The big news this week in the government IT sector was that federal CIO Vivek Kundra announced he was leaving government for Harvard University. One of his initiatives was creating data.gov, whose primary goal was to improve access to federal data. As any federal IT manager knows, managing data within their own agency is difficult enough. Image trying to accomplish this across all federal agencies! Fortunately, the challenge can be met by starting with an enterprise information management strategy.
Enterprise Information Management is the collection of people, processes and technology for gathering, managing and disseminating an organization’s information assets. It is broader than managing the data warehouse / business intelligence program to encompass managing the information in the agency’s mission-critical IT systems, including both packaged application software and custom developed applications.
Unfortunately, the industry is awash in an alphabet soup of acronyms referring to tools and techniques for managing information assets. As business intelligence consultants, we advise our client to ignore the vendor-speak and look at what each technology is attempting to accomplish. The approach presented below provides a simple framework of categorizing these tools and techniques based on what you do with the data:
These techniques link or “glue” together otherwise separate applications allowing them to exchange data with each other, in effect creating one virtual application. This reduces the need to manually input the same data into separate systems and eliminates inconsistent data. An application of this technique would be to glue together your agency’s IT systems to share citizen data.
These techniques get data from the applications and use it for a specific purpose, primarily to create reports. The data still physically resides in the application systems and is not moved into a separate repository. For example, code is written to query different applications, get the relevant data and combine it into an end user report.
these techniques physically move the data from the applications systems into a separate repository, typically a data warehouse and/or data marts. The data is exacted from the application system, transformed by a defined set of business rules, and loaded into the data warehouse and subsequently into data marts for end user reporting.
These three buckets broadly describe the technology part of an information management strategy. The exact solution for your agency depends on your business needs and could contain all, some or none of these techniques. And like all technology solutions, there is always a trade-off between cost, schedule and quality. By developing an information management strategy and thoroughly analyzing the people, processes and technologies needed to implement it, you will be in a better position to manage your information assess and deliver business value.
By David McIntire, Director of Federal Consulting
© DecisionPath Consulting, 2011