Do you want a simple and easy tool for making an important decision? The simplest “tool” is a heuristic. A short-cut like “take the default” or “do what everyone else is doing” may be fine for selecting a restaurant entrée or route to the grocery store. But what about more important decisions like taking a job offer, choosing a home or selecting an investment? That calls for a more powerful tool, like a decision matrix or decision tree. Since we’ve already discussed the decision trees, let’s take the red pill and dive into the Decision Matrix.
“You take the blue pill, the story ends. You wake up in your bed and believe whatever you want to believe. You take the red pill, you stay in Wonderland, and I show you how deep the rabbit hole goes.”
―Morpheus, to Neo
A Decision Matrix is simply a table where you list your criteria and score each alternative for each criterion. For example, “Anonymous” recently asked me on Quora, “How should I decide what job to choose?” This lucky person had two job offers and wanted help analyzing the decision. Here’s a more complete answer to how to do that:
- Define the decision opportunity, e.g., which job offer to take
- Find a quiet spot and mindfully, i.e. non-judgmentally, attend to the facts and feelings surrounding the decision. Be aware of potential biases and noise
- Identify the relevant stakeholders. Who is affected, besides Anonymous? Family, friends, current clients, future clients, even the environment may be included in the analysis
- List the stakeholders’ interests as criteria. Anonymous may be concerned about how well he likes management, the pay and the location for himself. A longer commute could disadvantage loved ones who spend less time with Anonymous and the environment which will be polluted more.
- Add weights to the criteria so that the weights total 100 percent
- Give each job a score, on a scale of, say, one to five for each criterion
- Calculate the weighted average score for each alternative by multiplying the scores by the criteria, summing up these products. You can see how the calculations work in this little spreadsheet (which you can download):
- Preliminarily choose the job with the highest weighted average score. (Job 1, above.)
- Discuss this preliminary choice with stakeholders and sleep on it
- Rescore and decide, secure in the knowledge that, while the future is uncertain, you’ve thought this important decision through.
A few cautions are in order. The first is about handling critical requirements. It is wise to create and use a checklist of must-have requirements before even creating the matrix. For example, ethics and legality may be of high importance to you. If Job 1 entails working at a gym that is really a front for a money launderer, well, that should obviously be ruled out on ethical and legal grounds. What other critical requirements do you have?
What about qualitative criteria like love, beauty, virtue, fun and personal or organizational values? The idea of converting these features to a number is especially challenging. One can handle some of these in a critical requirements checklist. For others, one just has to do one’s best and rely on Step 9. to permit one’s intuition to fairly account for them.
Finally, there’s uncertainty. Many of the scores and weights are subjective and could be more honestly represented as a range than a single number. Since the boss at Job 1 is still a bit of an unknown, her true score may be more in a range of 2 to 4 rather than just 3. A more complicated spreadsheet calculate a range of scores based on a range of inputs.
Of course there are many more nuances and caveats down this decision analytic rabbit hole. Indeed, as Daniel Kahneman has said, “No one ever made a decision because of a number. They need a story.” In my experience, numbers such as those calculated in a decision matrix can help justify a decision that people have already made perhaps intuitively. Even if this is so, the thoughtful decision analyst takes that red pill because, even today, facts matter.
- https://www.verywell.com/decision-making-strategies-2795483 See elimination of aspects for an interesting compromise between the matrix and heuristics