In designing, implementing, developing and maintaining the organisation of a multinational computer services company, one must first of all respect the constraints linked to service. In its development, as well as in its management, GSI has preserved its roots and the intentions that presided over its creation 20 years ago. With the passing of time, they have showed to be genuine key factors of success ; and the organisation that was set up to nourish them - in which great importance is given to the circulation of information - influences to a large extent practices and daily work habits, notably as regards marketing, communications both internal and external (where is the borderline between them?) and human resources management. Indeed, for "delocalized" service organisations, it is a vital challenge to preserve the nature, the form, the homogeneity of the service, while keeping relations of extreme proximity with customers who are often very different from each other - and that is particularly TRUE when you work in different countries. We are presenting here a few thoughts, and a few facts, from which the organisation of GSI was designed 20 years ago : this is the way we are facing the challenge ; we think that this method is the right one for GSI, we know how much tenacity it takes to keep on the same course ; and our remarks should not be taken as anything more than one example, without any pretention to make general recommendations
Several different organizational forms for a multinational manufacturing company are examined, and the implications of these different structures for the nature of the marketing research function: in particular the degree of centralization or decentralization. As an example of a matrix organization with a centralized marketing research department, Braun AG of West Germany is discussed. The nature of this company's approach to international marketing and research is examined, as are the operational advantages and disadvantages of centralized vs. decentralized market research. The optimal balance for this kind of operation is concluded to be a high degree of central control, coupled with flexibility and sensitivity in the way the research is conducted.
At the end of the first day, considerable discussion took place on the advantages and disadvantages of using chains of organisations. It was thought that the papers of Whitley, Haines, Berent and Jarvis nicely presented the arguments for: Central control of analysis and report but local autonomy of fieldwork (Whitley). Decentralisation of analysis and to a large extent, interpretation (Haines). International supervision and control using various local agencies (Berent). Specialized international department and tight control where local marketing staff is weak (Jarvis).
We understood this work essentially as an automatisation of the actual way of making budgets in a great international company. But the huge size of the model prevents a real back and forth movement of information between central management and decentralised units (feedback effects).