Decision Support: It's Time For Web ServicesDecision Support: It's Time For Web Services

Web services will let innovative business models evolve and will support current models by making the integration of applications easier. However, Web services standards and technology must mature before we'll have the most innovative and valuable models, and we have to learn from early Web-services experiences.

A recent example is Microsoft's .Net My Services, which originally required users to rely on Microsoft to host a private, secure, and highly available identity service. Several large companies have indicated that they don't trust third parties to host business-critical identity data. In response, Microsoft recently floated a model that provides products that let companies host their own sensitive identity data. This is positive, though Microsoft will probably need to ensure interoperability with other identity standards, such as those the Liberty Alliance is developing.

Most companies considering the business potential of Web services don't face the same issues as Microsoft, although one lesson is that the viability of a service depends as much on the relationship between service provider and user as on the nature of the service itself. Another is that security and trust are at the top of every company's list of obstacles to Web services. New standards will probably resolve these issues by late this year, with products implementing those standards emerging from then into mid-2003. Take-up will be gradual, so until later next year, the only viable Web-services business models will be those that don't require security, trust, or transactions, or that can use an interim workaround.

There are other obstacles as well:

The more a model requires the relationship between service provider and user to change, the more disruptive it is to the company using the Web service and its customers' perceptions. Consider a Web service that provides court-docket information from a provider of legal information: The provider already knows how to sell to those customers, and they know to look to that provider. New services that change the way they interact may require new channel strategies to reach the customer, as well as substantial marketing investments to change perceptions.More valuable services often require more significant and difficult business-process reengineering. A wise man once said, "You can get people to do a new thing in an old way or an old thing in a new way, but ask them to do a new thing in a new way, and you're asking for trouble." This is one factor driving early Web services to existing services or software interfaces, rather than to entirely new services.The most logical source for new Web services is existing service providers in financial, telecommunications, and travel markets. That's where the most active experimentation with the technology is. However, these areas also most often require high volume and a high quality of service that aren't yet available and may never be cost-effective for plain-text Web services.A credit-card processor would be unlikely to replace its existing interface to the approval system from card readers in the field with a Web-services interface. But it might well seek a new market in which Web services would let payment services be offered in a new way and where the volume and service levels required would be feasible. Ultimately, special forms of Web services may evolve to meet the needs of the high-volume transaction processor, but for now, such limitations have to be considered.The number of Web services with real business value will start to grow during the second half of this year, making it possible for the target users of these services to begin to consume them by early next year. It's likely that many companies with significant digital assets, or those that already generate significant revenue from services delivered from an IT infrastructure, will find real opportunities to translate new Web services into incremental revenue.It's important to recognize that the more interesting a Web service is from a business-value perspective, the more apt it is to require significant effort to implement. Consider the need in many cases to build a Web-services interaction layer on top of an existing set of transactions or interfaces. This will take time to design, implement, test, and release and will often require testing with a few partners. Long test periods are especially critical when a technology is relatively immature. This will allow feedback to be addressed and should significantly increase the chances of success.Web services will probably evolve much as E-commerce did: A few new companies may be created by this wave, but the bulk of the business will ultimately go to existing service providers that have the deep pockets to endurethe large investments and slow ramp-up that are likely to be typical of new Web services for years to come.Robert K. Weiler is chairman, president, and CEO of Giga Information Group Inc., a global technology advisory firm. Reach him at [email protected]. Research Fellow Mike Gilpin contributed to this column.To discuss this column with other readers, please visit the Talk Shop.