In this introductory lesson, we discuss why modeling is helpful in strategic decision situations, what constitutes goodness in decision models, and why Microsoft Excel is an appropriate platform for building and using these models.

Creating a decision model without first planning its structure is as unwise as constructing a building without a set of plans. In this lesson, participants are introduced to the decision diagram as a tool for describing the structure of a model. Then, working in small teams, they plan the structure of the model to be created later in the case exercise.

In this lesson, we review some of the features of Excel that are particularly useful in decision modeling. We discuss the use of range names - how they are created and identified and how they are usefully employed in models. We then discuss the distinction between local and global names. We continue with a review of the powerful but little-understood intersection operator in Excel. We discuss the difference between absolute and relative range names and show how relative names can be useful in decision models. Finally, we review a selection of built-in Excel functions and show how they can be used to describe a variety of time series profiles.

We start this lesson by presenting and discussing a few golden rules in decision modeling, such as making a clear separation in the model between inputs and calculated results. Then we take a detailed guided tour through a decision model that demonstrates the best practices that have been tested over many years of strategic decision consulting.

In the case exercise, the participants are assigned the task of creating a model for a typical business strategy situation. Using the concepts and skills learned earlier in the course, each participant builds a decision model for the case under the guidance of instructors. Through this exercise, participants gain firsthand experience in building decision models that they can later apply to real-world strategic decision situations.

Models are rarely created free of errors. In this lesson, participants learn that there are three distinct levels of quality in a model. Level 1 quality is whether the model runs without causing an Excel error, Level 2 quality is whether the model logic does what is intended, and Level 3 quality is whether the model adheres to best practices. This lesson teaches participants how to diagnose and correct model errors on each of these levels of quality.

Decision models are typically used in probabilistic analyses in which uncertainty in the inputs is transformed into uncertainty in the output. A variety of software packages are commercially available to do this probabilistic analysis. This lesson presents an overview of the software packages most commonly used. Participants then gain firsthand experience in using one of these packages with their case exercise models.

The reason for creating decision models is to generate insights that can guide and inform strategic decision-making. Which strategic alternative creates the most value? Why is it better than the others? How much risk does it entail? What are the most important sources of risk? In this lesson, we discuss these and other insights that come from modeling the decision situation. Participants learn how to use their models to provide these insights as well as how to create graphics that effectively communicate them.