Some Interesting “Alternative Facts” and Myths about Data Quality Management
A funny phrase has popped up in politics recently: “Alternative Facts”. Merriam-Webster defines “Alternative” as different from the usual or conventional and “Fact” as a piece of information presented as having objective reality. How does one get a fact that’s different from the “usual or conventional” fact?
I’m not here to get into a political fight. Maybe some people would say that “alternative facts” are really just another word for “errors”. Other people might say that if you think you know a fact, there might be something wrong with what you think you know, and other facts are actually the true ones.
We see the same thing is true with data. It can be hard to know whether you’re looking at good data, in which case you want to throw out data that contradicts it, or whether you’re looking at data that should be thrown out.
Thought leader David Loshin has worked with us to debunk 10 myths about data quality that will help you understand what you’re looking at, how to remediate it, and ultimately what facts – original or alternative – you can trust. Here are a few of them.
#1. A data quality tool is exactly that – a tool. It does not replace the human portion of what needs to be done. Assemble a program that combines good data management practices, data stewardship, and the use of tools and it will provide the greatest benefit.
#2. The data quality instrument you purchase will never make your data perfect 100 percent of the time. Data is an evolving set. There needs to be governance procedures, remediation processes, and cleansing priorities in order to make progress. Since you most likely have limited resources, it’s best to concentrate on the data which provides the best value.
#3. Data Scientists cannot manage every aspect of the data quality process. It needs to be a collaborative process whereby businesspeople as well as technical folks work hand in hand to provide an optimal outcome. IT and business users need to be encouraged to work together – to suggest otherwise would stall progress.
Check out David’s new whitepaper, Busting 10 Myths About Data Quality Management, to get a balanced view of best practices for implementing a data quality improvement program. The paper examines some common statements that can help you differentiate what you think you know, why it may be an alternative fact, or a myth, and some things to think about when planning your project.