule UAT.

I think that as usability practitioners, we need to add value where we
can on projects we have no control over and integrate appropriate
ethnographic and user centred design tools and techniques where we can
into existing development processes.

I wholeheartedly agree with Ash that we have to educate managers and
strive for a more rigorous approach.  However, we also have to be
realistic (and this is often more difficult as an outside consultant).
I fully advocate user involvement through all stages of development but
believe techniques such as expert reviews/heuristic evaluation can play
a role in any development methodology (especially early on in projects
that you don't have control over).

To put the use of different techniques into perspective, I have
conducted 5 heuristic evaluations this year (three of which led to a
redesign using a UCD approach). Conversely, I have conducted 64
usability test sessions.  I think this speaks pretty well as to which
technique I generally advocate.

Tania Lang
Peak Usability
Brisbane Australia

        Tip of the Day: Forward out-of-office replies to
                    mailto:[log in to unmask]
     CHI-WEB: POSTINGS: mailto:[log in to unmask]
              MODERATORS: mailto:[log in to unmask]