Timberjack Parts: Packaged Software Selection Project.
The case study “Timberjack Parts: Packaged Software Selection Project” discusses the process of selecting a packaged software for Timberjack, that could be used by the company globally. Timberjack is the leader in manufacturing professional logger equipment and machinery. The company has two service parts operations departments, one in Sweden and the other in the USA. The list of the requirements for the software package from both offices included UNIX software platform, strong support both in Europe and in the USA and excellent experience of the current wholesale users. Their goal, however, was to find a vendor, who would be able to provide all the abovementioned characteristics, satisfying the demands both the Swedish and American Timberjack offices. In the final selection round, the number of vendors was limited to two: QAD, suggested by the European side, and Oracle, favoured by the North American team.
The main problem in the software purchase was inability to come to the consensus between the two service operations parts of Timberjack. Despite the similarity of operations, volumes and number of employees, the demands from the two offices differed significantly. Already in the very beginning, the suggestion of the North American team to implement separate manufacturing systems for the company sites was rejected by the board of directors in favour of a more global software strategy and an integrated solution, which would provide standardized financial data across the company, simplify employee transition from one office to another and leverage negotiations with vendors. Further, during an RFP (request for proposal) preparation process, the difference of objectives and practices between European and North American teams have become more apparent. While in the U.S. RFP process has been highly formalized, leading to preparation of 200 pages report, in Europe selection process is much simpler. Moreover, since Swedish office was not satisfied with their current software solution, it was crucial for them to implement the new software as soon as possible. Highly formal and lengthy procedures for specifying requirements by the stakeholders and creating the RFP by the North American team were slowing down the implementation process. In addition to that, Swedish office was looking forward to a complete solution, which would be sufficient without further customization. North American representatives on the other hand valued flexibility in adjusting the software without touching the source code.
Therefore, the main problem for Timberjack was not the choice of the vendor, but rather coming to a unique set of software requirements, and assigning priorities. Should a consensus have been found in the first place, the two parties would have agreed on the common RFP procedure, common requirements and a joint approach to analysing vendors’ propositions. They would have similar objectives in terms of implementation time, level of detail and technical specifications, thus simplifying the selection process. Moreover, it would have reduced the negotiation time with the vendors, as it could help to specify clear requirements already during the proposal phase.
An alternative solution to the working process could have been closer cooperation between the two offices from the very early stage of the project, perhaps in one location. The RFP process could have been shortened, as suggested by the Swedish representatives, instead using the support of the selected vendor for more detailed description development.
Although implementing these strategies is straight-forward, the real world conditions are often different. Thus, despite closer cooperation and longer meetings between the two offices, the difference in objectives would most likely result in different system requirements. Developing detailed system description with the vendor would have at least two disadvantages. Firstly, it would significantly increase consulting cost, which Timberjack was trying to reduce during negotiations. Moreover, the final description would be biased in favour of the chosen vendor, while vendor selection process would be limited due to a small number of requirements, specified by the company.
References
Romanow, D., & Keil, M. (1998). Timberjack parts: packaged software selection project,
Harvard Business School, Boston, United States of America.
