BMC report examines DataOps practices

Systematic data management investment and effort is associated with outsized returns on data-driven initiatives, according to a newly released report on DataOps from BMC Software. Further, large amounts of organizational data frequently remain underutilized for insights due to challenges with active data management, the report states.

Released July 24, the report, titled “Putting the ‘Ops’ in DataOps: Success factors for operationalizing data,” was based on research examining organizational data management and DataOps practices. The report found there is no “one-size-fits-all approach” to data-driven practices, and that variables such as business size, geography, data management maturity, DataOps maturity, centralization or decentralization of data management and data delivery functions, and strategy to incorporate AI and machine learning into data management all influence how an organization chooses to refine its technology and processes in data-related pursuits.

Generally, both higher data management maturity and higher DataOps maturity tend to link to higher reported rates of success or adoption in other data-driven activities. For example, 75% of organizations with more mature DataOps practices report having a chief data officer, versus only 54% of those with less-mature DataOps practices. BMC defines DataOps as the application of more agile and automated approaches to data management to support data-driven business outcomes.

The BMC report also found that smaller organizations, with 5,000 employees or fewer, as well as those less mature in their DataOps practices, are more likely to report a pure DIY methodology where they build all relevant AI models in-house to support data management initiatives. Larger organizations and those that are more mature in DataOps practices are more likely to report a blended approach where they leverage in-house developed models alongside commercially purchased, AI-enabled technology.

BMC commissioned 451 Research to conduct the survey in late-2023, sourcing insights from 1,100 IT, data, and business professionals from large enterprises in global regions across industries in 11 countries. Other findings in the “Putting the ‘Ops’ in DataOps” report:

Larger organizations and European organizations lead in active data management to support adoption of emergent technologies, such as generative AI.

DataOps responsibilities are typically distributed.

Large proportions of organizational data frequently remain underutilized for insights, often due to challenges with ongoing or active data management.

The most obstructive challenges in the initiative to continually provide high-quality data for consumption include data quality, deployment and pipeline orchestration.

A lack of automation, stemming from both technical and cultural challenges, is a common pain point in the effort to consistently deliver data to relevant stakeholders.

Prescriptive and predictive analytics initiatives are driving enterprise data consumption today and are projected to increase.