The researcher combined the case study approach with a vigorous action research iterative cycle so that the current learnings were used to inform and test future actions in a focused, conscious attempt to add to the body of knowledge relating to the project management of ERP implementations in terms of the benefits of applying learning from a previous project to enhance the methodology applied in subsequent projects.
A full change management programme was put in place for each implementation to track, guide and inform the changes that were occurring within the context of the organisation ("the context of learning") and to suggest actions and decisions required at a project management level to reduce risk and assist business users in coping with the extensive changes to their work environment.
4. More than one ideal-type of the process of learning should be used as reference for interpreting observations and other data. Learning is a concept with different meanings. The process of learning is a complex phenomenon which can only be described with the aid of simplified or ideal-type models. It seems unlikely that any one of these models represents the reality of learning to a satisfying degree. Using two or more ideal-type models
simultaneously may help us to understand which model has more explanatory value under which circumstances.
This research made use of a number of research instruments to capture the learning taking place through the ERP implementations, ensuring a wide range of data sources were collected and analysed so as to give as wide a view on reality as possible, and to validate this through the triangulation of results. One of the recommendations
presented in this research was the proposal to adopt a new model of project learning that resulted from this research. This proposed model would require further validation through corroborative research before it could be accepted as a useful addition to the existing body of knowledge on ERP implementations. As previously discussed, the generalisability of action research is low, although due to the iterative nature of the research conducted, in terms of research in consecutive implementations, clear trends emerged.
5. A multiple of projects should be included in the investigation. Research on the causes of project performance has been dominated, so far, by cross-sectional surveys including large numbers of projects. Research on learning in projects, particularly the Project Based Learning stream of research, has been dominated by non-comparable case studies. What we need now is a stream of longitudinal research in which a multiple of projects is included.
The research conducted was clearly a longitudinal study in that all the ERP
implementations were run consecutively using the same basic project team, working under the same overall objectives, in the same broad business environment and strategy, with consistent project management in place, and consistent tools and measures by which to measure delivery against template plans and methodologies.
This research enabled direct comparison to be made between multiple projects in a stream of longitudinal research.
6. The comparability of the projects involved should be assessed or, preferably, controlled. It is quite likely that type of project has an influence on learning within the project. It has been shown, for instance, that learning is less likely in short-term project teams with a strong performance orientation (Druskat and Kayes, 2000).
Due to the action learning approach followed from the outset of the ERP implementations, combined with the factors mentioned in point 5 above,
comparability was high for this research. The projects were of sufficiently long duration to allow for in depth reflection and assessment. Indeed this was a
standardised requirement imposed on the project from its controlling body, The Group Project Management Office (PMO).
Storm (ibid, pg 11) further points out, as per the italicised points below, a number of challenges that should be addressed:
• To investigate "knowledge in action" the researcher must be rather involved in, or at least very close to, that action. But, being involved may imply influencing the process of learning itself.
The researcher was involved at a project management level in these projects thus was very close to all activities taking place. It was part of the researcher's mandate to ensure that a learning process was in place so as to optimise future implementations thereby meeting the stated objectives of delivering each successive project faster, cheaper and at acceptable quality levels. Indeed, the researcher went beyond merely influencing the process of learning to actively promoting it.
• Investigating the actions and corresponding learning behavior of project managers over the course of their projects requires heavy investments on the side of the researcher as well as on the side of the project managers. The risk that these investments will not pay off as expected are high. The dilemma is that the greater the length and depth of the study, the higher these risks will be.
With the researcher working as a project manager in the project, this risk was negated and the researcher was able to ensure the correct depth of study was conducted throughout the length of the projects.
• Selecting similar projects will increase comparability but reduce the chance of finding strong variances in learning. The quintessence of these dilemma's, it appears, is to design the investigation in such away that it optimises the usefulness to both the
researcher and the practitioner. Research suggests that the following measures may help to increase the perceived usefulness:
o Action planning is related to the research effort
o The research enables dialogue within the organisation o The research uses joint interpretative forums
o The research promotes mutual perspective taking.
As previously emphasised, by having the researcher working under a clear mandate, requiring a learning process to be part of the implementation, assisted in ensuring that variances in learning were evident. Action learning was a formalised, accepted part of the project, facilitating extensive dialogue at all levels, through the use of interpretative forums which promoted mutual perspective taking.