How Nurturing a Data Culture Can Unlock the Potential of DataOps

A recent study by 451 Research found that more than 80% of organizations report having made progress toward implementing a DataOps strategy. It’s no surprise that DataOps efforts are widespread. According to 451 Research, the top two benefits organizations anticipate are increasing sales and enhancing customer service and engagement. In other words, DataOps is an engine for growth.

Digging deeper, however, researchers discovered that only 9% reported that DataOps is an optimized, ingrained part of their company culture. What’s holding the other 90% back? And more importantly, what can your organization do to accelerate its adoption of DataOps?

In the session Reach Your Data’s Full Potential With DataOps at Hitachi Social Innovation Forum 2021, Paige Bartley, a senior research analyst with the data AI and analytics team at 451 Research, shared some surprising answers. One of the most interesting takeaways? Organizations may be addressing their DataOps obstacles in the wrong order. They’re too focused on addressing data literacy and other organizational barriers. Bartley says organizations should instead first deal with their underlying data management hurdles. While the two go together, without the latter you may never achieve the former.

Being Definitive About DataOps

According to 451 Research, DataOps is “the alignment of people, processes, and technology to enable more agile and automated approaches to data management” that deliver data-driven outcomes for the organization. The underlying assumption is that more data equals more unique insights. Therefore, the aim is to bring together an organization’s variety of structured, unstructured, and semistructured data to reveal coherent, actionable insights.

DataOps itself is not an end. DataOps is the enabling capability a data-driven organization needs to grow. Being data-driven almost always means that a data culture permeates the entire organization, an environment where every relevant worker is empowered to be effective with the data they use. When a strong data culture exists there are minimal points of friction and few barriers in the way of data driving business value.

Identifying Barriers to Growing a Data Culture

Bartley reveals that there are two categories of barriers getting in the way of data culture and DataOps. Organizational barriers, such as a limited budget or a lack of skilled resources, are cited by respondents as among the hardest to overcome. In the 451 Research study, budget and skills were both among the top three barriers cited most frequently out of more than a dozen possible responses.

The second category of barriers is made of technical challenges. As organizations mature into advanced data-driven practices, the complications of integrating new data resources and existing IT investments become more apparent and limiting. Bringing data together is a problem that spans organizations of all sizes, from the largest enterprises down to small- and medium-sized businesses.

Bartley offers a vital observation about technical barriers: “While people may be the basis of corporate culture, data is the basis of data culture. If you don’t have very-well-managed data, that data culture is not going to progress very far because the employees, the workers, can’t access and manipulate the data that they need.”

Enabling Self-Service Through Data Management

In other words, investing in data management is the essential first step toward establishing a data culture. While data management is complex, it is what enables the seeds of the data culture to take root. The self-service business intelligence tools that are the hallmark of a data-driven organization are only as good as the data they integrate with. IT must address that need.

This means dealing with some significant hurdles. Integrating data while maintaining security is complex and tricky. Increasing regulations and changing expectations regarding personal data have set the bar very high. When an organization integrates disparate data sources, it needs to be mindful of the results. Two data sources, neither of which are sensitive on their own, can easily yield unintended inferences.

Bringing data from legacy systems will be challenging. Organizations have sizable investments in their IT infrastructure. A universal rip-and-replace approach is likely to be unacceptable, even in some cases where a clearly better solution exists. At the same time, today’s data is everywhere, spread across applications and multiple clouds, and located on and off premises. All these resources will need to work together.

Quality of data will also need to be addressed. An organization’s “data past” needs to be confronted. A history of underinvestment in data architecture and management will be exposed and certainly should not be replicated.

But confronting these challenges today, head on, will enable the self-service data culture of tomorrow. A good DataOps program will reduce the friction that hindered self-service users in the past. It will also reduce headaches for IT department staff, who were the object of the resulting frustration. A good DataOps program will be all but invisible to the end users of the data. This approach will also enable the AI capabilities that 95% of respondents said was important to each organization’s ability to scale and make use of modern volumes of data.

If it sounds like a lot to handle, that’s because it is. But it is worth it. Investments in addressing these issues will create a self-service data model that line-of-business decision-makers will engage with. Those users will be the first generation of an emergent data culture.

Spotting the Value of a Data Culture in the Wild

The returns for those who fostered a data culture were made clear during the COVID-19 pandemic. Brick-and-mortar businesses found themselves having to accelerate their moves to digital delivery of customer service and engagement. That shift required massive data collection from new sources. Organizations that prepared for the necessary data integration found opportunities. Those that didn’t fell behind.

The improved efficiency with which organizations address data privacy regulation and compliance is another example of effective DataOps. Today, regulations, including the EU’s General Data Protection Regulation (GDPR) laws and California’s Consumer Privacy Act (CCPA), require companies to deliver data to any individual who asks to see what the organization is holding. Fulfilling such requests takes time and money. Good data management practices at data-driven organizations with reliable DataOps manage, find, and deliver data much more efficiently.

When done correctly, DataOps is a catalyst for creating a data culture. The result removes the barriers between people and data technologies, creating a positive feedback loop that strengthens the entire organization.

Samta Bansal is Global Consulting Strategy and Market Leader at Hitachi Vantara.

Check out more great stories on Insights.