Skip to content
CurriculumToolsLibraryCommunitySign in
Join the community

Curriculum/Part VI: Underwrite/Valuation

Lesson 6.4.1

Comps, DCF, DSCR

The three lenses and what each can and cannot tell you. Debt service coverage is the gating math for an SBA deal: the 1.25x floor is computed on historical earnings only, with no credit for projections.

Lesson 7 of 12 in this part · 10 minute read · Draft, under review

Short answer

How do I value a small business I want to buy with an SBA loan?

Use three lenses and take the lowest credible answer as your ceiling. Comparable sales show what other buyers paid. A discounted cash flow shows what the cash is worth to you. Debt service coverage shows what the loan can carry: under SBA SOP 50 10 8.1, for applications received on or after October 1, 2026, an initial acquisition needs at least 1.25x on historical earnings, not projections.

You now have an earnings number you trust. The next question is what to pay for it. There are three ways to look at that question, and each answers a different version of it. Comparable sales tell you what other buyers have paid. A discounted cash flow tells you what the cash is worth to you. Debt service coverage tells you what the lender will let the business carry. On an SBA-financed deal the third one is not a matter of opinion: it is a gate, and under the current SBA rules it is computed on history, not on the seller's forecast.

Principle

Three lenses, three questions. Your price is the lowest answer that survives all three.

First, the vocabulary.

  • A multiple is price divided by earnings. A $3,000,000 price on $750,000 of earnings is a 4.0x multiple.
  • SDE (seller's discretionary earnings) is profit before the owner's entire pay and perks. It is the earnings figure used for small owner-operated businesses, where the buyer will replace the owner by working in the business.
  • EBITDA is earnings before interest, taxes, depreciation and amortization, after paying a market wage to whoever runs the business. On the same company, SDE is always larger than EBITDA, so SDE multiples are always smaller. Never apply an EBITDA multiple to an SDE number, or the reverse. That one mismatch can add a turn or more to the price.
  • Comps (comparable transactions) are recorded sales of similar businesses, used to infer a multiple.
  • DCF (discounted cash flow) values a business as the sum of the cash it will produce, with future dollars discounted for time and risk.
  • DSCR (debt service coverage ratio) is earnings divided by the year's required loan payments. A DSCR of 1.25x means the business earns $1.25 for every $1.00 of debt payments.

What each lens is for, and where it fails:

Comps tell you what the market pays, not what this business is worth. Small-business transaction data is thin, self-reported and uneven in how earnings are defined. Asking prices are not sale prices: a 2026 study of small business acquisitions found businesses transacting for significantly less than the seller's asking price during periods of high economic uncertainty (C11). And public-company multiples do not transfer. Research going back decades finds private companies acquired at a discount to comparable public companies: an average 20 to 30 percent on earnings multiples in a 2000 study (C20), and 15 to 30 percent for unlisted targets in a 2007 study, wider when debt is expensive (C8). Those studies cover companies large enough to be compared with public ones. Yours is smaller and less liquid still.

A comp is only as good as the earnings under it. A 2010 study found expert valuers relying heavily on private firms' reported earnings even where those earnings could have been managed (C23). If the comp's multiple sits on unscrubbed earnings and you apply it to scrubbed earnings, you are comparing different things.

A DCF tells you what the cash is worth to you, and it is only as honest as its inputs. Small changes in growth or discount rate swing the answer widely. Built on the seller's projections, a DCF is an advertisement. Built on your recurring earnings with conservative growth, it is a useful check on whether the comps-based price can earn you a return for the risk of a personally guaranteed loan.

DSCR tells you what the debt can carry. It is a ceiling, never a value. A lender approving a loan at 1.25x means the business can probably make the payments in a normal year. It does not mean the price is fair, and it says nothing about your return.

Case

Coastal Comfort HVAC is the fictional company that runs through this curriculum. After the add-back bridge in lesson 6.2.1 and the spike removal worked in lesson 6.6.1, its recurring EBITDA is about $1,017,000, already net of a $225,000 market cost for a general manager. The broker claims $1,710,000 and asks $6,450,000.

Comps. The curriculum's answer key treats 3.0x to 4.5x recurring EBITDA as the defensible range for this business. That range is illustrative, set for the exercise, not a market statistic. On $1,017,000 it gives about $3,051,000 to $4,577,000.

DCF. An illustrative version, with inputs chosen to show the mechanics and not drawn from market data. Start from $1,017,000 and subtract maintenance spending on trucks and equipment and income taxes; suppose that leaves about $695,000 a year of cash available to owners before debt payments. If the business never grows and you require a 15 percent annual return for the risk, the value is $695,000 divided by 0.15, about $4,630,000. Require 20 percent, and it is about $3,480,000. Now let the seller's forecast in: assume the FY2025 spike repeats and grows, and the same arithmetic produces a number above the ask. Same business. The inputs are the valuation.

DSCR. This is where the ask breaks. An illustrative structure: 10 percent buyer cash, a 10 percent seller note repaid over five years at 6 percent, and an 80 percent SBA loan amortized over ten years at an illustrative 10 percent rate. Most 7(a) loans carry a variable rate, and your lender quotes the real one. Ignoring closing costs and working capital to keep the arithmetic clean:

Price Annual debt service DSCR on broker's $1,710,000 DSCR on recurring $1,017,000
$6,450,000 (the ask) about $968,000 1.77x 1.05x
$4,500,000 about $675,000 2.53x 1.51x
$3,814,000 (range midpoint) about $572,000 2.99x 1.78x

On the broker's number, the ask looks comfortably bankable. On the earnings that will actually arrive, it fails the SBA floor by a wide margin: anything under 1.25x fails the floor for an Initial Acquisition under SOP 50 10 8.1 (Y25), and the ask cannot be financed as structured. It also needs an SBA loan of about $5,160,000, above the $5,000,000 maximum for a standard 7(a) loan. Solve the other way, and in this illustration 1.25x on $1,017,000 supports a price of roughly $5,400,000. In a real structure, where closing costs and working capital are financed too, the $5,000,000 loan maximum binds before coverage does; lesson 6.5.2 puts that lender ceiling near $4,850,000.

Now line up the three lenses: the multiple range says $3,051,000 to $4,577,000; the illustrative DCF says $3,480,000 to $4,630,000; DSCR caps the price near $5,400,000. The lowest credible answer comes from the multiple range, so that is where the price lives, and lesson 6.5.2 uses its midpoint, about $3,814,000, as the value ceiling. The fact that the debt could carry more is your margin of safety, not an invitation to spend it.

The case's own letter of intent shows the lenses agreeing: $3,800,000, just under that midpoint. Every figure in this lesson rests on $1,017,000 from the seller's documents, though, and the price is above $3,000,000, so the SOP requires a quality of earnings report whose earnings the lender must use. In the case, that report moves recurring earnings to about $920,000 (lesson 7.3.1 shows what it found). The same lenses, re-run, put the value midpoint near $3,450,000, and the deal closes at $3,500,000, inside the lender's independent valuation of $3,550,000. Lesson 6.5.2 lays out both structures. A valuation is only as good as the earnings under it, and the earnings are not final until the QoE is.

Steps

Step 1: Fix the earnings basis

Decide whether you are valuing on SDE or EBITDA, write it at the top of the worksheet, and use only comps on the same basis. For a business you will run yourself, an SDE basis is common; once you pay yourself or a manager a market wage, you are on EBITDA. Use your recurring number from lessons 6.2.1 and 6.3.1, not the broker's.

Step 2: Build the comp table

  1. Collect recorded sales of businesses in your industry and size band. Sources include transaction databases (some are built from SBA lender data), business broker market surveys, your lender's and appraiser's experience, and brokers you know. Record the source and date of each.
  2. For each comp, note revenue, the earnings basis (SDE or EBITDA), the sale price, the multiple and the date. Discard anything that is an asking price.
  3. Compute the median and the middle half of the range. Look at outliers and write down why they are outliers.
  4. Adjust for this business: concentration, owner dependence, a declining trend or a spike year push the multiple down; diversified recurring customers and a team that runs without the owner push it up. Write the evidence for each adjustment.

Step 3: Build a simple DCF as the cross-check

  1. Start from recurring earnings. Subtract maintenance capital spending (what it costs every year just to keep the equipment and fleet running) and income taxes.
  2. Project five years with conservative growth anchored to the business's own history, not the seller's forecast.
  3. Choose a discount rate that reflects your risk: a personal guarantee, a single-owner business, an industry you may be new to. Write down why you chose it.
  4. Add a terminal value at the end of year five, usually by applying an exit multiple no higher than your entry multiple.
  5. Run a small sensitivity grid: three growth rates against three discount rates. If the answer swings wildly, the deal is fragile.

If the DCF and the comps disagree by a lot, one of them is wrong. Usually it is the growth assumption.

Step 4: Run DSCR the way the lender will

Under SOP 50 10 8.1 (which applies to applications received by SBA on or after October 1, 2026, Y25), Appendix 15 sets the coverage test for a change of ownership:

  • The floor. Debt service coverage must be at least 1.25:1 for an Initial Acquisition (a buyer becoming the new majority owner of a business they never owned or worked in) and for Owner Buyouts. Business Expansions need 1.15:1. Under the prior SOP 50 10 8 (effective June 1, 2025), the general standard was 1.15 on a historical and/or projected basis.
  • The earnings period. Coverage must be met using either the last fiscal year-end or an average of the last two fiscal year-end statements, on a historical or adjusted basis. The SOP states that the projection-based coverage method used for other loans is not applicable to change of ownership transactions. Your forecast does not count.
  • The formula. Historical DSC is EBITDA divided by the combined debt service after the transaction.
  • Adjustments. The lender may make prudent adjustments, such as unfunded capital spending, non-recurring income, distributions, distributions for S corporation taxes, seller discretionary expenses and owner compensation, and must justify each one in the credit memo. That means the lender's earnings figure can come out below yours, and lenders differ, so ask yours which adjustments it makes. An adjustment to owner compensation must be backed by a global cash flow analysis showing the owners meet their obligations at 1:1 on a global basis.
  • The QoE link. If a QoE is required (purchase price of $3 million or more, see lesson 6.3.2), the lender must use the QoE's earnings in the coverage calculation, and if coverage does not support the structure, the loan amount must be reduced.
  • The valuation link. Every change of ownership also requires a business valuation prepared for the lender, by an independent, qualified professional, except that at a Business Purchase Price of $350,000 or less the lender may value the business itself (unless buyer and seller are closely related). The technical updates effective October 1, 2026 confirm that small-deal option for 7(a) Small and SBA Express loans (Y26). If the price exceeds the valuation, the difference must be made up with equity. The total debt supporting the purchase, including seller debt that is not on full standby, is limited to the valuation amount.
  • Small loans get no discount on coverage. The same technical updates permit 7(a) Small and SBA Express loans to finance a change of ownership, but the full Appendix 15 coverage requirements apply to them; there is no reduced standard because the loan is small (Y26).

To run it yourself:

  1. Use your recurring EBITDA, with a market wage for the manager already deducted.
  2. List every loan payment for the first full year: SBA principal and interest on a ten-year amortization (the SOP's maximum for a change of ownership without real estate) and any seller note that is not on full standby.
  3. Divide. Then run it at the ask, at your comps midpoint, and at your walk-away price.
  4. Stress it. At exactly 1.25x, a 20 percent fall in earnings takes you to 1.0x: every dollar the business earns goes to the bank. That stress run is your test, not the lender's; the lender tests history only. Decide in advance how much cushion above the floor you need to sleep.

The site's SBA Deal Stress-Tester runs this math under the old and new rules side by side.

Step 5: Reconcile the three

Write the three answers in one table. Take the lowest credible figure as your ceiling, and note which lens produced it. If DSCR is the binding lens, the deal is a financing problem: the model in lesson 6.5.1 and the structures in lesson 7.4.1 may help. If comps or the DCF bind, the deal is a price problem, and no structure fixes paying too much.

Deliverable

A one-page three-lens valuation sheet:

  • The earnings basis (SDE or EBITDA) and your recurring number.
  • The comp table, with sources, dates, the median, the range, your adjustments and the implied value range.
  • The DCF inputs with a sentence of rationale each, the result, and a three-by-three sensitivity grid.
  • The DSCR at the ask, at your comps midpoint and at your walk-away price, under SOP 50 10 8.1's 1.25:1 floor, plus the price at which the deal hits exactly 1.25x.
  • One line: your ceiling price, and which lens set it.

This lesson is education, not valuation, lending, tax or legal advice. Lenders apply the SBA rules with their own credit policies; confirm the test with your lender, and have a credentialed valuation professional and a CPA review any deal you intend to finance.

Common questions

Asked about this lesson

What is DSCR and how is it calculated?

DSCR, the debt service coverage ratio, is earnings divided by the year's required loan payments. A DSCR of 1.25x means the business earns $1.25 for every $1.00 of debt payments. For an SBA change of ownership, historical coverage is EBITDA divided by the combined debt service after the transaction.

Can I use projections to qualify for an SBA acquisition loan?

No. Under SBA SOP 50 10 8.1, coverage for a change of ownership must be met using the last fiscal year-end or an average of the last two fiscal year-end statements, on a historical or adjusted basis. The SOP states that the projection-based method used for other loans does not apply to change of ownership transactions.

Should I use an SDE multiple or an EBITDA multiple?

Pick one basis and use only comps on the same basis. SDE is profit before the owner's entire pay and perks. EBITDA is after a market wage for whoever runs the business. On the same company SDE is always larger, so SDE multiples are smaller. Mixing the two can add a turn or more to the price.

Does an SBA loan to buy a business require a business valuation?

Yes. Under SBA SOP 50 10 8.1, every change of ownership requires a business valuation prepared for the lender by an independent, qualified professional. At a business purchase price of $350,000 or less the lender may value the business itself, unless buyer and seller are closely related. If the price exceeds the valuation, the difference must be made up with equity.

More answers on the buyer FAQ.

Sources

From the Library

  1. C20

    Koeplin, J., Sarin, A., & Shapiro, A. C (2000). The private company discount. Journal of Applied Corporate Finance, 12(4), 94-101.

  2. C8

    Officer, M. S. (2007). The price of corporate liquidity: Acquisition discounts for unlisted targets. Journal of Financial Economics, 83(3), 571-598.

  3. C11

    Kasten, C. L., Lambert, C. A., & Osborne, B. N. (2026). Small business valuations during uncertain times. Journal of Small Business Management, 64(5), 1999-2031.

  4. C23

    Elnathan, D., Gavious, I., & Hauser, S (2010). An analysis of private versus public firm valuations and the contribution of financial experts. International Journal of Accounting, 45(4), 387-412.

  5. Y25

    U.S. Small Business Administration (2026). SOP 50 10 8.1, Lender and Development Company Loan Programs, with Technical Policy Updates. Effective October 1, 2026.

  6. Y26

    U.S. Small Business Administration (2026). Information Notice 5000-882227: Issuance of Technical Updates to SOP 50 10 8.1. Published September 25, 2026; effective October 1, 2026.