Most companies have a rich store of metadata about their employees, sales leads, customers and suppliers, yet they fail to use this data when they build portals and knowledge management systems.
Here is an example: a lot of data is collected about target customers during the sales process. Sales people know what industries their customers participate in, who the key executives are, the size of the business and what products or services they deliver. In addition, sales people frequently involve subject matter experts from the company to help configure the product or service offerings for the customer. By mining the rich store of data in the CRM system, a knowledge manager can identify the customer, the solution that the customer bought and the names of the internal subject matter experts that have the tacit knowledge about the solution.
If knowledge managers identify the stores of information in the enterprise, then they don't have to collect all that information in an exercise called knowledge harvesting. Part of the job of a clever knowledge manager and system developer is to find the existing metadata and import it into the knowledge management system so that it does not have to be reentered or worse, recreated. Metadata can be inherited from many different applications, for example, data about people and their expertise frequently resides in HR systems. From electronic resumes we can find where people went to school, the languages that they speak and identify their previous experience. All of this can become metadata for applications that help to find people with specific experience.
Frequently, executives, employees, and knowledge managers worry about the volume of incremental work created by a knowledge management system, but analysis of data sources can actually lead to reducing the amount of work in business processes and accelerating critical information flows.