What is a decision matrix?
A decision matrix is a table that compares several options against the criteria that genuinely matter. Each criterion carries a weight, saying how much it counts in this particular decision, and each option gets a score on each criterion. Multiplying and adding gives a score that ranks the options.
The valuable part is not the final number. It is what you have to do to get there: name the criteria out loud and weight them. Plenty of people discover right there that what they said mattered is not what weighs most, and that is the conversation the session was looking for.
What it is for in a coaching session
- Taking the decision out of the loop. Whatever spins around in your head takes up less space in writing, and can be looked at.
- Making the criteria visible. It is the part nobody does alone, and the one that changes the conversation most.
- Separating weight from score. An option can be excellent at something that barely matters, and that shows at a glance.
- Spotting the criterion that decides. There is nearly always a single one that splits the decision. Naming it clarifies more than the result.
- Giving permission to decide. Sometimes the client already knows, and only needs to see their own reasoning laid out.
How to use it, step by step
- Write the decision in one sentence. If it does not fit in one sentence, it is probably two decisions.
- List the real options. Include "carry on as I am": it is usually the one nobody writes down and the one that wins most often.
- Draw out the criteria. Ask "what would the good option have to meet?" until no more come up.
- Set the weights before the scores. Done the other way round, the weights quietly bend so the favourite option wins.
- Score each cell. Quickly, without agonising over every box.
- Look at the result and listen to the body. If the number wins and the face says no, a criterion is missing.
How to read the result
- A large gap between first and second: the decision was already made, it just needed seeing.
- A tiny gap: the options are equivalent under the criteria that were set. Either a criterion is missing, or it genuinely does not matter and what is needed is to choose and move on.
- One criterion at weight 5 deciding everything: the decision may actually be about that criterion, not about the options.
- All weights equal: usually means nothing has been prioritised yet. Worth going back over them.
- An option you do not like wins: that is the best possible outcome. There is an unnamed criterion in there, and finding it is the work.
Questions to open the conversation
About the criteria
What would the good option have to meet? Which of these criteria belongs to someone else and not to you?
About the weights
If you could keep only one criterion, which one? Which would you say out loud, and which would you not?
About the scores
Which option did you score high without real information? What would you need to find out?
About the result
What did you feel when you saw which one won? Relief, resistance, surprise?
About what is missing
Which option did you not write down? What stops you considering it?
About the next step
What would you do this week if the decision were already made?
Common mistakes
- Setting the weights after the scores. They bend without meaning to, so the favourite option wins.
- Too many criteria. Beyond six, everything averages out and no option stands out.
- Criteria that are not the client’s. If a criterion belongs to their mother or their boss, it is worth naming it that way.
- Treating the number as an order. The matrix orders the reasoning; the person decides.
Frequently asked questions
How many criteria should be used?
Three to six. With fewer it adds nothing over thinking it through; with more, everything averages out.
What if two options tie?
It means that under those criteria they are equivalent. Either a criterion is missing, or it does not matter and it is time to choose.
Should "carry on as I am" be included?
Almost always. It is a real option and it usually scores better than the client expects.
What if I do not like the result?
That is the most useful part of the exercise: it means a criterion has gone unnamed. Finding it is the work.
Does it work in online sessions?
Yes. Share your screen and fill it in together, criterion by criterion.