What Is Decision Coaching?

Every piece of decision support I have ever used gave me more information. None of it made me think harder.

Reports, registers, scenario models, pros-and-cons lists. All of it tells you what could happen. None of it forces you to do the hard part: figure out what you are actually trying to achieve and what you are assuming to be true about how you will get there.

That gap between having information and actually deciding is where decision coaching sits. It is not more analysis. It is a structured process that walks you through your own reasoning until that reasoning either holds up or falls apart.

Consulting delivers analysis. Decision coaching makes you think.

Most decision support follows the same pattern. You bring a problem to someone. They go away, study it, and come back with a document. A consultant delivers a report. A risk team delivers a register. An advisor delivers recommendations. You get a stack of paper and a feeling of due diligence. You still have not decided.

My father, Grant Purdy, spent nearly fifty years helping boards, regulators, mining companies, and banks work through decisions. He and his co-author Roger Estall watched the same thing happen over and over: smart, experienced people accumulate analysis until they run out of time, then decide under pressure on instinct. The analysis gave them the feeling of progress. It did not bring them one step closer to an actual decision.

In their book Deciding, they argue that this happens because analysing and deciding are different activities. The tools designed for one are structurally incapable of performing the other. A consultant can tell you everything that might go wrong with closing a plant. That is analysis. Decision coaching makes you articulate what you are trying to achieve by closing it, what you are assuming will happen when you do, and whether those assumptions can bear the weight of the decision. That is deciding.

What decision coaching looks like in practice

The Universal Decision-Making Method that Grant Purdy and Roger Estall developed has five steps. Each one has to earn the next.

Purpose first. Before anything else, you articulate what you are actually trying to achieve, and for whom. This sounds obvious. In practice it is the step almost everyone skips. When I built the Walk, the product based on this method, early versions let people breeze through Purpose in ten seconds and spend all their time on risk and assumptions. My dad's feedback was one line: "You need to start with purpose. That's your purpose, not that of the decision." Without a clear Purpose, every option looks roughly equivalent because you have no basis for comparison.

Tentative decisions. State what you are actually thinking of doing. Not a brainstormed list of twelve options. The one or two things you are genuinely considering.

Assumptions. Every decision rests on assumptions. Some are well-founded. Some are guesses you have never examined. The Universal Decision-Making Method provides a significance matrix that classifies each assumption on two dimensions: how significant is it to the outcome, and how well-supported is it? The trigger question: if this assumption turned out to be wrong, how would it affect what you are trying to achieve? An assumption that could swing the decision by millions is Critical. An assumption about minor logistics is Limited.

A public safety organisation that Grant Purdy advised had been allocating 0.03% of its budget to one safety programme. When they surfaced their assumptions using the significance matrix, one Critical assumption about the effectiveness of existing controls turned out to be completely unsupported. They raised the allocation to 0.5%. Mortality in that area dropped 60%.

Sufficient certainty. The question is not "have we identified all the risks?" It is "do we have sufficient certainty to proceed?" You do not need to know everything. You need to know enough, given what is at stake. Australia once restricted blood imports to eliminate the risk of contamination. The policy achieved its narrow goal but the resulting shortages increased mortality overall. The Deciders had pursued maximum certainty in one direction and created worse outcomes in another. "Sufficient" is calibrated to the decision, not to an abstract standard of rigour.

Implement and monitor. Name the conditions that would tell you your decision needs revisiting. If assumption X turns out to be wrong, what will you do? This is how decisions stick instead of getting relitigated every few weeks when someone gets nervous.

Why assumptions, not risks

Most decision support starts with risk: what could go wrong? Grant Purdy has spent decades arguing that this framing is fundamentally broken. In Deciding, he calls the entire risk management apparatus a belief system. It starts with the answer, "you should manage risk," and works backwards. Four groups sustain it: insurers who need to price premiums, regulators who need to demonstrate oversight, academics who need to publish, and consultants who need to bill. Each has legitimate reasons to study uncertainty. None of them is trying to help a specific person make a specific decision on a specific timeline.

A risk register catalogues things that might go wrong. It assigns likelihood and consequence ratings on a five-by-five grid. At no point does it help anyone decide anything, because that was never what it was designed to do. It was designed to demonstrate that uncertainty had been "considered." Consideration and decision-making are different activities. Grant has written about this at length.

Decision coaching starts from the other direction. What are you trying to achieve? What are you assuming? Where are you most vulnerable?

The difference is not semantic. After the 2011 Brisbane floods, an inquiry into the Wivenhoe Dam produced 177 recommendations. None of them helped anyone decide anything. They were analysis, generated after the fact, pointing in every direction. If the original Deciders had been coached through their assumptions before the event, the conversation would have been about a handful of Critical assumptions connected to a specific Purpose, not a catalogue of everything that could theoretically go wrong.

Decision coaching is not advice

There is a reason the Walk does not tell you what to decide. Grant's position, developed across nearly fifty years of advisory practice, is that the person who has to live with the decision is the person who should make it. A coach does not know the answer. A coach makes sure the Decider has actually done the thinking.

I built the Walk around this principle. It does not generate options or recommend a path. It asks you to name what you are trying to achieve, state what you are assuming, and work out whether those assumptions hold. The output is a Decision Record that captures what you decided, what you assumed, how significant each assumption was, and what would trigger a review. It is not a report for your shelf. It is reasoning you can defend.

That is the gap most tools miss. They give you more information when what you actually need is a process that forces you to examine what you already believe. Decision coaching does not add to the pile. It makes you sort through it.

If you are facing a decision and want to see what this looks like in practice, the Walk might help.

---

Joseph Purdy is the founder of the Walk at Sufficient Certainty. His father, Grant Purdy, is the co-author, with Roger Estall, of Deciding (2020), and the architect of the Universal Decision-Making Method.