Andy Schriever, COO and CTO at Park Street Solutions gave an excellent talk at the first KM Chicago meeting on Expanding the Shared Information Environment.
The title doesn't do justice to the work he described, but it gives a good starting place. His idea of the Shared Information Environment is that companies share their information internally (across business units, departments, executives, sales) and externally with partners and customers.
The claim made by Schriever is that companies don't do a good job of making that shared information available either internally or externally. As a result, organizations have become "brittle" in that they can no longer grow or improve because they are limited by what they know about their business. The way information is organized and designed in the business has become a bottleneck in the desire to improve the business. In the specific example used in the talk, it took the company two years to bring a new sales person up to income-generating potential because of the complicated structure that had developed over the lifetime of the business.*
Schriever spent more time building the case for what was needed, including some very familiar problems with different areas responsible for generating information and not sharing that with others who need it. Or information stored only in paper form; or locked away in the heads of subject matter experts; or deployed in several media and kept current via manual processes.
The solution that Schriever described called for stepping back from the immediate problems and looking at the business as a whole, and where the business wants to go with service and capabilities. They saw that they needed to build a centralized metadata description of the entire business in order to provide first-class business information to everyone who needs it. They would then rebuild / redeploy applications such that they understand and use this data model. They would also break down silos of information by standardizing against this model for the organization, enabling much lower costs throughout the organization associated with rework or rewriting materials that already exist.
The enterprise metadata model that he presented covered just about every aspect of the enterprise: Navigation, Product Families, Products, Digital Assets, People and Organization, Applications, Industries. Then they tied this model to applications that take advantage of the model: product catalog; quote and proposal generation; product configuration; replacement parts identification; price lists; sales performance & reporting; digital asset management; and a people finder.
What impressed me most about the work he described is the connection to the business needs and that business understood its constraints around fragmented information stores.
* It would be great to learn what the new salesperson training time is with the new shared information environment for this company.