Skip to main content
Exploring what’s next in tech – Insights, information, and ideas for today’s IT and business leaders

6 steps to creating a data-driven culture

Your organization runs on data. It runs best when you get the most out of your data.


As organizations seek new ways to use data for strategic advantage, many are investing heavily in tools to collect, store, analyze, and share it more effectively. With so much data being consumed, such technology is needed to turn it into real, tangible insights.

But while tech plays a critical role in optimizing their use of data, organizations are finding that their data initiatives' success rests just as squarely on another less discussed factor: creating data-driven cultures.

In NewVantage Partners' latest executive survey, just 24 percent of respondents described their organizations as being truly data driven. And the overwhelming majority said the biggest barrier preventing them from achieving this goal was not lack of technology (8 percent) but rather their inability to harness nontechnical issues—organizations' people, business processes, and culture (92 percent).

Culture is on every board's agenda. It's become a key part of what should be considered a new data-first operating model designed to modernize the way an organization thinks and acts, using data as its guide.

But what are the key elements of data-driven culture? And what can the C-suite do to bring these cultural elements to life? Here are six moves executives should make to set their organizations on a—choose your term—data-driven, data-first, or data-inspired path.

1. Promote the concept of sharing

Historically, control of data has been seen as a source of power within an enterprise. Salespeople guard customer information closely, concerned that marketing will pester their account reps with requests for promotional activities. Business units competing against each other for internal bonuses keep insights in silos that could benefit the organization as a whole.

Please read: What it means to think data-first

The idea that "if people share their information, they lose power" needs to change. Data isn't a zero-sum game. It's owned by the whole organization, not by individual fiefdoms. Nobody should worry that if they share data to help a separate team, they may lose their job.

The solution is to set up clear rules around information distribution. These rules could involve allowing certain customer data to be shared and limiting the number of reach-outs marketing can make. In addition, create dataspaces where data producers connect with data consumers with mutual benefits to gain broader discovery and access, enhanced sharing and collaboration, and improved governance and trust.

Make the sharing of data an objective that teams are measured against—the more effective sharing, the more the goal is achieved. Data is valuable. Breaking down silos and democratizing the use of data benefits everybody.

2. Use data to unite teams around common goals

Teams often maintain separate datasets that serve their own specific needs. Marketing amasses vast files of customer satisfaction stats while the customer service team maintains records dealing with on-time product delivery. What if these two datasets were paired? What if the data science team ran the numbers through advanced predictive analytics programs? Both teams could share notes and learn more about the features customers value most, which products create the most customer satisfaction when delivered promptly, and which delivery practices will keep them clamoring for more.

Insights generated by data analytics can help multiple teams improve their daily work while increasing overall efficiency and effectiveness across different departments. This linking of systems and data is important. However, it's just as important to connect the teams themselves. Setting up channels (on Slack, Teams, Google Hangouts, etc.) to communicate and hold mutual brainstorming sessions to iterate on common goals is a good start. Identifying mutually beneficial data sources and putting them to work helps the enterprise create real value.

3. Lead by example

To create real data-driven cultures, CEOs can't just declare a strategy and retreat to the corner office. They have to embrace the concept. They have to walk the walk. When they're talking strategy in meetings, they should frame issues in terms of how the organization will use metrics and data to measure performance, and they should bring their own data points to discussions. While they can always leave room for gut analysis, CEOs should try to incorporate measurement in most strategic decisions. Leaders should frequently ask, "Do you have data to support your point?" or "What KPI can we attach to this project to assess its progress?"

Please read: Getting the most from your data-driven transformation: 10 key principles

CEOs need to set a tone and push the message all the way down the organization, making it clear that this is how the enterprise will use data. Furthermore, the organization can showcase leaders who visibly use analytics and AI to make data the foundation for everyday projects.

4. Eliminate fear of failure

Since measurement is important, organizations should set goals for how much data projects will impact specific KPIs. But if a project doesn't meet a goal, it shouldn't be considered a failure. It could be that the projection was off or that the initiative revealed a problem with corporate strategy that needed to be solved.

No one likes to miss a target. So, rather than set up data projects as pass-fail mechanisms, leaders should change the vocabulary that's being used internally. Data projects should be conducted as experimentsproofs of conceptbetas, and trials, with multiple projects running parallel to one another. Each should embrace the larger goal of learning which data, or which combinations of data, have the most positive impact on revenue, customer satisfaction, or customer growth. That way "failure" gets replaced by learning exercises that will lead to company success. This will drive a culture of active growth from learning, rather than one of lack of action due to fear of failing.

5. Speak a common language

For a data-driven culture to function, people need to understand each other. There has to be a basic understanding of terms when people with different levels of data-related expertise talk to one another. And technical and nontechnical people need to understand each other's expectations about how data contributes to a particular initiative.

Most companies have a problem with basic terminology. Business leaders aren't always conversant with terms like clean datastale data, and model drift. If data scientists misinterpret business leaders' instructions, they may create a less accurate model. If business leaders don't understand the technical challenges a project might present, they may make incorrect time, effort, and revenue assumptions.

Business leaders and technical experts need to get on the same page. One way to do this is to create a data glossary—a list of terms and their definitions to ensure the same definitions are used company-wide when analyzing data. Organizations should educate their workforces on data and its corresponding terminology, including how it's used, how it works, and how it can benefit the bottom line.

6. Incentivize data-driven practices

It's not enough to just encourage widespread use of data in all aspects of the organization. Leaders should incentivize its use. One way is to hand out faster promotions and salary increases to those who make effective use of data and analytics. If data activists are getting rewarded, others will notice. Leaders will need to endorse the approach and enlist human resources' help in carrying it out. The reward system could extend to recognize when employees and teams successfully use data to improve company processes, customer engagement, product quality, and other innovations.

Please read: The key to understanding data? Seeing it properly

To establish consistency, organizations can draw up a list of key outcomes that the whole enterprise can tackle. Then they can break them down into smaller outcomes by department or division.

Culture creates a foundation for success

Technologies have advanced significantly over the past decade to enable organizations to make more strategic use of data, and tech will continue to evolve. To get the most value out of data, however, organizations need to build out two sets of infrastructures: the platforms that move and use data and the cultures that sustain its use. Creating an active, supportive, data-driven culture is essential to a company's success.

This article/content was written by the individual writer identified and does not necessarily reflect the view of Hewlett Packard Enterprise Company.