By Judee Beaman, Guest Contributor.
The business world is a strategic game that is full of events that may either spark your triumph… or lead to your downfall. The fundamental task of the game of strategy is to appraise the situation and then to organize your resources better than your competitors.
In the search for competitive advantage (and efficiencies, too) organizations are increasingly digitizing nearly everything, capturing information about locations, purchases, and preferences. Many people predict a “digital transformation” that will fundamentally change business models as they gain new ways to create scale and efficiency.
What Is Data Science?
In a nutshell, data science is a practice that involves the capturing, archiving, extracting, and analyzing data for the purpose of producing useful knowledge. When combined with enabling technology, individuals and organizations gain the capability to sense patterns and anomalies.
We often hear the buzzword, “big data” which refers to massive sets of data that aggregates millions of entities. Through data science, we have the capability to identify patterns that might have significant positive (or negative) impact on the business.
Data science provides a form of technology augmentation that enables evaluation of the information found in large data sets. Its techniques can help separate signal from noise. Data science has helped multiple industries extract crucial insights and knowledge through scientific methods, intricate processes, and intelligent algorithms.
Data Science Improves Strategic Thinking
Strategic thinking is a set of individual skills. No matter how intelligent or educated the individual, the fact remains that people have limited ability to sort through large data sets.
How to Think Strategically, describes 20 microskills of strategic thinking. Following are four key strategic thinking micro-skills and a brief discussion of how they can leverage the principles of data science.
Storytelling. In data science, the phrase “data storytelling” is typically associated with practices like data visualizations, infographics, dashboards, and other kinds of data presentations. In this narrow sense of story, the goal is to package data in so as to be easily digested by even those who are not familiar with the subject matter.
Story telling, as understood by strategic thinkers, recognizes that stories are a tool for sense making in ambiguous environments. Each person may look at the same event and interpret the facts differently.
People retain and tell stories based on a handful of familiar “story anchors.” Organizational culture is a repository of those story anchors and the strategic thinking task is to understand if a signal (that is, a story in the narrow sense) leads to a better understanding of reality. People have a discourse about “where we’ve been” and “where we are.” This naturally suggests speculations (and visions) about “where we’re going.”
Often the stories told in organizations are biased towards the status quo and explains why the pull of nostalgia is so strong. Strategy involves taking old stories and replacing them with better stories.
Sharpness. Strategic thinking requires the ability to detect nuances. The microskill of sharpness is the individual’s acumen to spot subtle discontinuities that can spark new understandings.
Data science approaches can help to identify the signals within the noise. Here, the data scientist helps to set the level of specificity versus sensitivity. Too high sensitivity and you get false positive signals that might be distractions. On the other hand, too high of specificity can cause you to miss patterns that have significance.
For example, many companies in the insurtech industry use chatbots that use algorithms to find out if a person is lying about a claim in only a few lines of conversation.
Reframing. The goal of reframing is to obtain better understandings of the situation.
In a purely objective sense, data science produces a set of digital patterns. Yet, those patterns are defined by definitions and assumptions which introduce ambiguity into the analysis.
The art of strategic thinking recognizes that one set of data can tell you one story, yet a different person can find yet a different story in the same data set. This kind of sense making is what sparks insights.
For example, instead of manually having to go through each employee company review, you can use a program that aggregates data. The resulting analysis might reveal that the “employee’s story” is different than the “manager’s story.”
Abductive reasoning. The “science” aspect of data science refers to an appreciation of scientific methods. The person generates a hypothesis, collects data and tests the hypothesis, resulting in a conclusion to confirm or reject the hypothesis.
It is through this process that science moves us toward understanding of truth and how the world really works.
Abductive reasoning is the mental process used to generate a hypothesis. Abductive reasoning basically means having a set of observations and predicting the next likely outcome. Using data science’s different statistical and analytical tools, in-house or outsourced data analysts can help organizations simulate and test crucial business and industry data. In turn, the analysis can help business leaders make more educated decisions that will advance their enterprise forward.
Many business leaders rely on deductive logic to cope with uncertainty. That is, they assume that everything that is worth knowing is already known. This feels safe to the individual, but often perpetuates the status quo.
Abduction requires the use of imagination: What could this trend in data mean? Will it continue? Or is it an anomality? Does the presence of an emerging justify making a large investment?
Conclusion: Integration of Data Science and Strategic Thinking
Data science offers significant benefits to organizations. But it is not a magic bullet. Like any tool, data science methodologies must be integrated into the organization’s culture.
The answers to these three questions can help to shape the direction forward:
- Are the data contextualized to the organization and its ecosystem? The external context of a business or industry affects the approach use by the data scientist. The data scientist needs to be sensitive to that context and the success factors of the organization. She uses her domain expertise and understanding of needs and goals so that she can design useful approaches to pattern detection, analysis, optimization, and prediction.
- What are the biases that are baked into the data structures? For example, the definition of markets and industries are convenient categorizations. By redefining the market or industry, organizations can create new value logics.
- Are people asking high-quality questions? Our strategies are often limited by the box that we put ourselves in. To think outside the box is to ask better questions with the expectation that better questions will spark better strategy.
For more on this topic: