To identifying solutions you must -1- identify all the possible solutions for your problem and -2- rank them. To rank them, you need to compare them; decision matrices are a handy tool to do that.
To build a decision matrix, you need to identify:
- The criteria against which you will rate options and its scale and
- The relative weight of these criteria.
Get the criteria right
Part of identifying criteria is understanding which criteria you want to consider. You guessed it, these should be reasonably MECE. In particular, the mutually exclusive part means that you need to use criteria that are not causes (or consequences) of others.
So, let’s go back to the example of going from New York to London, assuming that we want to go there for a business trip. You might decide that the criteria to consider are price, speed, comfort and visibility.
The other part of identifying criteria is developing a common scale for performance. Let’s say that you want to use a 1-to-5 graded scale for each criterion, 5 being the top score. The more quantitative you can be about what these grades mean, the better.
In the going-from-NYC-to-London example, having identified that the criteria are price, speed, comfort and visibility, we would also identify a graded scale for each. For instance, the price scale would scale from 1, which is very expensive—say above $100k—to 5, which is very cheap—say, less than $500. You might have to arrange your scale as you rate your options; what matters is that you ultimately rate all options with the same yardstick.
Get good weights
The other part is the relative weight of your criteria. I like to have the sum of the weights equal to 1 but, since what really matters is the comparative weight of one against the others, you can use other methods.
The weight of categories depends highly on your goal. If you want to go from NYC to London for a business trip, comfort and speed will be important and therefore have a high weight. If you want to do the trip for a vacation, maybe you’ll want to favor price. If you want to do it as a publicity stunt, say to attract attention to your new business, then visibility will have by far the highest weight.
Then plug the numbers in the matrix
Once you’ve selected the criteria, scales and weight, you’re ready to plug numbers in your matrix and see what it yields.
To learn more about decision matrices, visit the American Society for Quality page.
Chevallier, A. (2016). Strategic Thinking in Complex Problem Solving. Oxford, UK, Oxford University Press, pp. 141–160.