One of the bits that I've talked about at various conferences and in the book is using SOA to understand the business value that various services have.
I was chatting on the phone today with someone around this topic and I thought it was worth a post on why treating SOA as a technology thing missed the real power and opportunity of what IT can deliver. There was a research report from the Economist recently that said that IT would need to move away from reducing cost and towards delivering value and of course that the business and IT have different views on how this will happen and what the barriers would be.
But lets start first with the IT department being expected to deliver value, this means that you have to understand the bits that add value, and of course the bits that don't. If you are viewing SOA as a technology thing then it really isn't going to help, as you can't really start ascribing value to a WSDL or a BPEL process, its just to low level to consider investment or cost cutting down there.
This means you have to have some way of understanding the business, some way of understanding what you will need to deliver to the business, and therefore some way of understanding the different values and drivers of the different parts of the business. Some people might say "That is what enterprise architecture is for" but I'd have to disagree as its really just the first step in enterprise architecture and its also the first step in actually managing an IT department, and I don't see the concept of "value" early on in the likes of Zachman or TOGAF .
This is why I argue for a simple approach to the business service architecture, because it quickly gets us to the stage where we can understand what the services are and we can then use that information to understand the business value.
By understanding which services are "under the line" and which are above the line you can quickly understand where you should cut cost, and where the business would appreciate suggestions as to added value. Speaking to someone from work a week or so ago he mentioned working with someone who did a similar exercise a few years back and found that they had over 50 projects in areas below the line.
As the business starts expecting IT to deliver more value its going to be essential for IT to understand more about which parts of the business deliver actual value, and which are just IT things that we think add value but in fact no-one actually cares.
For instance, how many Finance or HR services would live above the line? How much competitive advantage is there in having a customised Invoicing process? How much advantage is there in having an EAI tool or an ESB? How much value is there in REST v WS? Once you understand the value you can start delivering actual benefit and realising that the technology isn't important, its the end-game that matters.
For SOA to give competitive advantage it means knowing what advantage looks like.