Prediction Versus Prescription in Customer Experience Optimization

One of my most favorite books is Competing on Analytics, co-authored by expert and university professor Tom Davenport from a school I attended back in the mid-1980's Babson College . While the whole book was informative and eye-opening, there was specific information that resonated most around the three levels of analytics; Descriptive, Predictive, and ultimately, Prescriptive. In short, as I describe these, they answer specifics about what happened? what might happen? and what should we do about it?

Careerwise, whether here at QuestionPro or for about 25 years leading up to it, I've focused my work on helping mid- and large-sized companies impact their customer experience as a means of driving increased revenue, ensure customer delight and in it, maximize customer retention and value and lower operating costs to serve them. Analytics has always played a pivotal role in this work as without data and related insight, even the best recommendations are subjective and based on fancy guesswork and as such, marginal and potentially error-prone. Clients who are familiar with some of Tom's teachings ask about how we can move up the maturity curve and strive for real prescription as that's where the rubber hits the road for most. A data-enabled, insight-generated answer to the "What should we do?" is the billion dollar answer, more often than not as the other two are largely just lead-ins to the same discussion, regardless.

Drilling into this a bit farther, Predictive analytics, as the name suggests, focuses on forecasting. It employs statistical models and forecasting techniques to understand future behavior based on historical data. Predictive analytics can answer questions like: "What could happen in the future? What are the possible outcomes?" It may not tell you how to respond to these potential events, but it provides a probabilistic prediction, enabling the identification of risks and opportunities. Applications of predictive analytics are widespread, from forecasting customer churn in telecoms to predicting stock prices in finance.

On the other hand, prescriptive analytics goes a step further to recommend specific actions that can take advantage of the predicted outcomes. It answers the question: "What should we do?" It uses advanced tools, such as machine learning, business rules, and algorithms, to advise on possible outcomes. The power of prescriptive analytics lies in its ability to optimize scheduling, production, inventory, and supply chain design to deliver better business results.

While both predictive and prescriptive analytics are rooted in data-driven decision-making, their focus diverges in purpose and application. Predictive analytics is more about forecasting, providing a probable outlook based on past data. It identifies potential future outcomes but doesn't necessarily suggest actions to take. Prescriptive analytics, however, is action-oriented. It provides actionable insights and suggests specific steps to achieve a desired outcome or mitigate an undesirable one.

Moreover, the complexity involved in both types differs significantly. Predictive analytics uses straightforward statistical techniques, while prescriptive analytics involves complex algorithms and computations. Despite this, the ultimate goal of both is to enhance decision-making and enable businesses to leverage data effectively.

Nevertheless, the two types of analytics aren't standalone. They complement each other in providing a holistic approach to data-driven decision-making. Predictive analytics can feed into prescriptive analytics, offering valuable foresight that can be used to devise optimal strategies.

Predictive analytics, as the name suggests, focuses on forecasting. It employs statistical models and forecasting techniques to understand future behavior based on historical data. Predictive analytics can answer questions like: "What could happen in the future? What are the possible outcomes?" It may not tell you how to respond to these potential events, but it provides a probabilistic prediction, enabling the identification of risks and opportunities. Applications of predictive analytics are widespread, from forecasting customer churn in telecoms to predicting stock prices in finance.

On the other hand, prescriptive analytics goes a step further to recommend specific actions that can take advantage of the predicted outcomes. It answers the question: "What should we do?" It uses advanced tools, such as machine learning, business rules, and algorithms, to advise on possible outcomes. The power of prescriptive analytics lies in its ability to optimize scheduling, production, inventory, and supply chain design to deliver better business results.

While both predictive and prescriptive analytics are rooted in data-driven decision-making, their focus diverges in purpose and application. Predictive analytics is more about forecasting, providing a probable outlook based on past data. It identifies potential future outcomes but doesn't necessarily suggest actions to take. Prescriptive analytics, however, is action-oriented. It provides actionable insights and suggests specific steps to achieve a desired outcome or mitigate an undesirable one.

Moreover, the complexity involved in both types differs significantly. Predictive analytics uses straightforward statistical techniques, while prescriptive analytics involves complex algorithms and computations. Despite this, the ultimate goal of both is to enhance decision-making and enable businesses to leverage data effectively.

Nevertheless, the two types of analytics aren't standalone. They complement each other in providing a holistic approach to data-driven decision-making. Predictive analytics can feed into prescriptive analytics, offering valuable foresight that can be used to devise optimal strategies.

Both predictive and prescriptive analytics play crucial roles in improving business outcomes. Predictive analytics offers foresight, allowing for better planning and risk management, while prescriptive analytics offers actionable insights for decision-making, process optimization, and achieving operational excellence. Both serve as vital tools for businesses seeking to leverage data for strategic advantage. Predictive and prescriptive analytics are both invaluable assets to modern businesses. While they have distinctive roles and methodologies, their combined use offers a comprehensive understanding of what the future might hold and how best to navigate it. By harnessing the power of both, organizations can gain a competitive edge, make informed decisions, and drive operational efficiency.

Marc Mandel, CCXP is the North American VP at QuestionPro, the only full-stack CX solutions company. (www.questionpro.com). These views, however, are his own.