EBITDA stands for Earnings Before Interest, Taxes, Depreciation, and Amortization. In M&A, it measures a business's operating performance independent of its capital structure, tax situation, and accounting choices. It is the most common starting point for valuing a business in the lower middle market, typically expressed as a multiple (e.g., 5x EBITDA). It is not a cash flow metric and should not be treated as one.
Everyone's Favorite Punching Bag
Charlie Munger had a ... ahem ... colorful name for it. Warren Buffett said not thinking about depreciation as a real expense is "absolutely crazy." Munger went further and suggested investors just replace the word EBITDA with "nonsense earnings" whenever they see it.
And now, apparently, everyone on X agrees.
The problem is, Buffett and Munger were talking about a specific context: Large, capital-intensive public companies where depreciation represents real, recurring cash that has to go back into the business just to stay competitive. Railroads. Manufacturing plants. Telecom infrastructure. Places where you are, in Buffett's words, spending the cash first and recording the expense later.
They weren't talking about a 12-person accounting firm in Des Moines doing $3.2M in revenue with two desks, a printer, and a software subscription. Context matters.
EBITDA isn't the problem. Using it without understanding what it leaves out is.
I've spent years doing buy-side quality of earnings work across the deal spectrum, from main street businesses doing $250k in EBITDA to middle market companies pushing $25M. The acquirers look different, search fund buyers and individual operators on one end, PE firms and independent sponsors on the other, but the question is the same: What does this business actually earn, and can you rely on it? In that world, EBITDA is not a dirty word. It is, used correctly, a very useful one.
So let me walk through what it actually is, what each exclusion is actually doing, and where it earns its keep ... and where it doesn't.
What EBITDA Is (and Isn't)
EBITDA is not a GAAP figure. Your accountant isn't required to report it, and it can be calculated slightly differently depending on who's doing the math. That lack of standardization is a real criticism and we'll come back to it.
What EBITDA is: A proxy for operating cash generation before the effects of financing decisions, tax strategy, and non-cash accounting charges. It answers a narrow but genuinely useful question: How much money does this business make from its operations, before we layer on how it's been financed and how aggressive its accounting has been?
In an M&A context, that question matters because a buyer is not acquiring the seller's debt load, their personal tax situation, or their depreciation schedule. They're acquiring the engine. EBITDA tries to measure the engine.
Now let's take each piece apart.
Why We Back Out Each Component
Earnings Before Interest
Two businesses can have identical operations and wildly different interest expense. One owner bought the business with mostly equity. The other used a ton of seller financing. Strip out interest and you can compare the underlying business performance without the noise of how each deal was capitalized.
In a sale process, the buyer is going to recapitalize the business anyway. They'll bring their own financing, their own cost of capital, their own debt covenants. The seller's interest expense tells you almost nothing about what the business will look like post-close. So we remove it.
Earnings Before Taxes
Tax burden varies dramatically depending on entity structure, jurisdiction, owner elections, and carryforward positions. An S-corp owner in Texas has a very different tax line than a C-corp in California. Neither tells you much about the core operating performance of the business.
More practically: In a deal, the structure often changes. An asset purchase, a stock deal, an F-reorganization ... they all have different tax implications. Taxes are a downstream decision from how the deal is structured. So we remove them from the operating picture.
Earnings Before Depreciation and Amortization
This is where Buffett and Munger have a real point, and where the most honest answer is: It depends.
Depreciation is an accounting mechanism that spreads the cost of a past capital expenditure over its useful life. You bought a $500,000 machine five years ago. You're charging $100,000 per year against income. The cash left the building years ago, but the income statement keeps absorbing the hit. In that sense, adding depreciation back produces a number closer to actual cash generation.
Buffett's counterargument: That machine eventually wears out. You have to replace it. So depreciation is just a delayed recording of a real future cash obligation. He's right about that too.
Both things can be true. Depreciation is a non-cash charge today and a real future cash obligation. The question is how material it is to this specific business.
For an asset-light service business, D&A might be a few thousand dollars a year. Adding it back doesn't distort anything meaningful. For a trucking company with a fleet that depreciates and then has to be replaced, D&A is telling you something real and important. Ignoring it is a mistake.
Amortization adds another layer. Post-acquisition, a business often carries amortization of intangible assets acquired in prior deals: Customer lists, non-competes, trade names. These are almost always added back in EBITDA calculations because they're accounting artifacts of past transactions, not the business's own operations. Stripping them out is generally defensible.
The discipline here is knowing which type of D&A you're dealing with, and adjusting your analysis accordingly.
Where People Actually Go Wrong
The dunks on EBITDA are usually aimed at a different sin than the metric itself. Here is the real list of mistakes:
1. Mistaking EBITDA for free cash flow.
EBITDA ignores capex (both maintenance and growth), working capital requirements, and debt service. It is not the money you can take out of the business. A company with $2M EBITDA and $1.8M in annual capex requirements is not a $2M cash business.
2. Using seller-reported EBITDA as the final word.
"Adjusted EBITDA" is not EBITDA. Add-backs are where the creative writing happens. One-time expenses that recur annually. Owner compensation that understates the real cost of management. "Non-recurring" legal costs for a business that's been in litigation three years running. The quality of the earnings underneath the number matters enormously.
3. Applying a multiple without understanding the multiple.
A 6x EBITDA multiple on a stable, recurring revenue services business is a very different risk profile than 6x on a project-based business with two major customers. The multiple has to reflect the quality and durability of the earnings, not just the headline number.
4. Ignoring capital structure in the deal math.
EBITDA gets you to enterprise value. You still have to subtract debt, add cash, and account for working capital adjustments to get to equity value. Buyers who forget this step have a bad day at closing.
Where EBITDA Works and Where It Gets Complicated
Here is a practical breakdown across common deal types:
| Industry / Business Type | EBITDA as a Metric |
|---|---|
| Professional services (accounting, consulting, staffing) | Works well Asset-light, low capex, D&A is noise. |
| SaaS / software | Works well Low tangible asset base, recurring revenue. |
| Distribution (stable inventory cycles) | Generally works Watch working capital carefully. |
| Healthcare services (non-facility) | Workable Normalize owner comp and AR carefully. |
| Manufacturing / heavy industrial | Gets complicated Capex is real and recurring. |
| Trucking / logistics | Use with caution Maintenance capex is not optional. |
| Retail with heavy lease obligations | EBITDAR territory Adding rent back becomes the conversation. |
| Capital-intensive infrastructure | Inadequate alone Free cash flow is the real metric here. |
The pattern is straightforward: The more capital-intensive the business, the more you need to look past EBITDA to understand what the business actually requires to operate and grow. That doesn't make EBITDA wrong. It means you have more work to do after you calculate it.
What a Good QoE Actually Does With EBITDA
A quality of earnings analysis doesn't start and stop at the EBITDA line. It starts there.
The job is to validate the number (is the revenue real, recognized correctly, and durable?), normalize it (strip out genuine one-time items, adjust owner comp to market, isolate the true run-rate), and then contextualize it (what capex does this business need to sustain these earnings? What does working capital look like? Are there off-balance-sheet items the buyer should know about?).
At the end of that process, you might end up with a number that looks nothing like what was in the CIM. That's the point. The multiple applied to a clean, normalized, contextualized EBITDA is defensible. The multiple applied to whatever the seller's bookkeeper produced is a bet.
The Bottom Line
Buffett and Munger were right about the abuse. They weren't issuing a blanket indictment of the metric.
EBITDA is a starting point, not a verdict. It strips out the noise that varies across capital structures, tax elections, and accounting choices so you can compare businesses on a level field. That is a real and useful function. It just isn't the only function you need.
The professionals who dismiss it entirely tend to be talking about large-cap public markets where depreciation is enormous and meaningful. The professionals who treat it as gospel tend to be selling something.
The right answer is in the middle, which is also where most good analysis lives.
Know what EBITDA is. Know what it leaves out. Adjust accordingly. And if someone hands you a CIM with an "Adjusted EBITDA" number and no supporting schedule ... ask a lot of questions.
Timothy is the founder of QoEPro.com, a quality of earnings firm serving PE firms, independent sponsors, search fund acquirers, and individual buyers across the deal spectrum. View report options →
Frequently Asked Questions
Is EBITDA the same as cash flow?
No. EBITDA does not account for capital expenditures, changes in working capital, or debt service. It is a measure of operating earnings, not what you can take out of the business. For capital-intensive businesses, the gap between EBITDA and free cash flow can be substantial.
Why do private equity firms use EBITDA multiples to value businesses?
EBITDA multiples allow PE firms to compare businesses across different capital structures, tax situations, and geographies on a consistent basis. Because a buyer will recapitalize the business anyway, the seller's interest expense and tax rate are largely irrelevant to what the business is worth to a new owner. EBITDA strips those out so the comparison is apples-to-apples.
What is "Adjusted EBITDA" and should I trust it?
Adjusted EBITDA is EBITDA with additional items added back or removed, typically presented by sellers to show the "true" earning power of the business. Common add-backs include owner compensation above market rate, one-time professional fees, and non-recurring expenses. Adjusted EBITDA should be scrutinized carefully. A quality of earnings analysis exists specifically to validate or challenge these adjustments.
When is EBITDA not a reliable valuation metric?
EBITDA becomes less reliable when capital expenditures are large and recurring (manufacturing, trucking, infrastructure), when the business has significant lease obligations that a buyer will inherit (retail, restaurant), or when working capital requirements are high and volatile (project-based businesses, construction). In these cases, free cash flow or sector-specific metrics like EBITDAR are more informative.
What is a quality of earnings report and how does it relate to EBITDA?
A quality of earnings (QoE) report is a financial due diligence analysis, typically commissioned by a buyer, that validates the sustainability and accuracy of a business's reported earnings. It starts with EBITDA, normalizes it for one-time and non-recurring items, adjusts owner compensation to market, and then contextualizes it with capex requirements, working capital trends, and revenue quality analysis. The result is a defensible, independently verified earnings figure that supports the purchase price negotiation.