We live in uncertain times. Change is accelerating for businesses, while markets are volatile, complex and often ambiguous. In this increasingly unpredictable environment, senior leaders need to make complex decisions faster every day. These decisions are usually focused on driving operational performance and less frequently on creating shared value for the organization. This will no longer be enough.
To win and be sustainable, executives need to pivot from primarily making decisions focused on improving operations (making sure nothing goes wrong) to also identifying decisions that fuel innovation, growth and sustainability (driving business value).
So, what’s getting in their way? Organizations lack the right connected and contextual insights to make these decisions. In fact, I often hear our clients say: “We have a lot of data, but no real insights.”
Before we delve into the “what” and “how” of strategic decision-making, a good starting point is to define what an insight is. Our view of an insight is learning something unexpected. It’s something you did not know or recognize before. It can take you by surprise and leave you thinking, “Oh, that’s interesting!” An insight offers you a new perspective.
Work backward from the decisions
Rethinking the mechanics of decision-making starts with decision design. It begins with identifying critical decisions linked to enterprise value and working back to what type of insights you need, and then the data you need to understand the context. This process is supported by creating a decision-making framework based on a shared understanding of value and your organization’s key business imperatives.
The next step is to deliver the right reliable insights required for the prioritized business decisions. This calls for modeling and classifying value-based decisions, identifying the insights needed, acquiring the necessary data to generate insights by creating the right machine learning models, accessing training data, removing biases, and so on.
Keep the data-insight-decisions “adaptive wheel” turning
Organizations spend a lot of time cleaning, scrubbing and structuring data to try and make a model work. More important is finding a way to deliver insights regularly, on an operational basis, rather than as a one-off activity. By continually operating a “data-insights-decisions” cycle or what we call the “adaptive wheel,” organizations can operationalize the various science, data and cloud ops, adding new decisions and data and uncovering new insights as needed.
Rethinking the mechanics of decision-making requires focusing on the “what” and the “how” of strategic decision-making. This diagram illustrates the stages in the process to pivot from focusing only on making operational decisions to making data-driven decisions that support shared value, growth and innovation.
Make the right automation choices
With the insights and the analytics in place, automation can help build and operationalize these models quickly. Some of the prioritized decisions can be automated, but it’s essential to pick the right strategy that includes a code of conduct and recognizes both human impact and ethical aspects.
The good news is that organizations have been making massive inroads in automation. While most of this progress has primarily been around simple automation, we see clients beginning to invest more in algorithmic, enhanced automation and AI and explore the connection between data and AI.
However, leveraging the full potential of AI and data requires an organizational pivot, as data analytics and automation teams often work separately. Assessing how to bring these capabilities together and couple them with strategic insights will ensure you focus on making the right decisions linked to a shared value framework.
Notably, organizations also are focused on collecting more contextual data, for instance, weather, location, topography, etc., to make more accurate decisions that, in many cases, can have far-reaching economic, environmental and societal impacts. For example, we have partnered with the University of Louisiana at Lafayette (ULL) on a machine learning-enabled flood forecasting prototype project. The solution can accurately predict soil moisture levels using a deep learning model that analyzes variables such as precipitation, temperature, vegetation, plant canopy, surface water, and incoming long- and short-wave radiation to predict the likelihood of flooding.
Visualize decisions before acting on them
To truly exploit digital technologies to create shared value and minimize enterprise risks you need to visualize your entire enterprise digitally and simulate decisions before acting on them in the real world. Building a digital twin of your organization allows you to visualize and optimize your decisions, insights and automation choices and their impact on your business.
Decision-making often is complex and non-linear; it is part science, part art. We believe decisions, once “codified” and modeled, will soon be treated, managed and evolved as an enterprise asset class, similar to processes, critical data, business imperatives, value drivers and AI models.
I invite you to read our white paper Value in motion: Driving forward data-driven decision-making on how to ensure insights and enterprise decision-making create shared value for your organization. Reach out to me if you’d like to discuss this subject more.