How to Make a Difficult Business Decision

A VP of operations described her situation to me last year. She had a plant-closure decision on her desk and a board meeting in six weeks. She had spent $80,000 on a strategy consultant who delivered a 60-page report. Her risk team produced a 47-item register with a colour-coded heat map. She had a SWOT analysis, three scenario models, and a stack of papers she had not opened since the week they arrived.

She still could not decide.

I have spent nearly 50 years helping executives who need to make a difficult business decision, and I see this pattern everywhere. Smart, experienced people accumulate analysis until they run out of time, then decide under pressure on instinct, ignoring most of what they paid for. The analysis gave them the feeling of having done due diligence. It did not bring them one step closer to deciding.

Here is why: analysing and deciding are different activities. The tools designed for one are structurally incapable of performing the other. Until you recognise this distinction, no amount of consulting spend will help.

The apparatus that replaced thinking

Risk management, as widely practised, is a belief system. I use that phrase deliberately. Belief systems start with the answer and work backwards. The answer, in this case, is that organisations should "manage risk." Everything follows from that premise: the registers, the matrices, the frameworks, the consulting engagements, the board reports.

Notice what never gets asked. What decision are you trying to make?

The entire apparatus skips that question. It catalogues things that might go wrong. It assigns likelihood and consequence ratings on a five-by-five grid. It produces a register that sits in a shared drive or gets tabled at a board meeting. At no point does it help anyone decide anything, because that was never what it was designed to do. I have written about this at length elsewhere.

In Deciding, I traced how this industry sustains itself through four groups with strong self-interest: insurers who need to price premiums, regulators who need to demonstrate oversight, academics who need to publish, and consultants who need to bill. Each group has legitimate reasons to study uncertainty. None of them is trying to help a specific executive make a specific decision on a specific timeline. The compliance obligation created a lucrative circle, and the circle now perpetuates itself regardless of whether the work it produces helps anyone.

If risk management is the answer, what was your question?

That is not a rhetorical line. It is a diagnostic. If you cannot answer it clearly, you are commissioning analysis for its own sake.

The evidence is in plain sight

Enron had Arthur Andersen publicly praising the quality of its enterprise risk management. Boeing’s 737 MAX programme operated within a full risk management framework across the entire organisation. 346 people died in two crashes. Australia’s Royal Commission into banking misconduct found that every major bank had risk management committees, chief risk officers, comprehensive reporting, and assurance processes. Those banks had been engaged in systematic unethical and criminal conduct for years.

These organisations did not fail because they lacked risk management. They had all of it. Every register, every matrix, every committee, every report. The apparatus was fully assembled and fully useless when it mattered most.

The VP of operations has the same problem at smaller scale. Her 60-page report gives her the feeling of due diligence. That feeling is not the same as having decided. Her 47-item register gives her the sense that she has identified what could go wrong. That sense is not the same as knowing what she is actually assuming and whether those assumptions can bear the weight of her decision.

What good Deciders actually do

Roger Estall and I developed the Universal Decision-Making Method over decades of advisory practice. We did not set out to critique risk management. We set out to answer a more useful question: what do people who decide well actually do?

The answer was consistent. Good Deciders do not begin with a list of things that might go wrong. They begin with Purpose. They ask: what am I trying to achieve, and for whom?

This is Chapter 4 of Deciding, and we titled it "It’s the Purpose, Stupid!" because the number of organisations that skip this step is staggering. They jump straight to cataloguing hazards without ever articulating what they are trying to accomplish. The VP’s consultant wrote 60 pages about what might go wrong with closing the plant. Nobody wrote a single paragraph about what she was trying to achieve by closing it.

After Purpose comes a harder discipline: stating, out loud, what you are assuming to be true. Every decision rests on assumptions. The VP assumes her plant’s market will not recover. She assumes relocation costs will fall within a certain range. She assumes the workforce cannot be redeployed. Some of these are well-founded. Some are guesses she has never examined. Until she sorts the two, she cannot know where her decision is actually vulnerable.

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 matrix uses a single trigger question: if this assumption turned out to be wrong, how would it affect what I am trying to achieve? An assumption about raw material costs that could swing the decision by millions is Critical. An assumption about minor procurement logistics is Limited. The VP’s 47-item register treated every entry with equal weight. The significance matrix forces her to distinguish between assumptions that could break her decision and ones that barely matter.

A public safety organisation I advised had been allocating 0.03% of its budget to one safety programme. When we worked through their assumptions, 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%. No risk register would have surfaced that. A register would have listed "inadequate safety controls" alongside dozens of other entries, all coloured amber, none connected to a real decision about real resources.

Sufficient certainty, not maximum certainty

The question is not "have we identified all the risks?" It is "do we have sufficient certainty to proceed?" Confusing the two costs lives.

Australia once restricted blood imports to eliminate the risk of contamination. The policy achieved its narrow goal: no contaminated blood entered the supply. But the resulting shortages increased mortality overall. The Deciders had pursued maximum certainty in one direction and created worse outcomes in another. They were managing risk. They were not deciding well.

Sufficient certainty means something precise: for each significant assumption, do I have enough evidence to proceed, given what is at stake? Sometimes the answer is yes. Sometimes the answer is to investigate one specific assumption further. Sometimes the answer is to proceed now and monitor, because waiting will cost more than acting on an imperfect assumption. The point is that "sufficient" is calibrated to the decision, not to an abstract standard of rigour.

This is how to make a difficult business decision. Not by accumulating analysis until an answer reveals itself, because it never does. Not by building a register of everything that might go wrong, because that list has no end and no connection to what you are actually trying to decide. You decide by naming what you are assuming, testing whether those assumptions hold, and determining whether you have sufficient certainty to act. That is decision quality. Not the volume of analysis, but the rigour of the reasoning.

The VP of operations does not need another report. She needs a Decision Record that records what she decided, what she assumed, how significant each assumption was, and what would trigger a review. That is what her board needs. Not 60 pages of analysis. Clear reasoning she can defend.

---

Grant Purdy is the co-author, with Roger Estall, of Deciding (2020), and the architect of the Universal Decision-Making Method.