They all had a due diligence report. Why it did not save them

Large buyouts almost always have a commercial due diligence report. So do most of the ones that fail. In the deals we have scored, the report was rarely wrong. It answered the questions it was asked, at the time it was asked, for the people who asked. The buyer controls all three.

The warning is usually there to be read. We tested our fund assessment on 105 public buyouts, in an exploratory, AI-assisted retrospective test. It raised a warning on 43 of the 51 that later failed, and on 14 of the 54 that succeeded. The limits are on the method page: some answers rest on what the AI agents already knew, and the warning was tuned on these deals.

Three deals in our own files went badly for their owners. Each had a full commercial report. One of them had a good one. It covered the market, the customers and the competitors in depth. The report was right about all three.

That deal failed on a question nobody had put in the scope. Could the business build at several times the pace it had ever managed? Nobody asked, so the report did not say.

The public record shows the same thing. Healthscope was bought in 2019 after a contest that ran for a year. Shareholders were sent an independent expert's report.

The expert warned, in one sentence, that selling and renting back the hospitals would bring "higher operating leverage and increased volatility in earnings". Anyone could read it. Six years later receivers were appointed.

The information was there. It did not change the decision.

Three reasons a good report does not save a bad deal

It arrives after the decision. By the time the work is commissioned, the price is agreed in principle and the buyer is in exclusivity. People have told their partners, their lenders and their investors that the deal is happening. A finding at that point costs fees, face and months of work. So the buyer reads the report as confirmation, and the advisers write it for readers who want it.

The scope is written by the people who want the deal. A typical scope covers the market, the customers and the competitors. Those are the right questions, and there are more.

When we scored deals from their diligence reports, the same three questions were left unanswered almost every time. Can this team deliver the plan? Who will buy the business at the end, and why are they not buying it now? What does the real downside look like?

It is read for comfort. Every report has a risk section. The risks are listed, each is judged unlikely, and nobody works out what it would cost. In the failed deals we reviewed where one payer or one regulator changed the terms, the risk had been written down every time.

The companion piece covers why capable people do this. What follows is how to set the work up so it is harder to do.

What to ask for

A scope that gets the right answers has three parts.

1. The four ways earnings fall

In the failed buyouts we reviewed, earnings fell for one of four reasons. Put all four in the scope, as questions that need an answer.

• Is the segment growing, in volume? The segment the business sells into, with price rises taken out.
• Where does this year sit in the cycle? How far earnings fell in the last downturn, and how long they took to recover. Whether the base year contains one-offs.
• Who else can set the price? Every payer, regulator, licence, landlord and large customer, with renewal dates laid against the years of ownership.
• Has the business done this before? The plan bridged to the last three years, driver by driver. For every step above the run-rate, what changed, and who will deliver it.

Ask for customer calls that include customers the seller did not introduce.

2. The two questions reports tend to skip

• Can this team pull the levers? Named people, systems tested at the volume the plan needs, and one comparable thing they have already delivered.
• Who buys it at the end? Two or three named buyers, what each would pay for, and why they are not buying it today.

3. The risks, run as scenarios

Most scopes leave this part out, and it is the one we would add first.

Most reports list the risks and explain why each one is unlikely. Few show what happens to the deal if one of them occurs. The risk table has a column for likelihood and a column for mitigation. It rarely has a number.

Likelihood is the wrong place to start. Nobody knows whether a government will change a rule, or a large customer will leave. The deal team will always win that argument, because nobody can prove them wrong.

Impact is a different kind of question. It has an answer, and the answer is in the model.

So pick the three or four risks that matter. For each one, ask the advisers for a scenario.

• What happens to earnings if it occurs? By how much, and for how long?
• What does the business do about it? What can management change, how fast, and what does that cost?
• Does the debt survive it? Covenants, interest and rent, on the structure as it will be after the deal.
• What is the equity worth at the end, if it occurs in year two?
• What would we see first? The early sign that it is starting, and who is watching for it.

A useful scenario is a story, not a percentage. "Revenue down ten per cent" tests nothing. "The largest customer moves half its volume to a rival in year two" tests the deal.

Then look at the risks together. Five risks that are each unlikely are, together, likely to produce one event in a five-year hold. And some arrive as a pair. A downturn and a refinancing tend to come at the same time.

The answers sort the risks into three kinds. A risk the deal survives can be priced. A risk that can be passed to someone else, through the contract, the structure or the price, should be. A risk the deal cannot survive is a decision. The committee should make it, out loud.

The typical low case does none of this. In the deals we have scored, the low case still showed growth in most of them, winners and losers alike.

How to keep the answer honest

Scope is half of it. The other half is how the buyer behaves while the work is done.

Commission it early. A red-flag review takes a week or two. Run it before the price is agreed, while walking away is still cheap.

Write down what would make you walk. Do it before the work starts. Give the list to the advisers. Ask them to test those points first.

Have the advisers report to the committee. The deal team can hear them as well. Let the committee hear them at least once without the deal team in the room.

Ask the two questions that are rarely asked. What would make you say no to this deal? What in this report are you least sure of?

Do not edit the conclusions. Correct the facts in a draft. Leave the judgements alone. If the final report is softer than the first draft, the committee should see both.

Keep the fee independent of the outcome. An adviser paid more if the deal completes has the same problem the deal team has.

Let someone score the case who does not sponsor it. The person who found the deal is the wrong person to judge the evidence for it.

What the advisers need from you

Good advisers want to be asked the hard questions. Three things make it possible.

Time, because a report written in the last ten days of exclusivity can only confirm. Access, to customers, to data and to management. And permission to bring bad news, given out loud, at the start, by the most senior person on the deal.

A buyer who gives all three gets a different report from the same firm.

Where a ten-minute check fits

Before the diligence starts, have each member of the committee score the case separately. Our fund check takes ten minutes each. Point the advisers at the places where they disagree. Run it again when the report arrives, and see what moved.

The check stops short of saying whether to do the deal.

It gives one of four readings of the case. Keep going. Slow down and finish the homework. Slow down, there are real doubts. Or stop.

• The deal checks: buying as a fund, buying as a strategic acquirer, and selling
• How a Stages of Excellence check works
• No budget for due diligence? Four checks you can run yourself