What other EV multiples exist, and when do you use them?
EV/EBITDA is the headliner, but enterprise value pairs with several fundamentals. Each swaps the denominator to suit a different situation, and knowing which to reach for is a practitioner skill in itself.
The EV family
- EV/Sales — enterprise value over revenue. The EV-based cousin of P/S. Because it puts whole-business value over revenue, it fairly compares a debt-laden company with a debt-free one — something plain P/S can't do. Its use case is the same: businesses with little or no profit yet, where you still want to account for debt.
- EV/EBIT — like EV/EBITDA but after depreciation and amortisation. By leaving those charges in, it respects the real cost of assets wearing out, so it's stricter than EV/EBITDA for asset-heavy firms. Analysts who distrust EBITDA's add-backs often prefer it.
- EV/EBITDA — the balance point you already met, sitting between the two above.
Matching the multiple to the moment
A rough ladder from "most forgiving" to "most demanding" denominator:
- No profit, only revenue → EV/Sales.
- Operating profit, but heavy or distorting depreciation → EV/EBITDA.
- Operating profit, and you want the real cost of assets counted → EV/EBIT.
Moving down the ladder, each multiple asks the business to clear a higher bar, so the denominator shrinks and the multiple's number grows even though nothing about the company changed. That's expected — a company might show EV/Sales of 3, EV/EBITDA of 12, and EV/EBIT of 18 all at once. They aren't contradicting each other; they're measuring price against progressively stricter definitions of profit.
Why bother with the whole family
Because a single multiple can flatter or punish a company for reasons that have nothing to do with its quality. Looking at two or three EV multiples together is a cross-check: if a company looks cheap on EV/Sales but expensive on EV/EBIT, that gap is telling you its margins are thin — an observation you'd have missed with one number.
In the data
Two of the three are published, EV/Sales (shown as EV to revenue) and EV/EBITDA; the third you build:
EV/EBIT is enterprise value over EBIT from the income statement, and the EBIT line needs checking before you divide. In this data EBIT is measured at the pre-tax line, so for Apple's 2025 fiscal year it equals income before tax and sits slightly below operating income, with other income and expense in between. Two defensible "EV/EBIT" figures therefore exist for one company, depending on which line you called EBIT.
Try it now
- Note EV/Sales and EV/EBITDA in the table above. There is no EV/EBIT.
- Build it. Take enterprise value from the table above and divide by EBIT from the income statement:
Before dividing, find EBIT, operating income and income before tax in that same table and put them in order. Two of the three are equal and one is not, and which pair matches tells you where this filer's EBIT actually sits. Then compute the multiple against both candidates: the difference between your two answers is the whole of the ambiguity the section above describes, priced. 3. Line the three up: EV/Sales, EV/EBITDA, EV/EBIT. Confirm they rise as the denominator gets stricter, and that this is expected arithmetic rather than a data error — the company did not change, only the bar it had to clear. 4. Now a thin-margin business, where EV/Sales looks low but EV/EBIT looks high. The published multiples:
And the denominator for the one that is not published:
Divide enterprise value by EBIT, then line up EV/Sales, EV/EBITDA and your EV/EBIT beside the first company's three. The gap between the grocer's first and last is a margin story hiding in plain sight, and the operating margin in the first table says how thin.