Whenever we talk about data, data quality, or data governance, the topic of metadata invariably is integral to the conversation. The rules that define what we mean by data quality as well as the results of profiling the data and analyzing the output are all metadata. While data governance governs decision rights and procedures (among other things) for getting to managed data, the things we want to know about our data, and how those things are connected together into an information model are ALL metadata. And yet, the quality of the metadata is rarely discussed or practiced. But poor quality metadata – from incomplete definitions to missing values for information security classification to incorrect data types to inaccurate data quality rules – can have a catastrophic effect on the understanding and use of the data it describes. Improving the quality of your metadata is thus a worthwhile endeavor, perhaps as much as improving the quality of your data.

You will learn:

• How to think about metadata as data  
• The dimensions of metadata quality and samples of rules 
• The impact of poor metadata quality 
• Building the information model to specify metadata quality rules 
• “Profiling” metadata quality: Some art, some science  
• Recording and analyzing the results of metadata profiling
• Using data governance tools to improve metadata quality