Take a service-firm acquisition that was almost over the line. The buyer had made an offer based on a multiple of annual revenue. Due diligence was nearly complete. Then the seller came back with a new pricing logic. The market, the seller said, was now valuing service firms at three to five times EBITDA, and because the buyer would service the acquired book with the buyer's existing resources, the buyer's adjusted EBITDA on the acquired revenue would be roughly equal to the revenue itself. The seller was therefore asking for three to five times the same number on which the original offer had been built.
The firm has seen this attempt enough times to recognise it on arrival. It is not a different valuation. It is the same business, on the same facts, priced with a different method to produce a higher number.
The two methods are not independent. They land at the same number.
The revenue-multiple approach and the EBITDA-multiple approach to valuing a service-firm portfolio are not alternative views of the same business. They are the same view, expressed in two units. The bridge between them is the seller's margin.
Take a portfolio with operating margin X. X is the share of revenue that drops to EBITDA after the cost of running the book. At margin X, the portfolio's EBITDA is X times revenue. A buyer applying an EBITDA multiple of four is therefore paying four times X times revenue. A buyer applying a revenue multiple of 1.5 is paying 1.5 times revenue. For the two to land at the same number, X must be 0.375; that is, an operating margin of about 37.5 per cent. For a small advisory book, an insurance brokerage portfolio or an accounting practice, that is the middle of the realistic range. Asset-light service firms commonly run operating margins of 30 to 50 per cent, and across that band, EBITDA times three to five and revenue times 1.5 produce roughly the same number, often within a tight band of each other.
This is not coincidence. It is the arithmetic of margin. The two methods have been triangulating to the same answer for decades, and they diverge only when the underlying business carries features that one method captures and the other does not. Concentrated client risk. Regulatory overhang. Idiosyncratic working capital. When those features are present, both buyer and seller adjust the inputs. They do not, in either case, switch the method to capture a higher headline.
The valuation methodology is the cross-check, not the lever. Two methods that arrive at different answers on the same business mean the inputs are wrong somewhere, not that the seller gets to pick the higher one.
What the buyer brings and bears, stays with the buyer.
The seller's argument that the buyer's adjusted EBITDA should be applied to the seller's revenue rests on a misreading of what the buyer is offering. The buyer's existing platform, the fixed cost already absorbed by the buyer's other business, the integration capacity, the distribution: these are the buyer's contribution. They are the reason the buyer can pay anything above the seller's standalone economics. They are also the reason the buyer is the buyer.
Standard valuation practice draws a clear line between the standalone value of the target and the synergy value the combination produces. The standalone value is what the seller is entitled to be paid; it is the value the seller has actually built. The synergy value is created by the buyer's choices about how to integrate, who to retain and what to retire. Where synergies are shared with the seller at all, the sharing is not captured by switching valuation methods. It is captured through structures designed for the purpose: an earn-out tied to actual post-close performance, a deferred consideration tied to retention, a shared upside on a defined window. Anything else is an attempt to charge for the synergies without giving up the right to retain them on day two.
The asymmetry runs the other way too. Service-firm clients are not factories. They are relationships, and a book of business survives an acquisition to the extent that the people who maintain those relationships travel with it. The industry data on advisory, insurance brokerage and accounting acquisitions is consistent. A meaningful share of the acquired book rolls off in the first eighteen to thirty-six months after change of ownership. The range varies by sector; the existence of the runoff does not. That runoff is the buyer's risk to bear once the deal is done. So is the cost of running the book afterwards: the variable cost of servicing it, the integration effort, the share of the buyer's fixed overhead that the acquired book now attracts. None of that goes back to the seller. The buyer cannot, having paid, send an invoice for the retention work or the integration cost.
This is the structural reason a buyer-side multiple is what it is. It is not generosity, and it is not strategy. It is the buyer pricing in the risk and the cost that come with the book the moment the keys are handed over. The seller who tries to add the buyer's synergies on top of the seller's standalone economics is asking to be paid twice for the same business: once for what the seller built, and once for what the seller is not contributing.
The price is the price.
The firm's position is plain. A deal that has reached the late stages of due diligence on one valuation method does not get re-opened by switching methods after the question of value has been settled, and an offer that was sound at one multiple does not become unsound because the seller has discovered a different way of arriving at a larger number. Two methods that point to different answers on the same business mean the inputs are wrong somewhere. They do not mean the seller gets to choose the higher one.
What is on the table at that point is a negotiation move dressed up as analysis. It does not move the value of the business. It tests whether the buyer is willing to pay for the same business twice. Most disciplined buyers are not, and the seller who pushes the point past the buyer's patience usually finds that the original offer, the one that was on the table when the question of value was still in good faith, is the price the deal would have closed at if the seller had taken it.
A parallel conviction applies at the other end of the size spectrum. Below a feasibility floor, the conversation is not about which multiple but whether the portfolio is sold at all: the buyer's fixed cost of doing the deal consumes the margin, anchor-client concentration produces binary runoff risk, and long-tail fragmentation makes cost-to-serve uneconomic. The companion piece walks the floor and the alternative transactions the owner gets to choose between (peer-merger, internal MBO, sale to a smaller competitor, managed runoff) once the question shifts from which multiple to which transaction is feasible.