In Private Equity, the First Job Is Not to Lose the Money
I have sat through a lot of investment presentations, most of them as the investor on the other side of the table. The great ones and the awful ones all have the same page.
It is the returns page. 3x. 25% IRR. Multiple expansion. Operational alpha. Upper quartile performer.
The page I almost never see is the other one. The page that answers a much less fun question:
What is our probability of permanently losing this money?
I do not mean the downside case. The downside case in most models is just the base case with a haircut, built by the same person who built the base case, on the same Tuesday. I mean the real question. What are the odds this capital does not come back at all?
Ask it out loud in a room and watch what happens. People shift. Someone says “well, that’s why we have portfolio construction.” Then everyone goes back to the returns page, because the returns page is where the fun is.
The risk of failure is far greater than the odds of success
That is the sentence I would put on the first page of every deck, including my own.
Not marginally greater. Far greater. Most companies do not become the case study. Most of them land somewhere between disappointing and fine, and a real number of them go to zero. That is the base rate of this asset class and it is indifferent to how good your deck is.
Underneath the base rate sits the part that actually decides careers, which is that a win and a loss do not weigh the same. A great outcome adds to the record. A permanent loss can end the record, because it takes away the one thing you need in order to have a record at all, which is another turn.
So we are underwriting an activity where the bad outcome is both more likely and more consequential than the good one, and we build the entire presentation around the good one.
The math gets uncomfortable very quickly
Take a number, purely as an illustration. Say an investor carries a 5% chance in any given year of a catastrophic, permanent loss of capital. Arguably a low risk for private investing. Now, hold that risk constant year over year. Just a 5% risk of an investment failure.
What are the odds of getting through 20 years without hitting it?
36%.
Cut the horizon in half, to 10 years, and the odds only climb to 60%. A decade of work, and the survival of the original capital is a coin flip with slightly better manners.
That should sound daunting. It is.
Now move the annual probability to 10%, which is not a wild number for a strategy that reaches for the fences. Over 20 years the odds of survival fall to 12%. Even 10 years is only 35%.
Read that again. A 10% annual chance of ruin means you almost certainly do not survive a career.
This is the part of compounding nobody puts on a slide. We love to show what capital does when it compounds. Risk compounds in exactly the same way, quietly, in the same spreadsheet, in the other direction.
So the first obligation is embarrassingly simple
Do not lose the LPs’ money.
That is it. That is the job before all the other jobs.
I want to be precise here, because this gets misread as timidity. It is not a call to avoid risk. Private equity exists because we are willing to underwrite risk that other people will not touch. If you want to avoid risk, there are Treasuries for that, and they do not require a data room.
The distinction is between taking intelligent risk and quietly accepting a real probability of permanent impairment. Those two things get filed under the same heading in most memos, and they are not remotely the same.
Staying in the game is the whole strategy
Here is what protecting the downside actually buys you: another turn.
A manager who produces solid returns and avoids the crater gets something that compounds harder than any single deal ever will. More chances. Another fund. Another platform. Another operating improvement that finally lands in year four. Another exit into a window that finally opened. Another decade.
The spectacular win gets written up. Survival is what builds the fortune. Those are two different games and only one of them is repeatable.
I have never met anyone who got rich in this business on a single deal. I have met plenty of people who got taken out of the business by one.
How this changes underwriting
Once you believe the above, the order of the questions has to change.
Most diligence starts with “how much can we make?” and then works down through the risks as a compliance exercise, because by then everybody in the room already wants to do the deal. Start at the other end instead:
How do we lose the money?
Then work backward from that. What happens if revenue misses plan by 20%. What happens if the multiple compresses and the exit comp set gets cut in half. What happens if the CEO leaves in month nine. What happens if the financing market simply closes, which it does, on no notice, for reasons that have nothing to do with your company. What happens if the integration takes twice as long as the plan says, which it will. What happens if we own this thing for seven years instead of four.
Then the question that matters more than all of them:
Can this business survive every one of those outcomes without permanently impairing the equity?
If the answer is yes, you have a deal worth arguing about. If the answer is “yes, as long as two of those do not happen at the same time,” you do not have a downside case. You have a hope.
The pre-mortem, which is the part we actually built
Believing all of this is easy. Turning it into something that happens whether or not anyone in the room feels like doing it is the hard part. So we made it a step in the process.
Before a deal goes to our investment committee for approval, it gets a pre-mortem. Not a risk section buried in the memo. A separate exercise, with one instruction:
Assume the deal was done and it failed completely. It is 24 months later. Write the autopsy.
The framing is the whole trick. A risk section asks people to imagine what might go wrong, and people are politely terrible at that, particularly once the room has decided it wants the deal. An autopsy starts from the failure as a finished fact and asks what killed it. Different question, different answers.
Five things come out of it.
The autopsy. The seven most likely causes of death, ranked by likelihood. For each one: how it unfolded quarter by quarter, the assumption the committee was holding that allowed it, and the first warning sign the deal team would have seen.
That middle item is the one that earns its keep. Most failures do not come from a risk nobody listed. They come from an assumption nobody noticed they were making.
The verdict. The most likely killer, the most dangerous killer, and why those two are usually not the same. The single biggest hidden assumption in the thesis, the one the deal team does not realize is an assumption. And where the record supports it, a fatal flaw, stated in plain words. The exercise is explicitly not permitted to reassure us. If there is no fatal flaw it says so. If there is one, it says the words, at the top, in red.
The rebuild. What would have to change for this to be a good deal instead of this deal: structure, price, terms, the first hundred days. Then a short list of things that must be verified before the vote, and beside each one, the result that means walk away entirely. Not “look into it further.” Walk away.
The adversary. Whoever benefits most if we fail, played straight. The rival bidder, the incumbent, the competitor, the founder already halfway out the door. Where they would attack the thesis, what they do the quarter we close, and the move we would never see coming.
The tripwires. For every failure mode, one measurable signal that it has started, and the exact month after close when someone goes and looks. Tripwires, not vibes. A failure mode you cannot measure is a feeling, and feelings do not make it onto a board agenda.
There is one more rule, and it is the one I care most about. The pre-mortem is only allowed to read what is genuinely on the record: sourced facts, current documents, the actual screen. It cannot fill a gap. Where something is missing it prints what it could not read, and the absence itself gets treated as a risk. Every run carries two lists at the top, what it actually read and what was missing or withheld, so nobody can mistake a confident paragraph for a well-evidenced one.
Every run is kept. Who ran it, when, what it said.
That last part is the real point, and it has nothing to do with prediction. It is a record of what we believed at the moment we wrote the check. A year later, when something is breaking, we can go back and find out whether we saw it coming and approved it anyway. That is an uncomfortable file to keep. It is also the only way a committee ever actually learns anything.
What experience taught me about this
I built my first company and ran it straight through 2008 and 2009. Nothing in the good years taught me what that stretch did. You learn very fast that the businesses that make it are not the cleverest ones. They are the ones with enough room to be wrong for a while.
I have also walked into a software company that needed a turnaround, with the plan already written and the clock already running. The thing that saved that situation was not a brilliant new strategy. It was that the business could survive long enough for the work to matter. Time is the resource every operating plan assumes and no operating plan protects.
That is why, at Bambu, we start with survivability and work up. Not because we are afraid of risk, but because we would like to still be underwriting deals in twenty years, with the same investors, out of a bigger fund.
You do not need every deal to be a home run. You need to not own the deal that ends the career.
Over a twenty year horizon, even a modest annual probability of catastrophic loss becomes the dominant variable in your results. Not your sourcing. Not your operating playbook. Not your thesis. That one number.
Protect the capital.
Compound the capital.
Stay in the game.
Everything else comes after that.