A searcher finds a business listed at $1.8 million with seller's discretionary earnings around $380,000. A purchase multiple under 5x. Annual debt service on the planned SBA loan comes to about $262,000.
Quick math: $380K over $262K is roughly 1.45x coverage. The deal works. The LOI goes out.
Six weeks later, the lender declines the file. The underwritten DSCR came back at 0.86x.
Not because of the industry. Not because of the buyer's resume. Not because the seller changed the story.
Same deal. Same price. Same lender. Different numerator.
Two Formulas, One Decision
Buyer math:
Listing SDE ÷ annual debt service
Lender-style math:
Lender-style DSCR = (Verified cash flow − buyer salary − capex reserve − unsupported add-backs − required payments on non-standby debt) ÷ annual debt service
The second number is the one that decides whether the deal gets financed. Everything in this article is about finding that gap before you sign anything.
What the SBA Actually Requires
SOP 50 10 8, the SBA lender rulebook effective June 1, 2025, sets the floor for standard 7(a) loans: the business must show at least 1.15x DSCR on historical or projected cash flow. For change-of-ownership deals that rely on projections, those projections generally need to support coverage within roughly two years of funding. Lenders also run global cash flow, which means your personal income and obligations matter too.
But 1.15x is the floor, not the target. Many lenders underwrite closer to 1.25x, and conservative shops want more when a deal carries customer concentration, thin margins, cyclicality, messy add-backs, or heavy operational risk. Thresholds vary by bank, which is exactly why you want the number in hand before the bank runs it.
So the question is not, "Does the listing show enough SDE?"
The question is, "After the lender normalizes the cash flow, does this deal still clear the bank's actual DSCR threshold?"
The Listing DSCR Is Not the Lender's DSCR
Walk a cleaner deal through both formulas. Purchase price of $1.5 million, verified earnings of $420,000, an SBA loan of about $1.35 million after equity injection, and annual debt service near $219,000 on a 10-year term.
Line item | Buyer's quick math | Lender-style math |
|---|---|---|
Seller/listing earnings | $420,000 | $420,000 |
Buyer salary | $0 | ($120,000) |
Capex reserve | $0 | ($10,000) |
Cash available for debt service | $420,000 | $290,000 |
Annual debt service | $219,000 | $219,000 |
DSCR | 1.92x | 1.32x |
This is the same business, same purchase price, and same debt. The buyer counted all SDE as debt-service capacity. The lender did not.
Note that this deal still works. 1.32x clears most credit boxes. The point is not that lender-style math kills deals. The point is that 1.92x was never the real number, and the buyer who knows that can underwrite price, structure, and salary with open eyes.
The Salary Line Is Where Good-Looking Deals Break
The most common mistake in searcher models is treating SDE like free cash flow.
SDE adds back the seller's compensation because the seller is leaving. Reasonable. But you are replacing the seller. Unless you have another acceptable income source, the business still has to support an operator, and that operator is you.
You are not buying a salary. You are buying a job plus a return, and the lender models both.
The lender is not asking only whether the business can pay the loan. It is asking whether the business can pay the loan, pay you a reasonable salary, absorb normal capital spending, and still hold a margin of safety.
Now rerun the $1.8 million deal the way the bank did.
The buyer started at $380K of listing SDE against $262K of debt service and saw 1.45x. The lender verified $360K, because $20K of claimed earnings had no documentation. It subtracted a $120K market salary for the new operator. It reserved $15K for capex. Cash available for debt service: $225K.
$225K against $262K of debt service is 0.86x. That is not a pricing discussion anymore. That is a failed credit file.
The issue is not that the buyer was foolish. The issue is that listing SDE answers a different question than lender DSCR. One measures what the seller took out of the business. The other measures what is left after the business pays everyone it has to pay, including you.
Every Unsupported Add-Back Is a DSCR Cut
Add-backs are usually framed as a valuation issue. They are also DSCR inputs, and that is where they do quiet damage.
Lenders credit what can be verified. Unsupported personal expenses, vague one-time costs, missing revenue claims, and aggressive normalization come out of the numerator. Every dollar removed lowers DSCR, and a deal priced off the full add-back schedule can slip below threshold on haircuts alone.
This is the practical case for a quality of earnings review on deals where cash flow is the thesis: it tells you what earnings are likely to survive lender scrutiny before you spend months in diligence. A QoE that confirms the earnings protects your DSCR. One that cuts them tells you to renegotiate now, while you still can.
Run This Before You Sign the LOI
Five lines. Ten minutes. Do it before the LOI, not after the term sheet.
Start with tax-return-supported earnings, not listing SDE.
Remove every add-back you cannot document.
Subtract the salary the business needs to support for you.
Subtract a realistic capex and working-capital reserve.
Divide what remains by annual debt service on the actual loan structure.
If the number is under 1.25x, the deal is not automatically dead. It means the structure has to change before you sign.
Thin DSCR Is a Structure Problem, Not Always a Bad Deal
A thin lender-style DSCR is information. Serious buyers treat it as the starting point for structure, not a reason to walk. The levers, roughly in order of impact:
Lower the purchase price. If verified cash flow cannot carry the debt, the price is too high for this financing.
Increase seller debt on full standby. Payments on full-standby notes sit outside annual debt service, which lifts the ratio.
Add more equity. A smaller loan means smaller debt service.
Clean up the add-backs. Better documentation, or a QoE, can move verified earnings up instead of down.
Use the full available amortization. A 10-year term you compress to 7 only tightens your own ratio.
Find the right lender. Credit boxes differ. A deal one bank declines at 1.2x may still get considered elsewhere if the industry, collateral, buyer profile, and global cash flow are strong.
Improve global cash flow. Outside income strengthens the file; heavy personal debt drags it.
Real DSCR gives the buyer leverage. It turns a vague financing risk into a specific negotiation input: how large the standby note needs to be, how much price needs to move, which add-backs need proof.
The Number Is Either Yours or the Bank's
Every deal gets a lender-style DSCR eventually. The only question is who calculates it first.
Run it yourself and you walk into the LOI knowing your real coverage, your real salary load, and exactly which lever to pull if the ratio is thin. Let the bank run it first and you find out in week six, after the fee clock and the deal momentum are already gone.
The buyer who knows the real DSCR before LOI controls the negotiation. The buyer who learns it from the bank usually just loses time.
Open your model. Subtract your salary. Cut the add-backs you cannot prove. Reserve for capex. Divide by the real debt service.
Then decide with the real number, not the listing number, in front of you.
Disclaimer: This guide is for educational purposes only and does not constitute legal, financial, tax, or investment advice. Business acquisitions involve significant risks, and outcomes can vary widely based on individual circumstances. Always consult with qualified professionals including attorneys, CPAs, and financial advisors before making acquisition decisions. The EBIT Community does not guarantee the accuracy of information provided or the success of any acquisition strategy. Past performance and examples do not guarantee future results.
