How to Build an LBO Model That Actually Works

A leveraged buyout is a simple idea wrapped in a model that looks more complicated than it needs to be. The idea: buy a company mostly with borrowed money, use the company’s own cash flow to pay that debt down over several years, then sell the company. Because so little of the purchase was actual equity, even a modest improvement in the business — or just paying down debt — can produce a large return on the equity that was put in.

The model that supports this idea has four real pieces: how the deal is financed, how the debt gets paid down over time, what the company is worth at exit, and what return that produces on the equity invested. Most of the complexity people run into comes from one specific mechanical wrinkle in the second piece, which is where this piece will spend the most time.

What a Basic LBO Deal Looks Like

Before the mechanics, it’s worth seeing the shape of a simple deal, since every number in the rest of this article is just a more detailed version of this picture.

Say a company is worth $100M, based on its cash flow and what similar companies have sold for. Instead of one buyer writing a $100M check, the deal gets financed like this:

• Senior debt: $50M (50%) — the largest, safest layer, usually from a bank, at the lowest interest rate, and first in line to get paid back if anything goes wrong.
• Subordinated (mezzanine) debt: $20M (20%) — a smaller layer, riskier for the lender than senior debt, so it carries a higher interest rate, and gets paid back only after the senior debt.
• Equity: $30M (30%) — the buyer’s own cash. This is what’s actually at risk, and what the return at the end is measured against.

That’s the entire anatomy of a basic LBO: a large layer of debt, sometimes a second smaller layer of debt at a higher rate, and an equity check that’s deliberately kept small relative to the total price. A $100M company gets bought with $30M of real money. The other $70M is borrowed, and the rest of this article is really just: how does that $70M get paid down, and what does the $30M turn into by the time the company is sold.

Piece One: Sources and Uses

Before any modeling of the business itself, a buyout starts with a simple accounting question: where is the purchase price coming from, and where is it going?

Uses — what the money is needed for:

• Purchase price for the company (usually expressed as a multiple of EBITDA)
• Transaction fees (legal, advisory, financing fees)
• Sometimes, refinancing existing debt the company already carries

Sources — where that money comes from:

• Debt (usually several layers — senior debt, sometimes subordinated or mezzanine debt, each with different interest rates and repayment priority)
• Equity (the buyer’s own cash, contributed to cover whatever the debt doesn’t)

The sources have to equal the uses exactly. This table is the starting point of every LBO model, because everything downstream — the debt schedule, the interest expense, the equity return — depends on how much debt was used and how much equity was actually put in.

Example Sources & Uses, for a company being bought at $80M:

Uses	Amount
Purchase price (8x EBITDA of $10M)	$80.0M
Transaction fees	$2.0M
Total Uses	$82.0M

Sources	Amount
Senior debt (4.0x EBITDA)	$40.0M
Subordinated debt (1.5x EBITDA)	$15.0M
Equity (the plug — whatever’s left)	$27.0M
Total Sources	$82.0M

Debt here totals $55M against $27M of equity — roughly a 67/33 debt-to-equity split. That ratio is the entire premise of an LBO: the buyer is putting in $27M of actual cash to control an $80M company, and the debt is doing the rest of the work.

Piece Two: The Debt Schedule — and the Circularity That Trips People Up

This is the part that actually determines whether a model works or breaks.

Each year, the company generates cash flow. Some of that cash flow is used to pay interest on the debt. Whatever’s left over (called the cash flow available for debt paydown) is used to pay down the principal early — this is often called a cash flow sweep, because it “sweeps” available cash straight into debt reduction rather than letting it sit as cash on the balance sheet.

Here’s the wrinkle: interest expense in a given year depends on how much debt is outstanding. But how much debt gets paid down this year depends on how much cash is left over after paying interest. And how much cash is left over depends on interest expense. Interest determines paydown; paydown determines next year’s interest; next year’s interest depends on this year’s paydown.

That’s a circular reference — the model is asking two numbers to each depend on the other at the same time. This is normal and expected in a real LBO model, not a mistake. It’s usually resolved one of two ways:

• Iterative calculation — letting the spreadsheet recalculate repeatedly until the numbers settle (most spreadsheet software has a setting to allow this, since by default it treats circular references as an error)
• Average balance method — calculating interest on the average of the beginning and ending debt balance for the year, which avoids a true circular reference while still being a reasonably accurate approximation

People trying to build their first LBO model usually hit a wall here, either because their spreadsheet throws a circular reference error, or because they didn’t realize the interest and paydown calculations needed to reference each other in the first place. Knowing this is coming, and picking one of the two methods deliberately, is most of what separates a model that works from one that doesn’t.

A simplified one-year debt schedule:

	Senior Debt
Beginning balance	$40.0M
Interest rate	6.0%
Interest expense	$2.4M
Cash available for paydown (illustrative)	$5.0M
Principal paydown	$5.0M
Ending balance	$35.0M

Repeat this, year over year, for every layer of debt, and the ending balance in the final year of the hold period is what gets used to calculate the return at exit.

Piece Three: The Exit

At the end of the hold period — commonly three to seven years — the company is assumed to be sold. The exit value is usually calculated the same way the entry price was: a multiple of EBITDA at that point.

Exit Enterprise Value = Exit-Year EBITDA × Exit Multiple

From that enterprise value, subtract whatever debt is still outstanding at that point (from the debt schedule above) to get the equity value at exit:

Exit Equity Value = Exit Enterprise Value − Remaining Debt

This is the step where an important assumption quietly does a lot of work: what exit multiple to use. Using the same multiple paid at entry is the conservative, defensible default. Assuming the exit multiple will be higher than the entry multiple — called multiple expansion — is a real phenomenon in some deals, but it’s also the easiest place for an overly optimistic model to manufacture a return that didn’t actually come from operating the business better. A model that only works because of assumed multiple expansion is a model built on hope, not on the mechanics above.

Piece Four: The Return

With the equity invested at entry and the equity value at exit, the return is calculated two ways, and both matter.

MOIC (Multiple on Invested Capital) — the simple version:

MOIC = Exit Equity Value / Entry Equity Invested

IRR (Internal Rate of Return) — the version that accounts for time, since a return earned in three years is worth more than the same return earned in seven:

IRR = (Exit Equity Value / Entry Equity Invested)^(1 / Number of Years) − 1

Worked Example

Figures below are illustrative, built to walk through the mechanics — not a real transaction or sourced deal data.

Entry (from the Sources & Uses above):

• Entry EBITDA: $10.0M
• Entry multiple: 8.0x
• Equity invested: $27.0M
• Total debt at entry: $55.0M

Over a 5-year hold, assume:

• EBITDA grows from $10.0M to $14.0M (modest, realistic operating improvement)
• Debt is paid down from $55.0M to $20.0M through the cash flow sweep described above
• Exit multiple: 8.0x (same as entry — no multiple expansion assumed)

Exit calculation:

Exit Enterprise Value = $14.0M × 8.0x = $112.0M

Exit Equity Value = $112.0M − $20.0M = $92.0M

Return calculation:

MOIC = $92.0M / $27.0M ≈ 3.4x

IRR = (3.4)^(1/5) − 1 ≈ 27.7%

Reading it: the equity invested more than tripled over five years, and none of it came from assuming the business would sell for a richer multiple than it was bought for. It came from two unglamorous sources: the business grew EBITDA modestly, and debt got paid down using the company’s own cash flow. That’s what a real LBO return usually looks like when the model isn’t quietly leaning on an optimistic exit assumption to do the work.

Why Models Break

Three mistakes account for most broken LBO models, and none of them are exotic:

Ignoring the interest-paydown circularity, either by hardcoding an interest number that doesn’t actually reflect the declining debt balance, or by letting a spreadsheet’s circular reference error silently produce a wrong number instead of resolving it deliberately.

Assuming exit multiple expansion without justifying it. If the return only works because the exit multiple is assumed higher than entry, the model isn’t describing a real value-creation story — it’s assuming the market will simply pay more for the same business later, for no stated reason.

Overly optimistic EBITDA growth or margin assumptions, carried through every year of the hold period without being stress-tested against a more conservative case. A model with only one growth scenario isn’t really a model — it’s a forecast wearing a model’s clothes.

Explain It Like I’m Four

Imagine you want to buy a house that costs $100. You only have $30 saved up, so you borrow the other $70 from the bank.

Every month, you have to pay some of that $70 back, plus a little extra called interest — the fee the bank charges for lending you the money. Here’s the tricky part: how much interest you owe depends on how much you still owe the bank. But how much of the loan you pay off each month depends on how much money you have left over after paying that interest. Those two things go back and forth, depending on each other, every single month. That’s the confusing part that trips people up — it’s not a mistake, it’s just how the math actually works, and you have to handle it on purpose instead of guessing.

A few years later, you sell the house. Let’s say the house is now worth $140 — maybe you fixed it up a little, or houses in the neighborhood just got more valuable. By then, you’ve also paid the bank back down to $40 still owed.

So when you sell:

You get $140 for the house.

You pay the bank back the $40 you still owe them.

You keep $100.

You only put in $30 of your own money to start with, and you walked away with $100. That’s a much bigger jump than if you’d bought the house entirely with your own $100 in cash and sold it for $140 — same house, same price increase, but because you only used $30 of your own money, your own money grew a lot faster.

That’s the whole idea behind a leveraged buyout. Buy something mostly with borrowed money. Pay the loan down a little each year. Sell it later. Because so little of the original purchase was your own money, even a modest gain in the value of the thing turns into a much bigger gain on the small amount you actually put in.

One sentence version: borrowing most of the purchase price means even a small improvement in what you bought turns into a much bigger return on the little bit of your own money you actually risked.

What’s Next

The mechanics above are the whole skeleton of a real model — sources and uses, the debt schedule, the exit, the return. Two natural follow-ups sit on top of this: what happens to the return when the operating assumptions turn out to be wrong (sensitivity and downside cases), and where the leverage in this structure actually creates risk rather than just return — which connects back to the same power law and information-edge thinking running through the rest of this site.

Where in your own numbers — pricing, debt, growth assumptions — are you quietly assuming the optimistic case without stress-testing it?

About Me

I'm Michael Philippou, co-founder of Big Love, a plant-based ice cream business in Santa Monica that I've been running with my wife Victoria for over ten years. Before ice cream, I was a lawyer. I write about the real, unpolished lessons of running a small business — no gurus, no hype, just what's actually worked and what hasn't. Watch more on YouTube (Real Business Real Lessons) or follow along on LinkedIn.