Data management: guardrails or guidelines?

Paolo Platter at Agile Lab and Witboost argues that enterprises need data guardrails, not guidelines, to make data-driven decisions

Data is the lifeblood of modern business, with organisations everywhere now claiming to be ‘data driven’ in an attempt to deliver everything from better customer service and competitive advantage to bottom line performance.

However, the way data is managed is not always conducive to organisational effectiveness, particularly for large enterprises that have been producing, collecting and storing huge amounts of data for decades.

The potential benefits of taking a deep dive into large data sets to make better decisions are well known. The problem is that businesses struggle to align objectives with strategy to deliver the kind of insight they are looking for. Whether they lack the expertise, technology infrastructure or process discipline to meet their performance objectives and governance requirements, many continue to collect as much data as possible in the hope that, at some point, they can unlock its latent value.

In particular, the lack of a coherent data management strategy can easily result in sprawling data lakes and little in the way of actionable business intelligence or insight.

Yet, businesses are investing heavily, with the global enterprise data management market pushing through the $100 billion barrier as the opportunities created by data intensive digital transformation and AI strategies steer leadership thinking.

Addressing this situation is not easy, with a range of challenges often emerging that prevent enterprises from capitalising on what should be a major business asset. These include:

Governance

The typical large enterprise will have a massive data footprint, and to make better business decisions, leaders need to know what insights their data can reveal. Key to this process is good governance, the role of which is to ensure data can be trusted, is secure and available.

Clearly, this is a complex process that relies on the design and implementation of effective policies and standards. It comes as no surprise, therefore, that more than 90 percent of all data governance initiatives fail, according to Gartner.

Bridging the gap between data governance objectives and effective implementation means organisations must go beyond the use of standardised processes. Instead, using data management ‘guardrails’ that define boundaries and decision-making processes can help deliver data management consistency and integrity.

In contrast to guidelines that can be ignored, guardrails represent an enforceable approach that operates throughout the lifecycle of each data project, including the effective use of documentation, compliance checks and proper data protection measures. By adopting this approach, organisations can ensure they meet their own internal standards, as well as those set out by regulations such as GDPR.

Control

Given the scale and rate of growth of modern datasets, it’s simply not practical to keep it hosted in a single place. When different teams and departments are also buying their own tools and working in silos, keeping data strategy under control becomes an onerous task.

Without appropriate control guardrails, individual teams might also feel it’s more effective to develop their own approach to data management in an attempt to meet their specific requirements. The problem here is that the lack of standardisation and uniformity significantly increases the risk of cross-team errors, compliance and security breaches.

Performance

With the volume of data collected, stored and managed by businesses growing at an exponential rate, any lack of control can lead to major performance problems. For example, teams may spend excessive amounts of time locating and understanding data. This, in turn, slows down decision-making and diverts resources from core business activities. 

Similarly, the need to integrate data policies across disparate technology tools can put the brakes on any effort a company makes to create a coherent approach.

Ideally, data management systems should be both customisable to meet specific business needs and scale to accommodate growing data volumes and complexity. This kind of adaptability ensures that the data strategy can evolve with the organisation, support larger and more intricate environments and deliver the kind of performance organisational leaders are looking for.

With the right guardrails in place, organisations can overcome the barriers to success that have become a common feature of modern data strategies. In doing so, they can balance governance, control and performance so that data becomes a genuine business enabler.

For organisations that deliver on these objectives, using data as a transformative agent of change will move from a distant objective to an everyday operational activity that drives innovation and success.

Paolo Platter is CTO & Co-Founder at Agile Lab | Product Manager on Witboost

Main image courtesy of iStockPhoto.com and PeopleImages