Aligning data strategy and digital transformation

It’s a conversation we have with some of the most strategic executives we work with. The organization’s priority is enterprise digital transformation, but their data landscape isn’t fit for purpose and lacks the right foundation to leverage data and AI to accomplish their business objectives. Faced with needing to realize value from their investment, what takes priority – digital or data?

Our perspective is to stop approaching data/AI and digital strategy as separate, independent corporate imperatives and instead, think about them holistically. Each one’s ability to deliver value is highly dependent on the other. Data and AI programs need digital processes enabled by enterprise-wide technologies, and the computing power and ability to scale enabled by cloud to realize their value.

To that end, there are three key points to get right to align your data strategy and digital transformation to recognize value from your investments:

1. Connect data to business value
2. Create reusable data assets
3. Architect a scalable data platform

If you’re not sure what data has value… you’re not alone. In a recent Accenture research report, 190 data and AI executives rated the perceived value of their data.

It can be hard to break from the idea that all data could make a potential contribution to better insight. Massive data proliferation has almost guaranteed that some of it is not currently needed to address a specific business challenge – "data noise" if you will. We encourage data and analytics leaders to ask themselves, "I have X problem; what data is needed (existing or needs to be acquired) to build the assets and capabilities to solve it?" The focus can then be on identifying those critical data elements and how you articulate their value for the organization.

If you’re not sure what data has value… you’re not alone. Accenture recently published a research report in which 190 data and AI executives rated the perceived value of their data. We found that only 32% of them believed they could realize tangible and measurable value. Why is that? In my experience, it is most often because there is not a clear bridge connecting data to the business strategy. Alignment across the business on objectives that can be supported by data and analytics will drive more meaningful conversations and more measurable value.

You may even start to realize that there are common objectives or similar needs in multiple parts of the enterprise. Focusing on use cases where multiple problems can be addressed through insights from similar data sets allows you to start thinking about creating data assets that can support a set of priority use cases - which is the next point of the approach.

Understanding the critical data elements needed to solve multiple business issues is an essential first step in establishing reusable data assets. If this isn’t done, it is all too easy for an organization to end up with multiple parallel data supply chains driven by the different business functions. This leads to siloed reporting and analytics, but also inefficient processes and duplication of development effort.

Instead we encourage a focus on developing data assets that are reusable to accelerate speed and scalability for creation of multiple insights from the same data. Think of this as connecting and optimizing your data supply chain. When you have reusable data assets, you are creating an enterprise-level source of truth to underpin multiple analytics use cases.

Reusable data assets in action

Suppose a retail organization wants to reduce excess inventory by optimizing their supply chain. To tackle this business challenge, they leverage a predictive inventory management model that uses data on sales, by product type, customer segment and retail location to generate recommendations on how to reduce overstock. The model can generate new insights by combining first- and third-party data – from historical purchase records and stockout and demand data to external data like population growth metrics and business development trends for a 15-mile radius. As a result, the retailer can now better manage inventory for each specific location, but the use of this data doesn’t need to stop there. The organization can use the same dataset to feed into a model to optimize pricing strategy for their Commercial function. What could have been a dataset just used by the Supply Chain function for inventory management has become a data asset for multiple process improvements across the value chain.

This "one for all" (or "one for many") approach to data assets is the key to unlocking incremental-but-significant value from data at speed. A data strategy in tandem with digital transformation is critical here - ensuring the tooling and processes are in place to facilitate cross-functional benefit.

Often, clients approach us because their current data landscape isn’t providing what they need and they’re struggling to create new capabilities on existing in-house platforms. Even if one part of the business is advancing or successfully launching new capabilities, it can be a struggle to share that success and apply it to other areas. What is our advice? Create a single conceptual platform with the capability to scale over time. It is likely that Cloud will be a key part of the platform approach for most organizations - for the typical flexibility to scale up & down, but also for the native Analytics and AI services which can be leveraged for end-to-end. Designing the platform with an automation lens is critical, because it will simplify and accelerate the ingestion of data which can then be transformed into data assets.

The right underlying architecture that can support internal and external systems— particularly those in the cloud – can offer next-gen capabilities and layers to enable data and AI that can solve more widescale challenges. In this case, you’re building an ecosystem where data from various platforms, applications and users are assembled to leverage models and tools that bring cohesion to customer and other enterprise data. If your organization is just getting started on its journey or has legacy systems that can no longer be retrofitted to your needs (or even if you have frequently changing data), now’s a good time to investigate building one enterprise-wide architecture as part of your corporate priorities — instead of as a separate initiative altogether.

nvestments that thrive under high usage and have incremental or secondary benefits with new data should be prioritized now. The three points we outlined above are critical, but the fact is, data and AI literacy needs to be improved across the workforce for any type of data or digital transformation to deliver real business value. Executives outside of IT must embrace both major technological decisions and more routine checkpoints to stay the course of becoming a data-driven business. This includes formalizing and professionalizing AI as a trade with a shared set of norms and principles to be poised to achieve more value from AI.

A Chief Data Officer can evangelize the projects, but the ways in which data is being used across business units must be part of strategic planning and business outcome alignment, which helps improve governance and sets a clearer path to prioritization of investments. From cost optimization to modernization, knowing who needs to be involved and when, is a critical part of building a digital business. It’s an all-hands-on-deck mentality until ownership crystallizes and new behaviors become systematic.

We believe that organizations will only realize the full value of their data when digital and data/AI strategies are approached in tandem, and most importantly, when they’re anchored in the business strategy. Any technology transformation effort should be accompanied by the requisite evolution of data strategy, business models, processes, and culture. The resulting data-driven mindset across disciplines will fuel new ways of working and innovation, which when integrated, can help set you on the path to total data-driven reinvention.