In brief

In brief

  • Research has shown that organizations with a strong data culture have nearly 2x the success rate and 3x the return from AI investments.
  • A significant aspect of data culture in an organization is how employees perceive and respond to AI and data solutions in the workplace.
  • By regularly understanding which behaviors exist and to what extent, organizations can measure behavioral change over time.


We are in the midst of a data-driven revolution and exploding quantities of data are affording many companies with massive transformational value. Despite that, the value enterprises generate from their data is still underwhelmingly low, with only 32% of companies being able to realize tangible and measurable value from their investments in data. This begs the question - where is the investment in these capabilities going?

Usually, the investment goes to the technology and data itself instead of the integration between data, technology and the people who use it. This can perpetuate issues like the one we saw with a materials science client, where many managers and engineers are still spending 60% of their time gathering data to keep operations running. In other words, much of their time is spent working on the data versus working with the data. What’s preventing these companies from converting new data into new value? The answer lies in data culture – the meaningful integration of data, insight, and the experience of work.

Why is Data Culture Important?

Research has shown that organizations with a strong data culture have nearly 2x the success rate and 3x the return from AI investments than companies without. A strong data culture puts people at the center of any data transformation. It ensures they’re empowered to consume and use data in ways that make sense for them, and it aligns (and even evolves) with their level of data maturity. Along with strong data and AI capabilities, this kind of culture can help businesses to think about data more strategically and use it more broadly, helping them optimize operations, understand customers better and ultimately unlock new revenue streams through new products and services. It’s a compelling proposition as companies race to improve across all performance KPIs while reducing speed to market.

So, how can leaders get data culture right?

Opportunity 1: Focus on changing behaviors rather than ‘culture’

Aristotle said, “We are what we repeatedly do”. This brings to the fore the importance of observable behavior and habits around data; they are what shapes culture. To understand these behaviors, organizations can start by capturing the experiences and interactions people have with data, mapping some of the roadblocks they encounter when making effective decisions and listening to what they find most meaningful about their work.

For instance, Accenture worked with a materials science company to map the attitudes people held towards the use of data and analytics in their decision-making process. Many harbored negative attitudes towards data due to past experiences with it, but they still understood the value of data. We also discovered that many people were tackling even the simplest problem as if it were a big innovation problem. This was happening because the company was incentivizing highly innovative behaviors, and while innovative behaviors should be rewarded, they should not exist in an innovation vacuum. Unaccompanied by some form of data governance or value prioritization, this meant the organization struggled to unlock the full value of data because there lacked a clear path to implementation.

Solution: Adopt a behavior-led framework for change

By regularly understanding which behaviors exist and to what extent, organizations can measure behavioral change over time. We call this an organization’s Data Pulse and it is meaningful for two reasons. First, organizations can use it to track the existence of a data culture in an organization at any point in time. Secondly, it provides measurable evidence of those interventions that increased data adoption and those that did not.

The Data Pulse is a survey designed with behaviors at its core. We define the behaviors we need to see in relation to the way data is treated as an asset, how it is trusted as the basis for action and how data is used in the context of driving business value. In relation to these dimensions, we ask people to articulate the prevalence of certain observable behaviors in the environment around them. For example, whether people share insights they've discovered, making them accessible for others through to the way data and insights are used to create novel and valuable solutions for an organization.

By forming a clear understanding of people’s needs, challenges and values, organizations can use these as the basic design principles on which future data solutions are based. Mapping these principles over time paints a more accurate picture of an organization’s data maturity and the behaviors displayed.

Opportunity 2: If you want people to do something, make it easy

When it comes to changing behaviors, there is a universal truth: if you want people to do something, then make it easy for them. When we examine the technologies or processes many organizations use to make data-driven decisions, more often than not, people find it hard to do the things they need to do with the tools that they have. Data solutions are usually challenging to access. For instance, there might be too much friction or sometimes an overload of information with a limited amount of time to process that information. In the same materials science organization mentioned in our previous example, the lack of data governance led to poor data management, and siloed data sources discouraged many from leaning on data to generate better ideas. All the above act as restraining forces to data adoption, as these unintended forces end up overwhelming the user.

Solution: Solution-driven Behavioral Change

People are motivated to learn and change their behavior when a changing context demands new behavior Focusing on the environment in which people make decisions is key to understanding their experience with data, which is exactly what we delivered for a financial services client. We setup a room designed to enable the team to physically move across different areas as they mentally walk-through various stages of the decision-making process. Essentially, a guided physical movement acts as a nudge for following a lean and value focused decision-making process. At the same time, the environment is underpinned by data displayed on digital dashboards. Participants interact with real-time portfolio data resulting in more engaging and data-driven discussions in the room.

Participants interact with real-time portfolio data resulting in more engaging and data-driven discussions in the room.

The need exists to shift the burden of responsibility for changing behaviors from the individual to the system in which people operate. By changing the system to demand more of the behaviors needed, the focus is shifted and therefore less burden is placed on the individual.

Opportunity 3: Make data matter for people

There is a common story when it comes to the way people use data and analytics tools in many organizations, and the short version is: they often just don’t. This is less about data literacy and more about data relevance. When it comes to improving the way data and analytics tools are being used, instead of asking what training people require to use these tools, it would be better to ask ‘what opportunities exist for data and insight to help people do something meaningful as part of their everyday experience of work?’ Simply put, embed data and AI where the work happens.

All too often, organizations train people to do the things that aren’t happening. Training aimed at upskilling the workforce in data and AI may leave many feeling ill-equipped, especially if these new tools haven’t been integrated into their day to day work. The reason these things don’t happen in the first-place boils down to the processes, structures, and experiences of work, not a lack of time in the classroom. The challenge in building a data culture within a manufacturing organization does not lie solely in the lack of data literacy – on the contrary, engineers that make up a majority of workforce in this industry are frequent data users. The challenge exists due to data siloes, which disincentivizes cross-functional collaboration. This lack of opportunity to use data in the flow of work is the greater barrier to the adoption of data and analytics tools than a lack of training.

It is also interesting to observe how people in organizations are incentivized and thanked for their contributions. Very often, feedback loops and extrinsic rewards remain relatively stable over time, whether or not they are congruent with the organizations espoused values or beliefs relating to data. In other words, people are thanked, rewarded and often promoted for patterns of behavior that have been deemed valuable over time, often for decades. If feedback loops (to help people experience the value of data in practice) don’t exist, or the behaviors for which people are recognized don’t relate to data, we shouldn’t be surprised that the new patterns of behavior we need don’t always emerge. Accenture has found this to be particularly prevalent in relation to treating data as an asset, where effectively stewarding and protecting data is experienced as a thankless, side-of-desk activity. We believe there is the need to help enterprise leaders, mangers and those with influence to enhance their ability to reward, provide feedback and encourage data behaviors as a core component of their own data literacy.

Solution: Digital Twins

A significant aspect of data culture in an organization is how employees perceive and respond to AI and data solutions in the workplace. AI might present itself as a force of disruption and displacement of human work but, it is actually expected to empower people rather than to replace them. AI should be more about supporting the way in which people work and one such solution is the digital twin, which is a virtual model that is the exact counterpart (or twin) of a physical thing. Digital twins can be used to monitor, analyze, and simulate their physical counterparts through connected sensors on the object which are mapped onto the virtual model for anyone to view crucial information about it in the real world. It can help businesses monitor the state of equipment at remote sites, optimize product designs for specific needs, and even model the business itself.

Digital twins can be used to monitor, analyze, and simulate their physical counterparts through connected sensors on the object

For instance, using AI and other technologies, manufacturing companies can monitor a piece of machinery throughout its lifecycle and receive critical alerts, such as when it is at a high risk of malfunctioning due to faulty parts, signaling a need for maintenance. At a postal service client, we worked with, we helped create a digital twin that simulates the E2E postal network. The digital twin enabled the client to test business hypotheses through measurable experiments, and track impact against KPI. It helped create a simulation baseline against which future operational changes can be tried and tested.

Leaders that adopt digital twins would enable a paradigm shift in their organization which would in turn have significant and lasting impacts on the organization, such as real-time performance optimization, fully empowered autonomous teams with trusted data, and integrated yet flexible IT/OT architecture. By adopting digital twins to integrate data into day-to-day workflows, it encourages data behaviors and strengthens the data culture by reducing the barriers to adoption of data and analytics tools.

Shaping your data culture

Data culture is a complex problem, and we need to acknowledge that complexity. When considering how to develop a data culture, start by being closer to people and their interactions, the challenges they face, and the things they care about achieving. If we can support those things more effectively with data and insight, we can increase the adoption of analytics tools and in turn improve decision making. Shaping the right data culture blurs the lines between gut feel decisions, and data-driven decisions. This is critical to the success of any data-driven transformation. Over time, new experiences and interactions will change attitudes and sustain new behaviors relating to data, and the virtuous circle of data culture can really begin.

Benjamin X.Y. Lee

Managing Director – Applied Intelligence


Kate Morgan

Senior Manager – Analytics Strategy, Applied Intelligence


Samuel Netherwood

Senior Manager, Data Culture Lead – Applied Intelligence


Nicolette Wong

Data Science Consultant – Applied Intelligence

MORE ON THIS TOPIC

How to build a data strategy to scale AI
Why you need to capitalize on the rise of the data-driven enterprise
Getting AI results by "going pro"

Subscription Center
Visit our Subscription and Preference Center Visit our Subscription and Preference Center