Study: 'enterprise SOA' still a few years out

Quite frequently in this blogsite, I've talked about most shared-service implementations being more JBOWS (Just a Bunch of Web Services) than truly functioning SOAs. Now, a new study out of Saugatuck Technology research validates this point.

The study's authors conducted in-depth interviews with 40 end-user companies and 12 vendors. At least fifteen (or 37%) of the executives participating in the study said they were currently in a limited or full production stage of SOA deployment. However, according to Saugatuck, upon further probing, "it became clear that many are merely managing a collection of Web services, and have yet to make a strong commitment to SOA as a management discipline -- as opposed to an integration technology." 

One of the challenges of SOA is that often, service-orientation is in the eye of the beholder. What constitutes a "proper" SOA? ESBs?  Wrappered services from legacy systems? CORBA? All of the above? Not all SOAs will be elegant, and some may be downright ugly. Just last month, I reported on (some say refereed) a blogosphere tussle around the legitimacy of "generated" services as truly part of SOA. (Here, here, and here.)

Saugatuck concludes that SOA is now a technology-led approach, versus business-led, as found in an earlier study from 2005. This, of course, flies in the face of everything we're constantly being told about SOA -- a mantra that is regurgitated repeatedly at conferences and in trade journals: it's about the business... it's about the business... it's about the business... Okay, one more time -- it's about the business....

The perceived benefits of SOA at this time are mainly IT-centric advantages, rather than business gains, Saugatuck concludes. The study finds that by a wide margin, cost reduction outdistances all other motivations to move to SOA by a two-to-one margin. In the study, 57% cite cost reduction, 27% cite code reuse (which is cost reduction), and only 23% expect to increase business agility from their SOA any time soon.

Ironically, while the greatest perceived gain at this point is cost reduction, companies will actually see little of that, but, in the long run, will actually see greater agility. The study's authors conclude that through 2012, SOA implementations will not substantially reduce IT costs, blaming "steep learning curves, lack of technology skills and familiarity, lack of mature industry standards, and lack of user enterprise management ability (governance) to plan integrative SOA implementations." However, within a few years, they add, "enterprises adopting SOA will realize significant business benefits and ROI through improved business flexibility and responsiveness to competitive situations." 

The Saugatuck study pours cold water on the hopes and vision of many SOA proponents, but concludes that we'll get there, slowly but surely. SOA efforts are currently hampered by rework, failed projects and immature technologies within many early user enterprise SOA deployments. Users often confuse such typical project benefits as application integration with the true, longer-term benefits of SOA, such as more responsive and agile application architectures and reuse of services.

The study's authors point to three challenges that are slowing the journey to full-fledged SOA: 1) difficulties in securing funding for SOA projects; 2) changing business manager mindsets toward the sharing of computing resources; and 3) a lack of concerted governance efforts. As Saugatuck observes: "SOA is not suited for a short-term IT or business investment strategy. SOA instead is really designed to yield its greatest promise andbenefits over the course of years and decades."

Nevertheless, Saugatuck predicts that we're entering the "Third Wave" of SOA, in which projects are moving from pilots or confined departmental services to more of an enterprise reach. We won't see a lot of the third wave, enterprise-wide SOA, for at least three years, however.