Creating an Analytics Culture, Part III: Sideways-In Traits

Read - Creating an Analytics Culture, Part I: 12 Characteristics

Read - Creating an Analytics Culture, Part II: Top-Down and Bottom-Up Traits

A data analytics culture is hard to pin down, and even harder to create. It’s all about attitudes and actions, things that are hard to teach, but easy to learn. Culture, after all, is the unwritten rules of how things get done. Humans excel at absorbing those rules, often subconsciously, until they become habitual ways of thinking and acting.

Most data leaders struggle to define a great data analytics culture, but they know it when they see it. This series attempts tries to describe the ineffable—to pin the proverbial jello of analytics culture to the wall. The first article describes 12 characteristics of an analytics culture—the material evidence that an organization has adopted data analytics as a way of doing business. (See figure 1.)

The second and previous article in this series described six of those characteristics, three from a “top-down” or executive perspective and three from a “bottom-up” or business user perspective. This article finishes the job, describing the final six characteristics, the “sideways in” traits. Three of those characteristics describe departmental behavior indicative of an analytics culture, and three describe how an enterprise data team fosters an analytics culture.

Figure 1. Twelve Characteristics of an Analytics Culture

Business Unit Traits 

Business units and departments are both the first and last places where an analytics culture takes root. Some departments, such as finance, sales, and operations, are very data-driven and proliferate an analytics culture, while other departments, such as human resources and legal, may lag behind, seeing no compelling need or desire to analyze data. In either case, an analytics culture requires the active participation of every business unit or department in data-related initiatives.

Analytics Council. One sign of an analytics culture is the existence of an Analytics Council. This is a cross-functional committee of representatives from every business unit and department that uses data. The council functions like a board of directors, overseeing the data analytics function inside an organization. It reviews the data strategy created by the data team and approves budgets and funding. Most importantly, it works with the data team to define standards, policies, and initiatives that guide the implementation of the data strategy. These range from training and support to project prioritization and data governance.

The council typically contains a working committee that is the “workhorse” of the council. It works closely with the data team to set tool standards, prioritize projects, govern data and reports, create training and support programs, and evangelize data analytics throughout the enterprise. There is also an executive committee comprised of senior leaders who approve the strategy and funding and resolve any conflicts escalated by the working committee.

An Analytics Council ensures that data analytics is not siloed in one or two areas, but is an enterprise program. By its very nature, the council requires its members to adopt an enterprise view of data analytics, rather than a parochial one. By meeting regularly, members learn that it’s more efficient and effective to join forces rather than build data silos.

Governance. A major responsibility of the Analytics Council is to manage risk and avoid bad things from happening with data. This includes breaches of data security and privacy; lack of data quality and consistency that erodes business users’ trust in data; and lack of standards for key metrics and other data elements. To do this, organizations with a strong analytics culture create an enterprise data governance program to govern important data assets in operational and analytical systems throughout the organization. 

A data governance program defines standards for key data elements. It creates precise, unambiguous definitions of these data elements and documents them in a business glossary. It also defines policies and procedures to create and update those definitions, track data quality, detect problems, and resolve conflicts. The program also assigns owners and stewards to critical data elements to ensure the accuracy and security of data elements under their purview and escalate problems to the appropriate committees for review.

Data governance is not sexy—it’s arduous, time-consuming committee work. But organizations with a strong analytics culture embrace these tasks with gusto, knowing the downsides if they don’t.

Collaboration. Another hallmark of an analytics culture is when people feel free to share information with each other and across departmental boundaries. Collaboration is the inverse of data hoarding and data siloing, phenomena that afflict many organizations. Ironically, data sharing requires strong data governance to ensure that people can’t access data they aren’t authorized to see. Organizations without strong governance can’t easily collaborate or share data.

Collaboration also involves connecting data analysts and data scientists to each other across the enterprise. An online collaboration tool enables data analysts to share code and reuse workflows so they aren’t continually reinventing the wheel. Communities of interest (CoI) enables analysts to meet monthly, in person or virtually to share tips, tricks, and techniques. The CoI may also extend to periodic events, such as hackathons or internal trade shows, so data analysts can hone and demonstrate their skills.

Enterprise Data Team 

Finally, an analytics culture is characterized by an enterprise data team that balances data access and data governance. On one hand, the data team needs to foster self-service and data sharing, while on the other, they need to maintain strict data access and security controls, high degrees of data quality and consistency, and strong levels of systems reliability, availability, and scalability. It’s not an easy role!

Center of Excellence. The first task of any data team is to foster alignment with business units that need data to run their business. To do that, data teams need to work in concert with an Analytics Council to deliver enterprise services. They need to manage efficient project intake and prioritization processes to ensure rapid delivery of new functionality to the business. They also need to support governance processes that certify and operationalize new insights to ensure the delivery of trustworthy data. And finally, they need to coach, mentor, manage, and support embedded data analysts and scientists to ensure they deliver value for their department without creating data silos.

Data Platform. The major lever the data team has to create an analytics culture is the platform it deploys to manage and store data and support end-user queries. Whether it’s a data warehouse or data lake or some hybrid of the two, a data platform needs multiple zones of data processing to provide universal access without undermining security. In an analytics culture, the data platform is an enterprise resource that provides departmental views of data to support localized self-service. It’s also scalable, reliable, available, and resilient.

Data Agility. The data team fosters an analytics culture by delivering valuable business solutions quickly. It uses agile development practices to work iteratively with business stakeholders and applies DataOps techniques to speed cycle times, reduce data defects, and lower costs. It commits to a spirit of continuous improvement, and continuously measures the impact of reengineering its internal processes on customer satisfaction. The goal is to build solutions faster, better, and cheaper.


An analytics culture is defined by the attitudes and actions of employees. Although it’s hard to create and measure, we believe there are 12 characteristics that correlate with an analytics culture. These can be grouped in several categories: top-down (executive leadership), bottom-up (knowledge worker behavior), and sideways-in (departments, including the data team.)

Creating an analytics culture is a journey that takes a long time. Organizations need patience and commitment to see positive results. The 12 characteristics can serve as guideposts to steer organizations in the right direction and determine whether they are making progress.

Wayne Eckerson

Wayne Eckerson is an internationally recognized thought leader in the business intelligence and analytics field. He is a sought-after consultant and noted speaker who thinks critically, writes clearly and presents...

More About Wayne Eckerson