Around ten years ago, just about everyone around the world began embracing a set of open standards for networking, content, e-mail and similar capabilities that we collectively viewed as "the Internet." But equally important, in an outside-in approach, businesses, governments and other institutions embraced these same standards internally, and built what we called "intranets," as well as reaching out to connect to their supply chains with "extranets."
This outside-in approach -- both for the Internet and WWW as a whole and for individual businesses -- is so obvious in retrospect that it is hard to imagine that our behavior was almost diametrically opposite for so many years. For the most part, businesses used products based on proprietary and incompatible standards from different vendors. You'd even find that for individual departments within a single enterprise. Not surprisingly, the result was a Tower of Babel that made integration difficult, costly and time consuming. Years ago, just sending an e-mail to someone outside your company was a challenge. Then we noticed that those early users of the Internet in universities and research labs were easily communicating with each other around the world, not only exchanging e-mail, but also sharing files and documents. After awhile, the light bulbs went on.
That standards-based success of the Internet, the continuing advances in technology and the rapid progress we have made in integrating the IT infrastructure are naturally leading us now to extend such integration efforts "up the stack." We are starting to tackle integration at a business-wide level, and, in particular, at the business process level. Helping businesses integrate their processes end-to-end across the company and with key partners, suppliers and clients has been a key objective of our On Demand initiative in IBM. For the first time in history, we can seriously contemplate building globally integrated businesses, industries and economies.
Moving toward such a globally integrated model of business requires an outside-in approach, similar to but harder than that which transpired in IT with the advent of the Internet. For over one hundred years, the vertically integrated firm has been the dominant model for organizing work, especially complex work, as opposed to having multiple, independent, self-employed people who contract with one another to get the same work done. The reason for this was first explained by British economist Roland Coase in a 1937 paper called "The Nature of the Firm," for which he won the 1991 Nobel Prize in economics.
As Coase explained it, there are additional transaction costs incurred in obtaining goods and services outside the firm, in the marketplace -- tasks such as searching for the right people, negotiating a contract, coordinating the work, managing intellectual property and so on. Firms will arise when they can get the work done internally with significantly lower transaction costs than going outside the firm for the same services.
In other words, businesses have followed an inside-out, "self-centered" model because of the high costs of orchestrating and integrating their various processes, many of which are still labor-based and one-of-a-kind and therefore require quite a bit of management and coordination, which is most efficiently handled with a "classic," hierarchical organization. Coase also pointed out that, for a variety of reasons, there is a natural limit to what can be produced efficiently within the firm, which is why all businesses also have a more or less extensive supply chain, and strive for an optimal balance between what work gets done inside and outside the firm.
This balance is now in flux. Since we can now use technology, the Internet and open standards to begin to automate, standardize and integrate business processes, those transaction costs described by Roland Coase are dropping precipitously. Consequently, the whole nature of the firm, and what it means to run an efficient business, is going through very extensive changes. These are not easy changes. Not only is there a great deal of innovation required to automate and integrate business processes, but perhaps more important, there are even greater changes in culture required to transform Industrial Age business models to something more appropriate to our Internet era.
In the end, the evolution toward the globally integrated business is inevitable, and I believe that it will put more of a premium than ever before on continuing innovation. For businesses as well as nations, this is one of the most important challenges for the 21st century.