Procter & Gamble Uses @RISK
and PrecisionTree World-Wide

by Palisade Staff, Spring 2001

"We've trained well over a thousand people throughout the company on @RISK," says Procter & Gamble's Bob Hunt, and now he's rolling out another Decision Tool, PrecisionTree. In fact, Hunt, who is Associate Director for Investment Analysis in P&G's Corporate Finance organization, and serves as a resource to the business units, was on his way to China and Japan to introduce PrecisionTree to P&G Finance managers in those countries.

P&G has been using @RISK since 1993 when Hunt first introduced it for modeling production siting decisions. The company was evaluating some cross-border siting options, and these decisions required them to take into account not only uncertainties involving the capital and cost aspects of plant location but fluctuations in exchange rates as well. The company has since come to rely on @RISK for its "entire range of investment decisions" including new products, extensions of product lines, geographical expansions into new countries, manufacturing savings projects, and production siting.

More recently Hunt and his colleagues have been working with PrecisionTree. "Its attraction is its capacity to value complex decisions, which often involve multiple, sequential decision steps". They find it particularly valuable in evaluating "real options". "We considered using financial option calculators to analyze the real options that are embedded in our complex decisions, but we found that they simply can't solve for the real option value in projects with multiple, sequential investment decisions. Decision trees are really the only tool that can correctly value multiple sequential decisions where uncertainty is private risk." Last spring, Hunt taught three (3) business units how to use PrecisionTree to test it as a tool for valuing complex decisions made under uncertainty. After a successful test, Procter & Gamble is now in the process of rolling out PrecisionTree to all of its major business units around the world.

Business units are evaluating investment options based on their impact on shareholder value, and PrecisionTree helps them make good choices and better decisions. "It has been very useful in helping us break complex projects down into individual decision options, helping us understand the uncertainties, and ultimately helping us make superior decisions." He also notes that a lot of the value derived from using PrecisionTree is realized during the process that the staff goes through in determining the probabilities, and laying out the decision sequence and the criteria for making those decisions. The combination of the different approach required to frame decisions, and the ease and effectiveness of the PrecisionTree software, says Hunt, "is really powerful for our company."

PrecisionTree

PrecisionTree is the Decision Analysis Add-In for Microsoft Excel. Build decision trees and influence diagrams directly in your spreadsheet! Create diagrams easily by selecting cells in your spreadsheet and clicking node buttons at the PrecisionTree toolbar. Enter probabilities and payoffs directly in cells in your tree. With one click, PrecisionTree will run a powerful decision analysis on your model, determining the best way to proceed with your decision.

Who should use PrecisionTree? Anyone faced with a set of alternative decisions should use PrecisionTree. Businesses use PrecisionTree to make decisions on introducing new products, factoring in decisions at each stage of marketing and production. Gas and oil companies use PrecisionTree to make the best decisions when developing an oil field. No matter what industry, PrecisionTree can work for you.

Decision Analysis

Decision Analysis is a process that allows you to examine decisions in a structured and efficient manner. Using decision trees and influence diagrams lets you detail all of your possible options and identify the best decision to make. Decision Analysis uses powerful analytical techniques developed at Stanford, Harvard and other universities over the past thirty years. The most efficient way to build a Decision Analysis model is graphically using influence diagrams and decision trees. Both methods use symbols, or nodes, and lines connecting them. Influence diagrams quickly define the general structure of a decision model. Decision trees provide a more formal structure in which decisions and chance events are linked from left to right in the order they would occur. (For example, before you decide where to have a picnic, you need to determine the chance that it will rain.) The result is a tree structure with the "root" on the left and branches for each chance event or decision extending to the right. Probabilities of events occurring and payoffs for events and decisions are added to each node in the tree.

Benefits of Decision Analysis

  • Model decisions efficiently and effectively
  • Formally analyze decisions, clearly laying out your goals and options
  • Decision trees capture a chronological sequence of choices and possible events
  • Make well-informed decisions
  • Influence diagrams capture the general structure of a decision
  • Explain and document your reasoning to others


Dave Bristol, Palisade Technical Sales, gave us permission to use this case at DSSResources.COM on Wednesday, May 9, 2001. For more information, check www.palisade.com. Posted May 22, 2001.