We work with data all the time, without fully realizing that wrong information, incomplete information, information of uncertain source, or of uncertain meaning, information taken out of context, or otherwise misinterpreted – can substantially subvert our ability to reason and make the right decisions.
For a business enterprise, the problem is compounded by a variety of factors, such as--increasing amount of structured and unstructured data we can access, its internal redundancies and deltas, its necessary (and sometimes unnecessary) duplication and spread; increase in reliance of decision-makers on the metadata we wrap around actual information to represent its provenance and indicate its meaning; increase real need to use data as inputs for decision-making, and then again, as metrics, to judge if decisions were sound. Everyone, in other words, is faced with too much data to check, but with the need to filter and act on that data effectively.
Even though most business leaders understand the need for high-quality data, they are often not sure how to achieve it. And yet, it has become clear that an investment in the infrastructure must be made to ensure measurably acceptable data quality.
Information Governance – which is about building a concrete methodology for discovering, combing out, labeling and schematizing, optimizing, protecting, permitting, and forbidding access to data – in short, everything you need to be sure that information and its use are being handled in sound ways, under control of enterprise policy.
Most enterprises which are heavily IT-dependent, are facing some very serious issues concerning the management and governance of information, effectively managing the risks inherent to information management and complying with regulations at the same time. They realize that having a clear understanding of customers, partners and suppliers can mean the difference between growing a business and failing to compete. Critical initiatives for information governance, compliance and master data integration simply will not succeed unless the quality of the data in systems is clearly understood and actively managed.