The Thinker's Toolkit

books1 create new tag

Author(s): Morgan D. Jones

Review

The Thinker's Toolkit is a marvelous book and a quick read. I recently went through it for about the 4th time. I like to read it between consulting engagements, and I have given it to favorite clients as a present several times. The author, Morgan D. Jones is a former CIA analyst and is the founder of a consultancy and training firm called Analytic Prowess. The book is filled with case studies and exercises to illustrate each tool. There are step by step instructions for each tool

The book is 368 pages and the author presents 14 techniques for solving problems and structuring the decision making process. The table of contents includes the following:

Part One - Why We Go Astray

Thinking about Thinking

Insights into Problem Solving

Part Two - The Fourteen Tools

Problem Restatement

Pros-Cons-and-Fixes

Divergent/Convergent Thinking

Sorting, Chronologies, and Time Lines

Causal Flow Diagramming

The Matrix

The Decision/Event Tree

Weighted Ranking

Hypothesis Testing

Devil's Advocacy

The Probability Tree

The Utility Tree

The Utility Matrix

Advanced Utility Analysis

Part Three - Where Do We Go from Here?

Your Next Steps

Solutions to Exercises

Several of the tools naturally lend themselves to spreadsheet solutions. I constructed a spreadsheet for my personal use and have attached the file Thinkers_Toolkit.xls to this page. The following sections are documentation for the use of the spreadsheet.

Pros, Cons, and Fixes

The spreadsheet is set up to evaluate three alternatives, but you can add more.

1. State the problem or the issue.
2. List all of the pros of each alternative under consideration.
3. List all of the cons of each alternative.
4. Review and consolidate the cons - merge and eliminate.
5. Neutralize as many cons as possible - list the fixes (hint: this is the valuable part that most people don't do.
6. Compare the pros and unalterable cons for all the options.
7. Pick one of the alternatives.

Weighted Ranking

The spreadsheet is set up to evaluate up to six items with up to six decision criteria. You are evaluating the items in a series of paired rankings for each decision criteria. The spreadsheet tallies the results which can be surprising. For example, suppose you were going to purchase a new car and you had five models (items) to evaluate with six criteria (purchase price, mileage rating, style, available financing, and so forth). Here is the procedure:

1. List up to six decision criteria in cells b2 through g2.
2. Pair rank the decision criteria in step one to determine the relative importance of each criteria.
3. Assign a percentage weight in cells b3 through g3 for each criterion.
4. Enter up to six items to be ranked in column A. The item pairs will show in the matrix below. Hint: keep names short.
5. Pair rank all the items in column A for each criterion. Enter the winner of each pair in the paired rankings section.
6. Evaluate the votes shown in column H and I

Hypothesis Testing

In this spreadsheet, the sheet doesn't do much except to provide a structure for listing and evaluating evidence and hypothesis. This technique is exceptional for problem diagnosis. In the software arena, I use it to diagnose bugs because the technique tends to isolate and define the root cause of the problem. In the business arena, the technique could be used to resolve issues like "why are overtime costs so high?" or "what is causing the quality issues with part 1234?" Here is the procedure:

1. Generate a mutually exclusive list of hypothesis and enter the title for each in row 3.
2. List significant evidence about the problem in column A.
3. For each row of evidence, test for consistency with each hypothesis. Enter a C, I, or "?" in the matrix if the evidence is consistent or inconsistent with the hypothesis.
4. Refine the matrix by adding or rewording the hypotheses. Add new evidence. Since it doesn't contribute anything to the solution, delete evidence that is consistent with all hypotheses.
5. Evaluate each hypothesis by retesting all inconsistent evidence and delete hypotheses with significant inconsistent evidence.
6. Rank remaining hypotheses with the weakest inconsistent evidence (the weakest is the most likely).
7. Perform a sanity check on the results.

Utility Matrix Analysis

This spreadsheet has two tables that accomplish the same thing and can evaluate six options. Utility is the benefit or expected value of a particular outcome. These sheets handle the calculation and relative ranking of the expected values for different options. The Yes-No Scenario can be used to evaluate two outcomes for each scenario. The odds of each outcome don't have to be 50-50. The multi-outcome table can evaluate up to four different outcomes with various probabilities. You define the options, the effect of the outcome(s), and the probability that the outcome will occur. The sheet will compute and rank the effective values of the options. Here is the procedure:

1. Identify the options and outcomes to be analyzed and enter labels in the blue boxes.
2. Identify the perspective of the analysis. Advanced analysis may consider multiple perspectives. The perspective is important to determine utility values of various outcomes. For example, employees, managers, and shareholders may have different perspectives about the relative utility of various capital projects in the budget.
3. For each option-outcome combination, assign a utility value of 0 to 100 (or use actual money amounts). Ask the question: "If we select this option, and this outcome occurs, what is the utility (benefit) from the perspective of ...?"
4. Assign a probability to each outcome. Ask the question: "If this option is selected, what is the probability this outcome will occur?"
5. The sheet will calculate the expected values and determine the ranking of the alternatives.
6. Perform a sanity check.

Reviewed by: JimCrum

Discuss This Book

Was This Topic Useful to You?

Choice

Select ...Very UsefulSomewhat UsefulNo OpinionNot Useful

Very Useful (1)