WHEN IS VIRTUAL VIRTUOUS - ORGANIZING FOR INNOVATION

Champions of virtual corporations are urging managers to subcontract a nything and everything. Because a number of high-profile corporate gia nts have been outperformed by more nimble, ''networked'' competitors, the idea of the virtual organization is tantalizing. But is virtual re ally the best way to organize for innovation? Henry W. Chesbrough and David J. Teece argue that the virtual corporation has been oversold. I nnovation is not monolithic, and it is critically important to underst and the type of innovation in question. For some innovations, joint ve ntures, alliances, and outsourcing can play a useful role. But for oth ers, they are inappropriate - and strategically dangerous. The initial success - and subsequent failure - of the IBM PC illustrate the strat egic mistake of using a virtual approach for the kind of complex techn ology that should have been controlled in-house. The authors present a framework to help managers determine when to innovate by going virtua l, when to form alliances, and when to rely on internal development. T hey provide a range of cases to illustrate how to match organisational strategy to the type of innovation being pursued. General motors, for example, used the wrong approach to develop disk brake technology and paid the price: getting to market later than its competitors. To real ize its vision of ''tetherless communication,'' Motorola must choose a n organizational strategy allowing it more control over the direction and timing of technological change than a virtual approach could provi de. In contrast, the virtues of the virtual organization are illustrat ed by Ameritech's use of alliances to influence the innovation path in multimedia.