I’m in the process of studying what web services (now just called ‘services’) management costs IT organizations. Specifically, I’m looking at what the various cost factors are considering both an environment with a service-orientation mindset (SOA, ESB, SOI) and without.
One thing I’m seeing so far is that the service discovery aspect is significant. Many companies don’t have a registry (UDDI or otherwise) that allows project teams to find services that may already exist. I find this very interesting since the number one (arguably) benefit IT shops hope to gain from adopting services is reuse. How can an organization take advantage of service reuse if there’s no viable way to discover them?
The cost to rebuild a service that may exist in another department in the company is probably significant. Not only that, the cost of maintenance and monitoring of the ‘new’ service is not insignificant.
Another thing I’m seeing so far is that companies don’t have the capability of treating a ‘service’ as business service from a management perspective. In other words, IT is still in the mindset of ‘managing’ individual services or components without looking at the whole application or ‘business service’.
I wonder what your experience is at your organization. Feel free to email me or comment here.
I say that service management is the tip of the the overall app portfolio management spear because I’m seeing a trend toward more service orientation where web-based applications are becoming the norm and they are being built from ever-smaller, autonomous, stateless components (i.e. services).
More insights will follow as my study progresses. In particular, I hope to quantify the costs of service management as well as bringing some quantification to the value side of the equation. Stay tuned for that.
All the best,