Web Services ROI Remains Tough to Prove

Computerworld - NEW YORK—With Web services technology still in its infancy, a disconcerting trend is emerging: Cost-justifying projects to senior management is proving to be a difficult task as many IT managers struggle to find effective and consistent approaches to quantifying potential payback.

That was a key message delivered by attendees at a Web services conference here last week sponsored by the Software & Information Industry Association, a Washington-based trade group.

But early adopters and industry participants said one way to help prove the value of Web services efforts is to develop "lightweight" pilot projects that quickly demonstrate the potential return on investment of larger initiatives.

"Find some pain and relieve it," said John Radko, chief architect of global technology operations at Global eXchange Services Inc., a Gaithersburg, Md.-based spin-off of General Electric Co. that provides electronic data interchange services to retailers and manufacturers.

Radko also suggested making a business case for a Web services project that can attack "an area of intense investment," such as a high-priced enterprise resource planning (ERP) system that isn't generating the expected returns.

Figures showing how a company could use Web services technology to improve the ROI of an ERP project "are the types of numbers that executives watch," Radko said.

A shortage of best-practices guidelines for Web services projects is another reason why IT managers should consider going after bite-size implementations that can demonstrate solid returns, said David Pawloski, a product director at Flamenco Networks in Alpharetta, Ga. Flamenco sells networking technology for Web services applications.

Uphill Battle

Companies can set up Web services to interact with other applications and manage specific business processes. But given current IT spending constraints, technology managers likely face an uphill battle in arguing the financial value of Web services projects to top brass, some attendees said. Complicating matters is an expectation that large corporate Web services initiatives will take three to five years to roll out.

James Adamczyk, chief technology architect at Accenture Ltd.'s global consulting unit for the financial services market, said a CEO who gets a proposal for a big Web services project may be skeptical because of what has happened with other highly touted technologies such as computer-aided software engineering (CASE) tools and e-commerce systems. "He'll probably say, 'I've heard this before with CASE tools. I've heard this with the Internet. I've heard this with object-oriented programming,' " Adamczyk said.

In addition, trying to measure the ROI of an internally focused Web services project is a lot different from trying to gauge the returns of initiatives that would have a direct impact on a company's customers or suppliers, said Bill Whyman, president of The Precursor Group, an investment research firm in Washington.