American Airlines’ loyalty program is worth an eye-popping $31.5 billion — almost three times more than the market value of the company — according to third-party appraisers. Not bad for what amounts to a database of names and mileage points. It’s no wonder people refer to data as the world’s most valuable resource. 

Most organizations realize data is a strategic asset. Very few know how to manage it with the same level of care they use for other financial and physical assets, such as real estate, machinery, inventory and more. 

That’s a mistake. From new privacy regulations (GDPR, HIPAA, CCPA) turning up the heat around the world to triple-digit increases in cyberattacks, data is under increased scrutiny if not downright attack. To protect it, data governance is an essential practice for organizations of every size in every sector. 

A data governance program is essential for protecting sensitive data and adhering to privacy regulations. The right approach can also improve how people use, analyze and share data across the company, increasing accountability as well as improving results. A TDWI study found that 36% of data leaders view data governance as a key priority to improving an organization’s success with business intelligence (BI) and analytics. 

In this blog, we dive into three ways in which organizations can employ data governance best practices to add business value beyond compliance.

Improving discovery with metadata management

Data intelligence is about transforming data into knowledge — combining data with business and operational context. This context can be captured via active metadata: data that defines other data, including data about the actions and users of data.

The improved context of metadata powers data discovery, data quality, data lineage impact analysis, and detection of personally identifiable information (PII). To realize those benefits, organizations need to employ metadata management to generate a common understanding of their data and how to leverage it effectively to drive business value. 

For example, the US Department of the Interior, which has 90,000 employees across 10 bureaus, developed an enterprise-wide metadata management program. They now use it to facilitate discovery, access and use of agency data so that it becomes a sharable resource for users at every organizational level. 

Data cataloging solutions such as Collibra and Alation can help enterprises move towards data intelligence by linking business and technical metadata in one place and providing the required context for all data users. Implementing the right metadata management solution is critical to derive business value as data volumes and breadth of connectors grow.

Fueling effective advanced analytics

It’s true that business value can still be derived from some kinds of data analysis even if there are data integrity issues. This is because, for the most part, there is a level of human interpretation involved in any analysis. In other words, even though you may have undesirable data quality issues, it may still be possible for a human to identify them and remove errors from the results. However, human intervention becomes very challenging in the enterprise as the volume of data grows at an unprecedented rate. 

In order to derive value from more advanced analytics, having well-managed data is not an option. For instance, machine learning-based predictive models learn and adapt without following explicit instructions. If you have underlying issues in your data, such models will only magnify them.

For example, consider a sophisticated set of machine learning algorithms used to analyze medical records and suggest possible cancer treatments. If the data in those medical records is not consistent or reliable, the system may have difficulty delivering useful and reliable recommendations. 

In fact, that’s exactly what happened with IBM’s Watson for Oncology. It turned out that a major barrier to getting successful recommendations was the inconsistency and messiness of the source data: acronyms, human errors, medical shorthand, and different styles of writing used by different doctors all presented an enormous data parsing challenge. The result: While Watson can beat a grandmaster at chess, its cancer recommendations have been consistently underwhelming.

Poor data management is a major reason that 87% of data science models do not make it into production. These models do not prove their business value sufficiently because they’ve been hamstrung by poor management of the data underlying them. 

This low return on investment from their data science models has pushed organizations to start managing their data better. That way, they stand a chance of realizing more value from the significant investment they have made. Consider that worldwide spending on advanced analytics is forecast to be $215.7 billion this year, an increase of 10.1% over 2020. It stands to reason that companies spending so much on analytics will want to ensure that the data going into it is well managed.

Data governance as a competitive advantage

Historically, most companies viewed data governance as a framework for enforcing compliance. This command-and-control approach to governance created barriers between people and data.

Today, leaders are realizing that the right data governance strategy is about more than compliance — it can give them a competitive advantage in the market. By making compliant data easier to access, leaders can ensure that their teams can make faster decisions and drive operational efficiency at scale. 

Companies like Apple are starting to use personal data privacy as a unique selling proposition. In its iOS 15 release, Apple introduced a number of features which protect users’ data and prevent third-party applications from tracking user interactions. Apple CEO Tim Cook has famously said that “we shouldn’t ask our customers to make a tradeoff between privacy and security.”

Apple’s move represents the shift from a compliance mindset to a value mindset. Previously, the primary impact of having poor data governance was having to pay fines. Now it could also result in your customers moving to a competitor. 

Given Apple’s dominant position in the technology market, these commitments to data privacy carry weight. They signal a change from data governance being considered as an optional “bonus” to a critical feature for modern data stacks. 

In other words, companies need compliant data to be accessible across the organization — while aggressively protecting customer data from internal and external bad actors. Organizations that fail to recognize the importance and value of governing data will lose out to competitors. 

Download IDC’s Technology Spotlight paper to learn more about the challenges that enterprises are facing around securing their modern data stacks and what you can do to better secure your data integration environment. Or start a free 14-day trial of Fivetran to experience a host of enterprise-grade features to meet your data security and governance needs.