Sometimes thinking differently is a good thing. I'm not sure if this is one of those times.
To be honest parts of the podcast are pretty good (btw - Dana's podcast is definitely worth subscribing to). The ROI discussion bogs down a bit too much on reuse for my liking. I don't think reuse should be the main driver for SOA - and I don't seem to be alone on that thought.
Reuse is important but its also incredibly difficult. Reuse requires locating and understanding what a resource provides - if I can't locate a resource that does what I need should I assume it doesn't exist or have I simply not searched properly? (even worse - has the resource I need been identified with a set of metadata that I can understand and locate?)
Reuse is but one aspect of ROI within SOA. There are many others that should be considered such as:
- Impact on TCO of the associated processes
- Revenue impact
- Impact on risk
In the end ROI, like SOA itself, is not an exact science - benefits may take a year or more to be recognized, depending upon the metrics being tracked.
Prescribing a "one size fits all" approach to ROI is just as wrong as trying to prescribe a single model of SOA for everyone. Every organization is unique with different needs and business drivers. What works for one organization may not necessarily work for another (be it SOA or ROI metrics). Keep that in mind the next time you read about the ROI of SOA or identifying what is or isn't SOA.