I had a simply amazing client inquiry Monday this week. It was with a bank. This bank was focused on how to relate/explain/integrate its own internal information model to other organizations. This information model, describing the organizations it works with, needs to “integrate” to a host of external organizations (other banks) and support external regulatory requirements. And “integrate” here does not mean “just make work with others”. It means “interoperability” so that agility in process change is baked in. Integration does not deliver that.
Clearly the original request was more along the reporting lines (regulations lead to reports – plain and simple) but the architect at this bank were taking the long view, and realized that “BI” was not the answer – it was seen as a band aid. The real source of the problem (inconsistent semantics and what means to business performance and outcomes) was centered on how business data is created in the first place across the bank. This the real starting point for this IM program was Master Data Management (for operational business systems, not the data warehouse for BI).
However, it turns out that this bank already has an MDM program underway. This was great, if new, news to me. The MDM program was not yet totally enterprise wide, but it was in flight, and focused on a couple of key master data objects. The program was going well – but the challenge this client was this: how to sell to the business the need to govern other data, not just master data, in order to increase quality and consistency of important data that drives better business outcomes. This question goes the heart of what we do in EIM research.
The guy on the end of the phone said something like (and I paraphrase), “We know that we actually need to govern metadata across a whole host of IM silos, including MDM, core banking systems, other business applications, all the way down through into the data warehouse, but how do we get the business interested in “governing metadata for enterprise wide information consistency?” This was like the most exciting inquiry for a long time. This is a tough question. We dare not mention the word “metadata” to a business person; their eyes will gloss over in 1.3 nanoseconds. When we mentioned, for the first time, “master data”, at least they asked, “What did you say?” which gave us a second bite at the cherry. This allowed us to say, “Single view of customer” and that was magical. It allowed us the opportunity to speak to business people on their terms. This is what we need to do with enterprise class metadata management programs.
This is not an attempt to “another metadata management” silos. We have too many of those. This is an attempt at:
- An enterprise wide (when its complete) discipline focused on aligning and so governing the semantics that underpin all the important information spanning all the important business processes of our origination
- Focused on metadata (but don’t use that term)
- For better business value
That is the key to the kingdom.
Alas after the call I wanted to drop everything; hold the phone, stop the email hell, and get all my thoughts down on paper and get a new research note drafted. Unfortunately there was not enough time for that. But the ideas are percolating nicely. More research will follow. Monday was a great day – I was invigorated.
Category: Enterprise Information Management (EIM) Information Architecture Information as an Asset Information Governance Information Innovation Yield Curve Information Leadership Information Lifecycle Management (ILM) Information Management Information Trust Tags: