Home » Technical Topics » Data Science

Transforming Day-to-Day Decisions in the Enterprise

6932733899My views on how to effectively align daily decisions to business objectives

At first glance, the odds of winning at rock-paper-scissors is one in three… right?

Well, if completely left to chance, then the answer is a simple yes. But people do not play randomly, according to a study at Cornell University, they follow hidden trends.

In a similar fashion, people tend to follow trends when making day-to-day decisions while performing their roles. These trends can introduce inconsistencies and cause a deviation from the organization’s high-level objectives. To mitigate the risk of going off course, a well-implemented workflow can go a long way to automating or optimizing those decisions. Better yet, implementing such a workflow doesn’t need to be costly or complex.

A lot of progress has been made with the automation of decisioning, driven by enterprise analytics, through digital transformation initiatives of varying scale (and varying success). But not all decisions are suited to full automation, and some will always require having a ‘human in the loop’. In a recent interview with SAS, Iwona and I explored the increased importance and some of the challenges with enterprise digital transformation. We focused specifically on the use of business-driven apps as a way to both scale the adoption of enterprise analytics and enable innovation through rapid app deployment.

The Last Mile

McKinsey refers to the practical consumption of analytic output as “the last mile of analytics”, and the article speaks to how often an organization’s biggest challenges are turning insights into outcomes. We believe that using apps is a great way to conquer the last mile by guiding decisions through well thought out workflows with modern, intuitive interfaces.

Decisions are where it’s at

Referring to a major US retailer, McKinsey notes that “The company had to overcome many challenges, including the toughest of all — bridging the ‘last mile’, or delivering the right insights to the right people at the right time in a way that informs their decision making to drive better business outcomes”. This made total sense to me, and it aligns with Boemska’s approach to helping customers make effective and consistent decisions. The article ignited a desire in me for a deeper understanding of how decisions are made and why people make them. Specifically, in the context of analytics, how Boemska can help ensure that strategic decisions made at the Executive level, cascades through to the thousands (if not millions) of daily micro-decisions.

The discipline of Decision Intelligence

My curiosity led me to the wonderfully quirky and refreshingly effective communicator Cassie Kozyrkov, Chief Decision Scientist at Google. She gives a detailed introduction to the discipline of Decision Intelligence. In the article, she states that “Data are beautiful, but it’s decisions that are important. It’s through our decisions — our actions — that we affect the world around us”, and I couldn’t agree more.

However, as with the game of rock-paper-scissors, decisions can be fraught with bias. If bias in decision making impacts business outcomes at a micro level, over time, these outcomes can contribute to significant differences at a macro level. Try out Cassie’s quick but fun quiz to gain some insight into some of the dynamics impacting our decisions and how we can make better ones.

Guiding decisions to drive better business outcomes

The next question I asked myself was how to effectively mitigate the risk of decision bias within an enterprise setting. I spoke to a number of existing customers and partners and found that simple, quickly deployed interfaces do indeed extend the value created by their data science teams. For example, we helped one of our customers by working with their data science team to integrate their SAS forecasting models into a planning app now used at all levels of the organization, including the CEO. Our industry partners found that creating interfaces that are use-case specific on top of their powerful analytical models help bring the value of those models to life. The most surprising was perhaps our academia partners who found value in using apps as a way to encourage collaboration across faculties within their university.

So what is the point of all this, I hear you ask

The interview with Iwona got me thinking of what the real problem is we’re solving. And what the true value is to our customers. Going through this thought process led me to conclude that in the end, what matters most is making the best decisions possible at all levels of an organization in an informed and timely manner. To support these decisions, streamlined data pipelines and structured model governance are essential. While providing a framework for decisions, with responsive data-driven apps designed by the business, significantly increases the effectiveness of those decisions and drive better business outcomes. Ultimately, enabling customers to get more from their investment in analytics quickly and cost-effectively.

Is this an approach you agree with? I guess that’s for you to decide!!!

You can view the original post here