Turn the customer into profit, weight the two outcomes, then find the price at which the loan still gets paid if the customer leaves.
Customer concentration reduces value through profit, not revenue: price the business on its EBITDA with and without the customer, weight the two by the chance the customer leaves, then check that the loan survives the loss. A customer worth 30 per cent of revenue carries 67.5 per cent of a 600,000 EBITDA; at a 25 per cent chance of losing it, a 2,100,000 asking price becomes 1,785,000, a 15.0 per cent discount. And the price at which the debt still gets paid without that customer is lower again: 1,646,884.
Worked in full in Buying a Small Business by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Brokers and buyers often talk about concentration as a percentage of revenue and a haircut on the multiple. Both understate the problem. The lost customer takes its whole contribution with it, while most of the fixed costs stay behind, and the loan payments do not change at all.
| Input | Value |
|---|---|
| Revenue | 3,000,000 |
| Largest customer's revenue (30%) | 900,000 |
| Contribution margin on that customer (revenue less labour, materials, direct costs) | 45% |
| Fixed costs avoidable if it leaves (a dedicated supervisor) | 45,000 |
| Adjusted EBITDA (20% margin) | 600,000 |
| Asking multiple of EBITDA | 3.5x |
| Probability the customer does not renew at its year-two renewal | 25% |
| Debt: 90% of price, 10.5% over 10 years, monthly | annual constant 16.19% |
Step 1: the customer's share of profit. Its contribution is 900,000 times 45 per cent, 405,000: 67.5 per cent of EBITDA from 30 per cent of revenue. That is the number to put in front of a lender, not the 30.
Step 2: EBITDA without the customer. Remove the contribution and add back only the fixed costs that genuinely go with it.
EBITDA without the customer = 600,000 − 405,000 + 45,000 = 240,000
The business loses 60.0 per cent of its profit. At the same 3.5x it would be worth 840,000.
Step 3: the probability-weighted price.
Weighted EBITDA = 75% × 600,000 + 25% × 240,000 = 510,000
Price = 510,000 × 3.5 = 1,785,000, a discount of 15.0% (315,000)
Excel: =(1-pLoss)*EBITDA+pLoss*(EBITDA-CustRev*CM+Avoid), times the multiple.
This simplification treats the loss as permanent from day one; with the renewal two years out, the buyer keeps the profit for two years first, which a full cash flow model would credit. It does not change the next step, which is the one that ruins buyers.
Step 4: does the loan survive the loss? At the asking price the buyer borrows 1,890,000 and pays 306,033 a year.
| Case | EBITDA | Debt service | Cover |
|---|---|---|---|
| Customer renews | 600,000 | 306,033 | 1.96x |
| Customer leaves | 240,000 | 306,033 | 0.78x |
Without the customer the buyer funds 66,033 a year from personal cash, before capital expenditure and tax, on a loan they have personally guaranteed. The price at which the loss case still covers debt service exactly once:
Survivable loan = 240,000 / 16.19% = 1,482,195; survivable price = 1,482,195 / 90% = 1,646,884, or 2.74x the full EBITDA
Three prices for one business: 2,100,000 asked; 1,785,000 on the expected profit; 1,646,884 if the buyer wants to be able to pay the bank without the customer. A lender sizing on the full 600,000 at 1.25x would lend against a price as high as 3,293,767, so bank approval is no comfort: the bank's test is on the year it sees, not the year after the renewal.
The gap is a structuring problem as much as a price problem. Paying 1,785,000 at closing and 315,000 contingent on the customer renewing gives the seller the full 2,100,000 if the seller is right about the relationship, and means the buyer pays nothing for the lost profit if the seller is wrong. It does not by itself make the loan safe: debt sized on a 1,785,000 closing payment still exceeds the loss case's capacity, so the closing payment has to come down towards 1,646,884, or the contingent part has to grow, for the buyer to stay current with the bank without the customer.
| Probability of loss | Weighted EBITDA | Price | Discount |
|---|---|---|---|
| 10% | 564,000 | 1,974,000 | 6.0% |
| 25% | 510,000 | 1,785,000 | 15.0% |
| 40% | 456,000 | 1,596,000 | 24.0% |
| Share of revenue | Share of EBITDA | EBITDA without it | Weighted price | Survivable price |
|---|---|---|---|---|
| 15% | 33.8% | 420,000 | 1,942,500 | 2,882,046 |
| 30% | 67.5% | 240,000 | 1,785,000 | 1,646,884 |
| 45% | 101.2% | 60,000 | 1,627,500 | 411,721 |
At 15 per cent of revenue the survivable price sits above the asking price, so concentration is a pricing question only. At 30 per cent it binds. At 45 per cent the customer carries more than the whole profit, and no leveraged price is safe: the discount on expected profit, 22.5 per cent, badly understates the risk to an indebted buyer.
Cutting the multiple by a rule of thumb on the revenue share, say half a turn for 30 per cent. That gives 3.0x and 1,800,000, close to the weighted price, and it would pass a casual review. But it answers the valuation question and not the financing one: at 1,800,000 the loss case still covers debt service only 0.91x. The concentration test has to be run on the loan as well as on the price; the cover calculation itself is set out in what purchase price a 1.25x DSCR supports on an SBA loan.
Translate the customer into profit, price the business with and without it, and test the debt on the "without" case: 1,785,000 on expected profit and 1,646,884 to stay current with the bank, against 2,100,000 asked. Bridge the difference with a payment contingent on renewal. The free workbook for this case lets you remove a concentrated customer from a full acquisition model with one input.
A common rule of thumb flags any customer above 10 to 15 per cent of revenue, but the share of profit matters more. In the illustrative case a customer at 30 per cent of revenue and a 45 per cent contribution margin is 67.5 per cent of EBITDA, because most fixed costs stay when it leaves. At 45 per cent of revenue it would carry 101.2 per cent of the profit.
Pay less at closing and make the balance contingent on the customer renewing, through an earn-out or a seller note that is forgiven if the customer leaves. Here 1,785,000 at closing and 315,000 on renewal gives the seller the full 2,100,000 if the relationship holds. To keep the bank paid if the customer leaves, the closing payment would need to fall to about 1,646,884, with more of the price contingent.
Because the lender tests cover on current earnings. On 600,000 of EBITDA the business covers debt service 1.96 times at a 2,100,000 price, comfortably above 1.25. Without the customer EBITDA is 240,000 and cover falls to 0.78, leaving the guarantor to fund 66,033 a year.
This article is one calculation from Buying a Small Business. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: corporate finance, valuation and markets → · All 681 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.