SDE vs EBITDA: Which Number Matters for Your Deal?

Two businesses with identical operations can be marketed at very different numbers depending on which earnings metric the broker chose. Neither number is wrong. They measure different things.

What each one is

Seller's discretionary earnings is what the business generates for one working owner. It starts at net income and adds back interest, taxes, depreciation, amortization, one owner's total compensation and benefits, and discretionary or non-recurring costs.

The logic is that the buyer is stepping into the owner's chair. Whatever the current owner extracted, in salary, in perks, in profit, is available to the buyer who does the same job.

EBITDA is earnings before interest, taxes, depreciation and amortization, and it does not add back owner compensation. It assumes the business pays a market-rate manager to run it, and it measures what is left after that.

The gap between them is roughly one market-rate salary. On a business doing $1.2 million of revenue, that gap is frequently $100,000 to $150,000, which at a 3.5x multiple is $350,000 to $525,000 of apparent value.

Where the line falls

There is no official threshold, but in practice:

• Businesses under roughly $1 million of earnings, sold to individual owner-operators, are almost always quoted in SDE.
• Businesses above roughly $2 million of earnings, sold to private equity or strategic buyers, are quoted in EBITDA, because the buyer is not going to run it personally.
• Between those, either can appear, and you have to read the recast carefully to know which you are looking at.

Most SBA-financed acquisitions sit squarely in SDE territory. The buyer is going to work in the business, and the lender is underwriting the buyer's ability to service debt and take a living out of the same cash flow.

The mistake that actually costs money

It is not choosing the wrong metric. It is pairing a metric with a multiple that belongs to the other one.

EBITDA multiples are higher than SDE multiples for the same business, because EBITDA is a smaller number. If someone applies a 5x EBITDA multiple to an SDE figure, the result is not aggressive, it is meaningless. The same thing happens in reverse when a 3x SDE multiple is applied to EBITDA and the seller is talked out of real value.

Whenever you see a valuation, confirm three things: which metric it is built on, whether owner compensation was added back, and whether the multiple came from comparable deals quoted on that same metric.

What a QoE does to both numbers

Neither metric means much until the inputs are verified. In a Quality of Earnings review we rebuild the number from source rather than accepting the recast, which usually means:

• Tying reported revenue to bank deposits, and reconciling the books to filed tax returns.
• Testing each add-back against documentation instead of description.
• Separating historical expense from post-close assumption, so the market adjustment is visible rather than netted.
• Correcting cash-basis timing that flatters or depresses a period.
• Producing a trailing twelve month view, not just fiscal years, because a deal closing in September should not be priced off last December.

On a business where the owner does the bookkeeping, adjustments of 10 to 20 percent from the marketed figure are ordinary. Not because anyone lied, but because books built for a tax return are not built to be sold from.

The practical version

If you are buying a business you intend to run: think in SDE, and subtract what you would have to pay someone to do your job if you ever stopped. That number is closer to what the business is worth to the next buyer.

If you are buying a business you intend to hire a manager for: think in EBITDA, and make sure the manager's full cost is in the model before you apply a multiple.

If you are selling: pick the metric your buyer pool actually uses, present it consistently, and be ready to support every adjustment. A number that survives scrutiny is worth more than a bigger number that does not.