Accelerating Your Data Culture Journey: Customer Best Practices in Data Governance

Data Governance – The Pillar No One Wants to Talk About

This blog builds on a previous blog that explored our customers’ best practices in building the data culture pillars of Search & Discovery and Data Literacy. Now, we’ll dive into what many consider the most difficult, but most critical, pillar of data culture: Data Governance.

Every organization that takes data seriously must embrace data governance. However, for many organizations, the term data governance has earned a negative reputation that must be overcome. Done right, data governance simply ensures that the right people receive the right data at the right time, so they can make data-driven decisions based on trusted information. In the words of one Alation customer, data governance is “what helps turn data into dollars.”

An earlier blog series describes in great depth how the data catalog enables data governance within organizations. But what if no one even wants to talk about it? Even the term is intimidating. It sounds like data needs to be under lock and key. But data governance has evolved. For too long, data governance was focused on restricting access to data behind high walls. Yet we know that when people want data, they’ll find their way to it.

Our most successful customers have found ways to take down those walls and integrate governance into their data culture, and thus more effectively protect their most valuable asset – data. A data platform that integrates governance into its user interface will reinforce the correct, compliant behaviors at the right time.

A People-First Approach to Data Governance

So how do you get people to embrace data governance? One pro-tip from our customers: Start by calling it something else. Susannah Barnes, formerly of American Family Insurance, says, “I like the term ‘Data Enablement’ because it describes more broadly what data governance should be providing to data users.” To Barnes, data enablement (governance!) supports data literacy, improves data engineering through automation, and empowers SMEs and natural data stewards to actively govern data.

Sharing those success stories turns doubters into believers. At American Family, the Enterprise Data Governance team worked with the data community to demonstrate how data governance provided immediate and positive business impacts to users at all levels. “The negative connotations of ‘data governance’ started to fall away,” says Barnes. “It wasn’t the formal, command-and-control governance program of the past, but it became one that users looked to when they had a data problem.” When it came time for broader policy implementation, the Data Governance team didn’t have to sell it – the value of data governance had already been demonstrated.

The value of good data lies in its usage. After implementing a people-first approach to governance, Yash Bhagde, Data Governance Manager at Oportun, is on a mission to “inspire confidence in our business stakeholders that data … is available, governed, cataloged, trustworthy, and protected.” That means business leaders can confidently use the data to drive business outcomes, enhancing the capabilities and competitiveness of the business.

As our customers build on this critical pillar, most note that governance can no longer be a top-down, imposed proposition. “The term governance smacks of edicts handed down from above and dusty documents that no one reads,” says Sebastian Kaus, Data Governance Lead at Vattenfall. “But a thriving data culture must be grown from the bottom up. And that starts with listening to people.” At Vattenfall, there is no “governance program.” It’s something people naturally do every day. “We’re embedding a data culture around information governance. The more people trust the data and collaborate around it, the more efficient and profitable Vattenfall will be.”

Data stewards play a critical role in data governance. When stewards are active and engaged, governance flows organically from their work, and guides others using the data catalog. Kaus notes that at Vattenfall, the stewards and catalog users become the best evangelists for data governance, which drives a stronger data culture.

Stewardship is a key pillar of governance. The new CDO of a large cloud software solutions company was “delighted to discover when I arrived that, although data governance wasn’t necessarily a formalized construct within the company, we did have people who thought of themselves as data stewards.” The unofficial stewards had formed a loosely coupled network of data experts across different functional areas of the business. That network ultimately formed the foundation of the company’s non-intrusive data governance program. The CDO concludes, “Sometimes the hardest part of getting started with a governance program is getting people to realize that they need to put some energy behind it and work together. But when you can activate that network from day one, the hardest part is behind you!”

How to Build a Successful Data Culture

Companies with a strong data culture – one where people feel confident in their ability to find, use, and trust data – are more likely to meet or exceed their revenue targets. Culture is what makes technology investments and organizational change “stick” as practices and processes become embedded in the organization.

How do our customers make culture stick? Here are some lessons from the trenches:

Where there’s no will, there’s no way.

There must be an organizational will to adopt new practices around data. Executive buy-in and a data strategy sponsor are critical to developing a data culture. And, from the executive suite to the call center, there needs to be an understanding of how data drives revenue. At Vattenfall, Kaus has developed a strong data culture. He does this by publicly tying the Vattenfall mission to data management; Kaus spreads the message that many of the company’s business challenges – such as electricity production that’s more efficient and cost-effective, or risk avoidance through better regulation compliance – can be solved by better use and sharing of data. And it’s working. Today, collaboration around well-governed data is becoming part of Vattenfall’s organizational DNA.

It’s all about people.

Technology exists to support people. And while investments that enable users to quickly find and use well-governed data are necessary, they alone are not sufficient for building a data culture. Building a data culture is all about people. Below are some examples of our customers’ best practices for investing in people to build organizational capabilities.

Invest in training.

Our successful customers invest in data literacy education for their people – at all levels of the organization. Training needs to be functional; not everyone is a data scientist. Any data literacy program must adapt to the specific business function, for example, addressing what data literacy means for the sales, finance, customer service, and other staff.

Start small and go deep.

Our most successful customers have focused their initial data transformation efforts on a single area of the business. As users discover that they can trust the data for strategic decision-making, they want to share their success with others. Momentum to become data-driven then spreads organically throughout the organization.

Spread the word.

Our customers report that their data catalog users quickly become “data evangelists” once they see how easy it is to access and use data for their purposes. Analysts that once spent the majority of their time searching for the right data can now quickly find and apply analytics. Naturally, they want to spread the word to other data scientists and analysts throughout the organization. Let them.

Train often, but don’t train too early.

Timing is everything. Make sure the tool you debut is ready for use by everyone. Introducing new technology, such as a data catalog, to the organization before it is fully functional and ready to use will backfire, driving down adoption rates instead of boosting them.

Build a community.

Successful customers constantly educate, train, and develop. After launching Alation, one customer provided training, office hours, lunch and learns, and sharing sessions about both Alation and other complementary tools. But they went beyond this by creating a community of people who are super passionate about data and data culture that meet on a regular basis to discuss how to advance that data culture across their very large enterprise. Now, three years later, they have a growing community of people excited about working with data and attending their events and training – voluntarily and regularly!

Conclusion

All companies have an organizational culture surrounding how people think about data and their role in using it. How they access data is a reflection of your governance program. Some may say, “I ask the guy who has been here the longest where to find the data I need” while others may report, “experts handle the data and give me reports when I ask for them.” An active governance program, supported by a data catalog, can effectively create a self-service BI environment at your organization, so that people say, “I am confident in my ability to find the right data for my purposes, and I can generate insights from it.”

If you remember one thing: One size doesn’t fit all. There’s no one way to build a data culture. But we hope this blog series has provided some insights into best practices, both in theory and in our customers’ best practice. The bottom line: data culture starts at the top, with a willingness to invest in and build a data culture, and a strategic focus to build on all three capability pillars: Data Search & Discovery, Data Literacy, and Data Governance.

Want more information? Be sure to explore our other data culture resources:

• “What is Data Culture and Why Do You Want One?” blog series

• “The Data on the Chief of Data” podcast with Randy Bean

• Check out our Customer Page to learn more about how our customers are building their data cultures

1. Originally quoted in Why data governance is like Fight Club.https://www.computing.co.uk/analysis/4043937/governance-fight-club