Restaurant EBITDA multiples: why recovered cost is the cheapest enterprise value

What sets the multiple

Restaurant businesses trade on adjusted EBITDA, and the multiple reflects scale, concept durability, unit-level consistency, franchise mix, remaining lease and franchise agreement terms, and the credibility of the adjustments. Small single-concept operators typically transact at low-to-mid single-digit multiples; larger platforms and asset-light franchisors command materially more.

The multiple is not a lever an operator controls in the near term. The EBITDA it multiplies is.

The arithmetic of recovered cost

Recurring cost recovery flows to EBITDA at one hundred percent. At a six-times multiple, an annualized $400K of recovered spend is roughly $2.4M of enterprise value, created without capital, price increases, or new units. The equivalent EBITDA from sales growth requires several million dollars of incremental revenue plus the labor and marketing to support it.

This is why sponsors run procurement and payments workstreams in the first hundred days. It is the only value creation lever that is both fast and fully within management's control.

Make the recovery diligence-grade

Buyers discount savings they cannot verify. A recovery survives diligence when three things exist: the executed contract or amendment that produced it, before-and-after statements from the vendor showing the new rate in effect, and a run-rate calculation tied to actual volume rather than projected volume.

One-time credits, retroactive refunds, and implementation rebates are real cash but are not run-rate EBITDA. Separate them in your own reporting before a buyer separates them for you, because a quality-of-earnings adjustment you disclosed reads very differently from one they discovered.

Where it fits in a hold period

Cost structure work is front-loaded: it compounds across the whole hold and it improves the base that every later initiative runs through. Sales initiatives, remodels, and development are slower, capital-intensive, and less certain. Doing them in that order is not a preference — it is the sequence that maximizes the exit number.