quantitative decision analysis

An organization thrives or fails based on the countless decisions, large and small, its leaders and managers make on a daily basis. An entire discipline – decision science – has emerged to help decision-makers do something deceptively simple: make better decisions.

However, how decision-makers today often arrive at the choices they select for their organizations includes bias, subjectivity, and unscientific methodologies that are based on gut feel and seldom better than chance. From a weighted score “decision-making matrix” to qualitative risk assessments based on expertise and experience, not statistics, organizations use many decision-making tools and processes that lead to unfavorable outcomes.

What is needed is a decision-making process for managers based on scientific quantitative methods fueled by data and embodied in mathematical models – all for the purpose of helping decision-makers objectively assess risk and determine the right course forward.

A good decision is one that makes the best of whatever resources are available, given a certain set of circumstances. What is “best” isn’t determined by intuition or wishful thinking, though; it ideally should be defined by a quantitative approach that gathers data, measures uncertainty, assesses risk, and determines the path forward that offers the highest probability for a favored outcome of all the potential choices.

In quantitative decision analysis, we use scientific methods to inform the decision-making process. These methods help data scientists put a value to several critical pieces of information, which include:

• Problem facing the organization

• Impact of the problem

• Uncertainty involved in the decision

• Desired outcomes

• Cost of various courses of action

• Overall risk involved in the decision

• Likelihood of various scenarios occurring

If it seems like a good decision can more or less be embodied by something as impersonal and inorganic as a mathematical equation, it can.  In fact, the best decisions made by executives, managers, and employees alike are typically informed by quantitative data and science in some way.

Put another way, when we provide leadership with actionable insights derived from quantitative decision analysis, we are doing something that is key to the success of a decision: reducing uncertainty.

The more uncertainty there is in a decision (and there is always some), the more complex the decision will become. What trips up many decision-makers is their inability to quantify what they don’t know or what they’re not sure of – to put a value on the difference between what will actually happen and what they believe will happen.

Quantitative decision analysis helps decision-makers choose better solutions for the challenges they face through measurement. The biggest reason why there is so much uncertainty in business today is because leaders aren’t doing a good job of putting a value to it. By measuring the right things in the right amount, decision-makers can make critical decisions faster and with greater confidence.

Decision models can be built that help managers make strategic decisions based on proper risk assessment. These mathematical models can evaluate various factors and criteria that can lead to a range of outcomes for a particular choice or set of choices. In other words, quantitative decision analysis can tell a decision-maker what is most likely to happen if a choice is made, and what the impact of that event is likely to be – thus pointing toward the most correct choice.