Same EBITDA. Eleven Million Apart.

Two businesses in the same industry. Both did $3.2m of EBITDA last year. One is worth $12.8m and the other $24.0m, and the owners of the first one have no idea.

Nothing in either set of accounts explains the difference. The profit is identical. You have to look at what a buyer is being asked to believe.

A multiple is the price of uncertainty

Enterprise value is the market’s confidence that future cash flows will arrive. Everything else is evidence for or against that proposition. Customer concentration reduces confidence. Owner dependence reduces confidence. Slow or unreliable reporting reduces confidence. Unknown liabilities destroy it.

So a multiple is not a measure of quality. It is the price of uncertainty. Company A carries one customer at 38% of revenue, a founder who still signs off on pricing, and management accounts that arrive six weeks late and get restated. Company B has no customer above 14%, a general manager running it, most of its revenue under contract, and accounts that close in five days. Same profit. A buyer prices A at 4.0x and B at 7.5x, and the $11.2m between them is the whole of the difference.

The number that makes it real

Flip each multiple over and you get the earnings yield: 25.0% for Company A, 13.3% for Company B. That is 11.7 percentage points a year, on the same earnings in the same industry. Nobody demands 25% a year from an asset they are comfortable with. They demand it from one they think might break.

None of the work that closes that gap improves this year’s EBITDA by a dollar. It takes two or three years, and it has to be evidenced before anyone comes looking, which is why it keeps not getting done.

We have laid the full example out driver by driver, including the table a buyer would actually build and the one risk that breaks the model entirely.

Read the full worked example →

Or put your own business through it: the Enterprise Value Gap Calculator estimates your value across fifteen drivers and shows the discount you may be carrying.