Recently, I interviewed Anne Thomas Manes of the Burton Group. Of course, we chatted about the SOA obituary she wrote earlier this year and the fallout from that piece.
Once again, she reiterated that she thought service-oriented architecture is a good idea, but what she wanted to get across is that it's gotten such a bad name, it's no longer a good idea to go to your CEO and announce you need funding for SOA.
As always, it's better to focus on delivering business value and how you'll do that. In fact, she said some of the most successful SOA initiatives she's seen never used the term SOA, even though that is, in fact, what the IT departments built. She pointed in particular to Bechtel, a company she mentioned in the original SOA obituary post:
"The latest shiny new technology will not make things better. Incremental integration projects will not lead to significantly reduced costs and increased agility. If you want spectacular gains, then you need to make a spectacular commitment to change. Like Bechtel. It's interesting that the Bechtel story doesn't even use the term 'SOA'-it just talks about services."
There are SOA success stories out there; we've highlighted or published interviews on quite a few SOA successes here on IT Business Edge. My personal favorite that I've covered is VetSource, which service-enabled all its IT functions so it could create new applications "on the fly" for new customers. (Gartner's Nick Gall would no doubt argue it's also a great example of WOA.)
But Manes is pretty particular about what she considers a true SOA success story. As she pointed out in March, companies have built beautiful SOAs that petered out, in part because "... the techies have not been able to explain to the business units why they should adopt a better attitude about sharing and collaboration -- which is the fundamental cultural shift required for SOA to succeed."
I suspect this is the source of a lot of frustration with SOA.
It's not that companies aren't trying still pursuing SOA-in fact, a recent CA survey shows SOA deployments are on the rise, with 73 percent of organizations claiming to have "deployed an SOA application."
And therein lies the problem. As Manes and many, many others have warned, SOA isn't "an application." It's an approach to building many applications-it's service-enabling all your IT functions, or, ideally, the business functions.
If you're trying to do it with one application, then you're very probably going to be disappointed with the big-picture results. Hence, as SearchSOA.com reports
"The survey results show that between 70.8% and 76.l6% of respondents felt their organization's SOA deployments adequately met expectations. But, in the area of SOA performance, for example, 10% said deployments failed to adequately meet expectations."
It seems to me what's needed are better role models for success. What does true SOA success really look like?
Here's how Manes described the one success story she had found in March:
"This company reorganized IT around functional capabilities (rather than business units) and established strong positive and negative incentives that encourage people to adopt a better attitude toward sharing. I'm beginning to think that this is the only path to SOA success."
SOA isn't dead as an architectural practice-it's just dead as a stand-alone business initiative. But, there are a lot of people out there feeling jaded and disillusioned about the whole concept.
Manes told me there are three case studies that she felt really demonstrate how SOA can transform IT and the business when it's done right. I thought it might be helpful to share success stories that show how SOA can transform IT from a bottleneck (and sometimes outright dam) to a business-enabler, when it's done right.
I've tracked down articles on what Manes says are the three best examples of SOA success stories:
- Bechtel: This Network World article explains how Geir Ramleth, CIO, rebuilt the IT system of Bechtel, a global engineering, construction, and project management company.
- The British Telecom: Joe McKendrick called BT the "poster child of good Big SOA" in September, 2007. SOA allowed the BT to close down to 800 systems, a number that may climb by 700 to 900 systems more in the near future. It's easy to find articles profiling BT's SOA, but for simplicity's sake, here's one we featured, published on TechWorld in 2007; it's still available. You can also read my 2007 blog post, "Three Key SOA Lessons Learned from British Telecom," for more links.
- Cigna Insurance Group: The Cigna Insurance Group presented its SOA case study at the 2008 Burton Group Catalyst Conference, which was covered by SearchSOA.com. It's also interesting to read this CIO.com article from 2003, which shows you the trials and tribulations of Cigna's IT department before a new CIO entered the picture in 2007 and started the company's SOA initiative.
If you'd like to read more SOA success stories, IBM maintains a list of IBM-related SOA success stories. I'm not sure they'd meet Manes' test of a true SOA success, but they might be of interest to you. This month, case studies of the Bank of New York Mellon and Ball State University were added.