Your Data Infrastructure Needs a Data Culture

By David Cochran and Joy Eakins 

We live in a time when making effective use of our data is not an option. Either we generate value from our data, or we lose a competitive advantage.

Rising to meet this imperative requires addressing two major challenges. One is building a data infrastructure: the systems, software, processes and expertise that provide access to our data and the capability to analyze it. This formidable challenge overshadows what is often the more difficult challenge: building a data culture.

Ignoring this step can be devastating. To help you avoid it, we have a few insights to offer. But let us start with the obvious question first.

What is a data culture, and why is it important? 

A data culture is composed of the daily flow of work that brings data analysts, business leaders and decision makers together to turn the data into information that makes your organization stronger and wiser.

Consider your data infrastructure like a car, and your data culture as its driver. The car may be a lightly used subcompact or a formula one race car. Whatever the car, a driver may do (or not do) many things with it. He may leave it in the garage, drive it in circles, steer it into a ditch or—much better—take the car out with purpose and direction and use it to go somewhere.

Your data infrastructure needs a healthy and purposeful data culture if it is to go somewhere. There is no magic formula for building such a culture, but we can share a few insights that may help along the way.

Recognize that building a data culture is an adaptive challenge more than a technical one

Too often we view data analytics as a technical endeavor. It is also a people endeavor and a learning endeavor.

As Ed O’Malley and Julia Fabris McBride (2022) explain, a technical challenge is relatively straightforward. Put the right experts, processes and systems in place, and it can usually be fixed. Building a data infrastructure is largely a series of technical challenges.

Building a data culture is a different kind of challenge. It cannot be “fixed” by institutional authority or technical expertise; it requires building mutual understanding among many stakeholders; it takes time, patience and a strong sense of purpose; and it benefits from curiosity, experimentation and learning over time. O’Malley and McBride call these adaptive challenges. Building a data culture is a web of adaptive challenges.

A data culture must be woven into your larger company culture. You’ll have to weave in many places at once. And if you do not approach it with purpose, you’ll wind up ensnared and stuck. Thus, we recommend you view the challenge as one that calls for multi-level cultural leadership.

Identify, consult and tap multi-level leadership 

Building a data culture requires leadership. But it cannot be single-source leadership. Data culture is a joint learning endeavor, and the learning you need cannot be limited to a few persons. Rather, it must filter through the organization. Such an endeavor, as Sean Hannah and Paul Lester (2009) have written, “requires not only powerful individuals at the top of the organization but…empowered formal and informal leaders” at every level.

John W. Rowe’s transformative leadership of Aetna in the early 2000s presents a vivid example. The fourth CEO in five years, Rowe needed to lead Aetna out of decades of stagnation and years of struggle. This required nothing less than reweaving the company culture. Rowe and his team began by selecting individuals from all levels of the organization—persons who “were well connected, sensitive to company culture and widely respected”—and listening. They gleaned important insights and recommendations from these persons. And they included them in the design and execution of the company’s transformation process (Katzenback, Steffen, & Kronley, 2012).

For Aetna, this approach proved successful. We have seen a multi-level leadership approach serve organizations well as they develop a data culture. We recommend it.

But somewhere very early in the process, we also recommend a second and more focused strategy.

Speed the learning with focused projects and quick successes 

Data analytics initiatives can easily get out of hand. They can become complex, costly, time-consuming and too often profoundly unsuccessful. One or two of those experiences, and your fledgling data culture, may find themselves limping along with a crisis of confidence.

Nurture your data culture and speed its growth by giving it quick wins to learn from. Start with a specific and focused problem. Ensure there is an ample supply of data to mine for relevant insights. Assign a crack team to tackle it—including at least one data analyst to mine the data, a business leader experienced with the nature of the problem and a decision maker with the authority to do something about it. Give the team a window of time to see what they can accomplish and have them report back.

Whatever they report, learn from it. If it was a failure, not much has been lost. Learn from it and identify ways to improve. If it was a success, document it. In either situation, celebrate those who took the risk. Then rinse and repeat. When possible, build incrementally on your early successes to gain increasing leverage.

In so doing, you will have stolen a few best practices from agile software development (Rigby, Sutherland, & Takeuchi, 2016). You will have reduced your early risks. You will have accumulated early successes—and a few failures. So long as you and your teams learn from the successes and failures, you will have greatly sped up the growth of your budding data culture.

Conclusion 

To derive value and information from your data, you will need both a data infrastructure and a data culture. Data culture is an institution-wide leadership challenge. Consider tapping people from multiple levels of your organization to help guide its growth. Speed up the learning curve by assigning focused projects to crack teams of data analysts, business leaders and decision makers. Treat your early efforts as experiments and learn from them.

Document the wins. Learn from the failures. Soon your multilevel leadership team will have processes, stories and examples ready to pave a path for others to follow, greatly reducing the friction points for those who would dig in their heels (Heath & Heath, 2010). You will have started building a healthy and vibrant data culture.

David Cochran, is the Dean of the School of Business and Technology at Newman University (NU) in Wichita, Kan. He established a networking group known as the Data Professionals of Wichita in 2018 by researching and developing a business data analytics program at NU. 

Joy Eakins, is the president of Cornerstone Data, Inc. in Wichita Kan. in 2008, she founded what would later become known as Cornerstone Data. With a multi-disciplinary background in mathematics, computer science and theological studies. Eakins was uniquely positioned to view data science in light of the people and organizations that produced it. The company was created to provide such a bridge.