EBITDA multiples and what drives them

What EBITDA is, and what it hides

EBITDA is earnings before interest, tax, depreciation and amortisation. Stripping those four out lets buyers compare businesses independently of how they are financed, where they are taxed, and how aggressively they depreciate assets.

That comparability is genuinely useful and it comes at a cost. Depreciation is excluded, but the assets still wear out and will need replacing. Interest is excluded, but the debt still has to be serviced. A business with heavy capital requirements and a business with almost none can show identical EBITDA while generating very different cash.

Sophisticated buyers know this and adjust for it, which is why capital-intensive businesses attract lower multiples than asset-light ones with the same EBITDA. The multiple is doing the work that EBITDA itself omits.

Why one sector supports several different multiples

Published sector ranges are useful for orientation and misleading if taken as entitlement. Within any sector, the spread between the weakest and strongest business is usually wider than the gap between sectors.

Size is the most reliable driver. Larger businesses consistently attract higher multiples for the same earnings, because they are less fragile, have deeper management, and open up a bigger population of buyers — including funds with minimum cheque sizes. This is the size premium, and it is the single clearest pattern in mid-market transaction data.

Growth raises the multiple because the buyer is paying a number of years of current earnings for a stream they expect to be larger. Margin quality and stability matter more than headline margin: a steady twelve percent beats a volatile twenty. Recurring revenue commands a premium over project revenue, sometimes a very large one, because it lowers the uncertainty of year eleven.

What pushes a multiple down

The discounts are more predictable than the premiums, and most are fixable given time.

Customer concentration is the most common. One customer at forty percent of revenue is not a business with strong sales; it is a business with a single point of failure, and buyers price it that way.

Owner dependence — if the relationships, the pricing judgement and the technical knowledge sit with one person who intends to leave, the buyer is acquiring less than it appears. The founder-dependence discount covers this in full.

Poor financial information. Unaudited accounts, management figures that do not reconcile, no reliable monthly reporting. Buyers price uncertainty, and the discount for opacity is usually larger than the cost of fixing it.

Regulatory or single-supplier exposure, deferred capital expenditure that the buyer will have to fund, and unresolved legal or tax contingencies all attract discounts or, more often, an indemnity or escrow rather than a price cut.

How the multiple is actually applied

The arithmetic looks simple and the inputs are where the disagreement lives. Normalised EBITDA multiplied by the multiple gives enterprise value; subtract debt and add surplus cash to reach equity value.

Worked through: a business with reported EBITDA of R12m, normalised to R14m once owner salary and one-off items are corrected, at a multiple of 5.5 gives an enterprise value of R77m. Carrying R18m of debt and R3m of surplus cash, equity value is R62m.

Notice what moved. The R2m normalisation adjustment added R11m to enterprise value at that multiple — which is why the earnings work matters more than most owners expect, and why it should be done and documented before a buyer does it for you.

Multiples other than EBITDA

EBITDA dominates mid-market conversation but it is not always the right base, and using it where it does not fit produces nonsense.

Revenue multiples are used where profit is suppressed by deliberate investment or where the business is too young for earnings to mean much. They are common in software and early-stage technology, and they are dangerous elsewhere: revenue says nothing about whether the business can convert it to cash.

EBIT multiples deduct depreciation and therefore respect the fact that assets wear out. For capital-intensive businesses, EBIT is frequently the fairer base and produces a more honest comparison across differently equipped competitors.

Earnings or P/E multiples work after interest and tax, which makes them sensitive to how the business happens to be financed — useful for a shareholder comparing returns, less useful for comparing operating businesses.

Sector-specific bases exist where the economics warrant them: per bed in healthcare, per subscriber in telecommunications, per hectare in agriculture. These are shorthand for capacity, and a buyer will still test them against cash generation before paying.

Using multiples honestly

Three habits prevent most of the disappointment we see.

Ask what the comparable transactions actually were. A multiple drawn from listed companies is not applicable to a private business without a discount for illiquidity and scale. A multiple drawn from a competitor's sale may reflect a strategic premium that no financial buyer will repeat.

Ask which EBITDA the multiple is applied to — reported, normalised, trailing twelve months, or forecast. The same headline multiple on a forecast figure and a trailing figure are different offers.

And treat a range as a range. An owner anchored on the top of a published band, negotiating against a buyer anchored on the bottom, usually reaches the same middle as everyone else — having spent three months and considerable goodwill getting there.