Data Governance: The key to successful digital transformation

Businesses must be data-driven and insight-led in order to take advantage of opportunities in the interconnected economy. And a digitised business generates no shortage of data from across the enterprise. But it’s not enough to just generate or amass data. To outthink and outpace competitors, businesses need to detect transformational opportunities before anyone else, and then act proactively on intelligence.

These competencies – typified by industry disruptors such as Netflix and Uber, as well as the ‘big tech’ companies Amazon, Apple, Google and Facebook – are fundamental to providing a stellar customer experience, developing and implementing new business models and monetisation opportunities. What’s more, it doesn’t only concern initiatives that are consumer-facing. Businesses need to be cannier in how they work together with business partners and suppliers by finding new, better and more efficient ways to collaborate.

How dirty is your data?

The intrinsic value of any DX programme hinges on producing actionable insights. Businesses are eyeing advanced tools and technologies like predictive analytics, robotic process automation (RPA), artificial intelligence (AI), machine learning, the Internet of Things and Blockchain, to help improve or invent new products, services and business models. For example, almost a quarter (23%) of businesses are actively using predictive analytics, while over half (51%) are considering or evaluating tools to enable predictive capability.[2] But no matter what the headline innovation, it’s data that’s the unsung hero, whether fuelling processes behind the scenes or in front-end applications.

In the hot pursuit of actionable intelligence, what’s often overlooked is that any indicator or predictor is only as reliable as the quality and availability of the underlying data. As we become accustomed to taking analytics at face value, incomplete or inaccurate data will inevitably translate into misleading insights and poor decisions – ones that could negatively impact operations, planning, projections and the bottom line. And any high-profile DX project will almost certainly fail to deliver on its promise if the underlying data isn’t up to scratch. To illustrate, almost eight out of ten (78%) AI and machine learning projects reportedly stall due to poor data quality, while a staggering 96% have run into problems with data quality, the data labelling required to train AI, and building model confidence.[3]

Over time, these issues will be compounded. Not only do services and processes based on tech such as AI and RPA feed on data – they generate mountains of the stuff, too. This data needs to be collected, aggregated and analysed to be of any use or value. As the volume, variety and scope of data created and acquired grows exponentially, pity the business users who are expected to find, understand and trust information to do their jobs.

Left unaddressed, companies’ legacy of siloed data and systems will only entrench siloed practices, processes and cultures. This will keep them firmly stuck in a state of being process-defined, rather than data-driven, and prevent them from detecting blind spots or delivering the frictionless experiences today’s digitally-savvy customers have been conditioned to expect.

Data governance: The blueprint you’ve been looking for

While data science is a hot commodity, the foundation of DX is, in fact, data governance: the blueprint of policies, processes and stewards for the end-to-end lifecycle of data that, with enterprise data management, turns data into a shared, company-wide resource through continuous availability, usability and integrity.

Before running with ambitious DX initiatives, businesses must learn to learn to walk, by getting better at maintaining trusted, accurate and complete data until this becomes a core competence – and a competitive differentiator in itself. 

Data cannot be democratised without giving the consumers of that data an understanding of its trustworthiness and relevance to the business. That means having a firm grasp of the context, quality and business value of all available information sources – both inside and outside the organisation. Data governance initiatives must straddle silos within the business to drive collaboration between the people who actually use the data. Doing so provides a framework to provide executives with a holistic view of metrics, which empower them to make agile, insight-led decisions. The key message here is that data governance must be built into the enterprise by design. It cannot be an afterthought. 

Effective data governance can give organisations an opportunity to release data trapped in ageing legacy systems (often with the potential to retire them) and deploy cloud-based applications that can open up possibilities to innovate by providing self-service access to data and appropriate tools. It provides an increased number of workflows to be transformed by inserting contextual insight in the reach of non-technical decision makers. And when implemented within a framework of privacy, data can work to maintain and protect customer trust alongside delivering outstanding and memorable customer experiences. Ultimately, data governance promotes greater agility and speed of delivery – a central tenet of digital transformation.

As tempting as it can be to jump straight in, business leaders should be clear on the end goal of their digital transformation and treat their data like all other strategic assets: managed with the appropriate tools and governed by the appropriate policies and practices. At a time when data can determine or wreck business outcomes, how organisations manage this mission-critical asset will dictate the success of their DX initiatives.

[1] Gartner Inc., 2019 CEO Survey: The Year of Challenged Growth by Mark Raskino, April 16, 2019

[1] Dresner Advisory Services, Advanced and Predictive Analytics Market Study, 2017

[1] Dimensional Research, Artificial Intelligence and Machine Learning Projects Obstructed by Data Issues, 2019