It’s obvious that an enterprise data strategy involves data – but we sometimes disregard the fact that it should also involve metadata. Why? Because it’s key to unlocking the value of data. Metadata shows you what data is available and how people can use it. It also reveals which data is being used for which decisions.
It’s invaluable to have an automated system for metadata collection and analysis in any size or type of data management environment – but it’s particularly useful in highly regulated industries like banking and insurance. The demand here is created by regulations like BCBS 239, which require organizations to provide information systems that automate the governance of key data flows. These information flows need to capture metadata from the end-to-end process of information delivery.
Let’s look at what’s required for metadata integration, and then consider the benefits for users.
Imagine that you’re a new employee. You want to get to work quickly and be as effective as possible at decision making and running the business.
One of your challenges is getting up to speed with (or learning about) the large number of data sources available to you. You need to understand how data is used for decision making. In many cases, the underlying processes implemented for decision making use different methods of defining terms, vocabulary and taxonomy. They sometimes use different logic for calculating values and for extracting and massaging data. And they may use different data quality rules for manipulating data.
To complicate matters, the technologies used for ETL, business intelligence and analytic processes store metadata in different repositories, often using different methods.
A strategic look at metadata
How can you provide consistent information to both IT and business users amidst all these variables? You need to empower users so they can accurately search for and identify all available information sources, discover attributes of each information source, and perform data lineage and other analysis on the data.
As you consider how to accomplish this, you’ll probably have several questions:
- What data is available for me to use in my decision making?
- What form does the data have?
- How is it collected?
- How often/frequently is it updated?
- What processes are used to update the data?
- Who is the data owner?
- How do I compare reports that come from different systems – what data feeds the reports? What fields are used in which reports?
- When do you use one data source versus another?
One way to address these concerns is by incorporating a metadata solution as part of your data strategy. It should include:
- A business data glossary.
- Automated metadata collection and integration.
- Data lineage and metadata analysis.
- Proactive data monitoring.
Accessible metadata = data that's relevant, useful and trustworthy
With these capabilities in place, both business and IT users can access all kinds of metadata from a governed environment. They can search, define, create and edit the metadata. They’ll be able to examine relationships, use workflows to manage change and ensure granular security. Plus they’ll be able to enforce business and data rules, and set up alerts so they can be more proactive with data monitoring.
A strategic metadata solution lets business users search, locate and use the information that’s truly relevant for them. Then they can answer questions about the data, like determining what data is available, whether it’s suited for their purpose and where to go for more information.
IT can see who uses data sources and who will be affected by changes. They can analyze how information and metadata flows from data model design to database through ETL processes, BI tools and analytic environments. They can monitor data and detect changes that might be important to the information environment. Plus they can identify duplicate information and help resolve misunderstandings about the data.
Making metadata accessible is a sure-fire way to ensure your data is relevant, useful and trustworthy for everyone who needs it. With a strategic metadata solution, your organization can provide reliable information to regulators and meet objectives around regulations that require documented data flows.
Sure, any good data strategy encompasses all of the basics – all of your data, all of your processes, your people and governance. Just remember to include metadata as a core component of your overall enterprise data strategy.Download The SAS Data Governance Framework: A Blueprint for Success