Page 78 -

NEW ORGANIZATIONAL FORMS THAT SUPPORT KNOWLEDGE WORK 67

organizations struggle with their ICT projects – the Standish Group (2007)
estimated that 70 per cent of software projects fail. Successfully implementing
a so-called ‘knowledge processing’ system like sharepoint, therefore, requires
managing organizational change. Understanding how to effect change using
ICT is therefore extremely important. There are two important points to
remember in relation to this (Wagner and Newell, 2007):
1. Achieving a workable solution does not mean that within an organization
there needs to be consensus over all issues related to the ICT. Rather, the
important issue appears to be coordinating action that will allow broad goals
to be achieved, even if this involves compromise along the way. This is because
consensus about how to configure an ICT is likely to be very difficult, given
the diversity of users who will be affected. Different groups will want different
things included/excluded and if consensus is sought this is likely to lead to a
stalemate. We will see this illustrated in the Uni case at the end of the chapter.
2. Achieving a workable solution is more likely if treated as multiple cycles of
design (by which, in relation to packaged software, we mean configuration
and perhaps some customization), implementation and use rather than a sin-
gle phase of design followed by implementation and use. There are a number
of reasons for this:

a. Legacy thinking: Evidence suggests that users find it difficult during the
early stages of requirements definition to see beyond their current prac-
tices and anticipate how things could be done differently if they had new
tools, for example, to enable more knowledge sharing within and across
functions. This is because much work practice is rooted in everyday inter-
actions (Suchman, 1987) so that trying to appreciate a new way of work-
ing by just looking at the technical system is difficult. Add this to the users’
limited technical knowledge (Beath and Orlikowski, 1994) and it becomes
understandable why, when users are involved in the initial configuration/
customization phase, their main concern is that the new system will enable
them to do what they did before – with ‘as little change as possible’.
b. Vanilla implementations: Packaged software does allow configuration
options, but only within fixed parameters and the general advice is to
avoid, as far as possible, customizations to the package. Given this context,
it begs the question of the point of user involvement if users are not lis-
tened to because of the desire for a ‘vanilla’ implementation?
c. Motivation: A final problem during the pre-implementation phase is actu-
ally getting users to be interested in participation. This lack of engage-
ment is a reflection of human nature, where we only become interested in
something when it is salient to us and when we can actually begin to learn
about the technology through practice and participation (Wenger, 1998).
Moreover, even where users genuinely want to be involved, they are often
distracted by more urgent things that they are faced with, especially for
knowledge workers who are typically very busy.

6/5/09 7:01:07 AM
9780230_522015_04_cha03.indd 67 6/5/09 7:01:07 AM
9780230_522015_04_cha03.indd 67