Late last year, at an SOA-oriented IT industry event held in Las Vegas, visitors to the BEA booth were asked to fill out a short survey on how their respective companies address SOA governance. While thoroughly unscientific, this informal survey offered a glimpse into:
Anyway, let's look at the data.
On the road to SOA Governance
Question #1: At what stage is your SOA governance program?
No real surprises here. More than 90% of the companies represented in the survey are somewhere along the path the SOA governance. But what's the deal with the 7% with no plans? The next question sheds a little light on that issue.
No Governance, no SOA
Question #2: How important is governance in your SOA strategy?
Again, nothing particularly surprising here, though for those who believe that SOA governance is "somewhat important," I wonder about the tipping point. Do they have a specific milestone in mind -- a certain number of services, or a certain number of service consumers -- or are they waiting until the need becomes obvious? When they reach that point, will they regret waiting?
As to those who responded that their SOA is doing fine without SOA Governance, it's interesting to put that response in the context of the "no plans for governance" group in Question #1. If we can assume that those who responded "not important" in Question #2 represent some percentage of those who responded "no plans" in Question #1, what does that say about the SOA programs at the rest of the companies that have no SOA governance plans?
I'm no statistician, and my math skills are exactly why my wife hides the checkbook, but the numbers suggest that there are ungoverned SOA programs out there that are doing a bang-up job of wasting lots and lots of money. Failed or stalled SOA initiatives are hardly uncommon, but if those initiatives had placed a higher priority on governance, how might the outcome have changed?
The people problem
Question #3: What is the biggest challenge you have encountered in adopting SOA?
What's interesting here is that "Introducing governance" turns up so low on the list. That's not to imply that getting people to go along with SOA governance is a walk on the beach. But in the experience of survey respondents, it appears to be significantly less of an issue than cultural, educational, and organizational challenges. If the introduction of SOA governance is the path (or at least a path) of least resistance in SOA adoption, doesn't it make sense to start there, rather than approaching governance as an afterthought?
What's in the toolbox?
Question #4: Which of the following is the most important SOA governance technology?
The registry repository comes out the clear winner here, a response that makes sense given that the registry repository has the broadest scope with regard to the entire SOA lifecycle. Overall, the responses indicate varied perceptions of what SOA governance is, and where the governance focus should be placed. That perception and focus -- and the initial selection of tools -- is undoubtedly driven by the specific, unique pain points in each organization's SOA governance experience. But the bottom line is that each of the products listed plays an integral role in end-to-end SOA governance. So from that perspective, the right answer is "all of the above."
Note to vendors: make sense of the mess
Question #5: Which of the following criteria is most important when selecting a governance vendor?
While the responses to Question #4 may indicate some uncertainty or even confusion about the specifics of SOA governance, the responses to Question #5 offer some indication of a broader understanding of the general issues, or an indication that the respondents at least know what they don't know. Even without a calculator, that adds up to the need for overall solutions that will help companies fit SOA governance into their respective ecosystems.
Common goals: agility and efficiency
Question # 6: Complete this sentence: SOA governance will help my organization become…
Among the responses -- virtually all of which expressed the idea of better, faster, cheaper response to business requirements -- more than 30% of those who reported no plans for governance in Question #1 responded in Question #6 that SOA governance would result in improvements in agility and efficiency. One possible explanation is that the people who responded to the survey understand the need for SOA governance, but their bosses do not.
This modest survey paints a picture of widespread recognition of the importance of SOA governance, and a general migration toward SOA governance in some form. That's good news. The indications of confusion about SOA governance are of some concern, but experience is an excellent teacher.
How does your experience with SOA governance compare to that of the people represented in this survey? How would you respond to the questions?
Previous posts in the SOA Governance@Work series: