Buyer's guides · Reading the numbers
How much is a small business worth — and where the multiple comes from
A small business is worth its earnings to a working owner multiplied by a number that summarises how safe those earnings look. Main Street businesses changed hands at roughly 2–3× SDE in mid-2026, depending on the size of the deal. Your deal is not the average: the multiple moves with proof, concentration, the lease and who has to work there.
What this answers
- The formula everyone uses — earnings × multiple — and why the first term decides far more of the price than the second.
- What the median multiple actually describes, and why quoting it back at a seller proves nothing about this business.
- The things that genuinely move a multiple up or down, in the order a buyer can check them.
- Why a valuation calculator cannot answer how much is a small business worth, and what it is still useful for.
- Whether the multiple goes on profit, revenue, SDE or EBITDA — and what changes when someone switches between them mid-conversation.
How much is a small business worth?
Almost every small business in the United States is priced the same way: take the earnings the business produces for one working owner, and multiply them by a number. That number — the multiple — is not a market rate you can look up. It is a compressed opinion about how likely those earnings are to survive the owner walking out of the door.
Which means the question has two halves, and buyers spend their attention on the wrong one. Arguing the multiple down from three to two and a half is a negotiation. Establishing whether the earnings figure is real is arithmetic, and it moves the price by more.
The first term: what earnings, exactly
A small-business profit and loss statement is written to show the lowest legal profit. The owner's salary, the owner's vehicle, health insurance, a phone, sometimes a family member on payroll — all of it sits in expenses. Read literally, a business that comfortably supports a household can look like it barely earns.
Seller's Discretionary Earnings undoes that: net profit before tax, plus one owner's compensation and benefits, plus interest, plus depreciation and amortisation, plus genuine one-off costs that disappear with the seller. It answers the question a buyer actually has — if I take this over and run it myself, what does it produce before I pay debt and tax?
Two traps live inside that definition, and both inflate the price:
- One owner, not two. SDE contains the pay of a single working owner. When two people work full time, the second person's replacement cost is a real expense. Add both back and you have priced a business that needs an employee you have not budgeted for.
- Add-backs without documents. Every adjustment is a claim that an expense will not exist after the sale. Some are obviously fair; the ones with no invoice, bank line or payroll record behind them are arguments, not earnings — see add-backs. They are also the cheapest thing in the whole packet to inflate, because each dollar added comes back multiplied.
This is why the reading order matters more than the negotiation. A packet whose earnings have never been tied to a tax return has not produced a price at all; it has produced a proposal. The documents that settle it are listed in what to ask a seller for.
The second term: where the multiple comes from
Multiples are observed, not calculated. Broker associations survey completed sales and publish what actually happened, and appraisers and brokers price new listings against those observations, adjusted for the story in front of them.
The benchmark most often quoted on Main Street comes from the IBBA and M&A Source Market Pulse survey, which covers businesses valued up to $2 million — the segment priced on SDE rather than EBITDA. In its second-quarter 2026 survey — 255 advisors reporting 181 closed deals — the multiples advisors reported across the Main Street bands ran from about 2× to 3× SDE, with the smallest deals at the bottom of that range and the largest at the top.
Read that sentence carefully, because it is routinely misused in both directions. It is a median of completed sales across every industry in the cohort. It is not a fair price for a business, not a ceiling, and not evidence about the deal in front of you. A seller quoting it to justify an asking price and a buyer quoting it to argue one down are making the same mistake: treating a summary statistic about hundreds of other businesses as a fact about this one.
What the median is genuinely good for is calibration. If the asking price implies a multiple far above it, there should be a reason you can name, written down, and supported by something. "It is a great business" is not one of the reasons.
What actually moves the multiple
Every item below is a risk to the earnings continuing after the sale, which is all a multiple ever encodes. They are ordered roughly by how cheaply a buyer can check them.
| Factor | Pushes the multiple down when… | Pushes it up when… |
|---|---|---|
| Quality of the records | Earnings live in a spreadsheet the seller maintains and never reconcile to tax returns | Returns, statements and bank deposits tell the same story without explanation |
| Owner dependence | The owner holds the relationships, the licence or the technical skill, and nobody else can do the work | A manager already runs it, and the owner's absence has been tested |
| Customer concentration | A handful of customers carry most of the revenue, with no contracts behind them | Revenue is spread, recurring, and contracted |
| The lease | The term left is short, assignment needs a consent nobody has sought, or the rent resets on transfer | Years remain, assignment is clean, and the rent is at market |
| Earnings trend | The business is shrinking, or one good recent year sits on top of two flat ones | Growth is steady and the reason for it is something other than the owner working harder |
| Staff | Key people are unaware of the sale, or their pay is below what replacing them would cost | The team is stable, paid at market, and stays |
| Transferability | Licences, permits or a franchise approval stop with the seller | Everything needed to trade transfers, and the process is known |
Notice what is absent: industry averages by trade, the seller's retirement plans, and how long the listing has sat. Those explain a price after the fact. They do not survive contact with a lender or a resale.
Is the multiple on profit, revenue, SDE or EBITDA?
All four get used, often in the same conversation, and switching between them mid-sentence is the most common way a price gets quietly inflated.
- SDE is the Main Street standard for owner-operated businesses. It includes one owner's compensation, so the multiple applied to it is lower than an EBITDA multiple on the same business.
- EBITDA excludes owner compensation, because larger businesses are priced as though a manager is hired to run them. The same business shown as SDE and as EBITDA differs by an entire salary.
- Revenue multiples are a sanity check, not a valuation. Two businesses with identical revenue and different cost structures are not worth the same, and a revenue rule of thumb cannot see the difference.
- Net profit as reported is almost never the basis, because of everything described above — which is exactly why "how many times profit is a small business worth" has no stable answer.
The practical defence is one question, asked early: which earnings figure is this multiple applied to, and can I see how it was built? A listing that says "cash flow" without saying which is not being evasive on purpose, usually — but the ambiguity favours the seller every time.
Why a valuation calculator cannot answer this
Search for how much a small business is worth and half the results are calculators. They take a revenue figure, an earnings figure and an industry, and return a number with decimal places. The arithmetic is correct and the output is close to meaningless, for a reason that has nothing to do with the calculator's quality: it takes the earnings figure from you, and you got it from the seller.
Multiplying an unverified number by a plausible multiple produces an unverified number with more confidence attached. That is worse than no estimate, because a number on a screen is remembered as a finding.
A calculator is still useful for one thing: running the deal backwards. Take the asking price, divide by the earnings the seller claims, and look at the multiple that falls out. If it lands far above the Main Street median, you now have a specific question to ask rather than a general unease — and the answer is either a documented reason or the beginning of a renegotiation.
What the price is not
Three things sit next to the valuation and get folded into it by mistake.
- The price is not what you need in cash. Financing, seller participation and closing costs all change the money that leaves your account — see seller notes. A business you can afford and a business you can pay for are different questions.
- The price usually assumes normal working capital comes with it. If it does not, the business needs funding on day one that nobody has priced.
- The price is not what a lender will accept. Lenders test whether the earnings cover the debt they would be creating — the coverage arithmetic is in DSCR. A price the lender will not support is a price you cannot pay with their money, whatever the seller and the broker agree.
Where the asking price came from
On most listings it came from a broker's opinion of value, prepared for the seller, using the seller's own adjusted earnings and a multiple from comparable sales. Every step of that is defensible, and the one underneath it is the problem: the earnings are taken as given. What such a document is and is not — and the five questions worth asking about any of them — is set out in what a broker's opinion of value does not cover.
When a deal is far enough along that the price is real money, the earnings get verified properly rather than argued about. That is a Quality of Earnings review, and what one contains and what it costs is a separate question from valuation — but it is the one that decides whether the valuation was built on anything.
None of this tells you what to offer. It tells you which parts of a price rest on documents and which rest on a story, so that the number you decide on is yours.
Work out the multiple the asking price implies
Divide the asking price by the earnings the seller claims. That single division turns a price into a comparable figure, and it is the only number you need before the first conversation.
Find out which earnings figure that multiple sits on
SDE, EBITDA, "cash flow" and net profit are different numbers for the same business. Ask which one, and ask to see how it was built — the schedule matters more than the total.
Tie the earnings to a tax return before you touch the multiple
Compare the revenue and profit in the most recent return to the figures in the listing. A gap the seller cannot explain in one sentence makes the multiple irrelevant, because the number it multiplies is in question.
Test each add-back against a document
Every adjustment either has an invoice, a bank line or a payroll record behind it, or it moves to a list of questions. At the Main Street median, each unproved dollar of earnings is nearly three dollars of price.
Price the risks you found, in words first
Concentration, a short lease, an owner who is the business, staff paid below market: write down what each one would cost you to fix. A multiple argued from named risks is a negotiation; one argued from a survey median is a debating point.
Check the price against what a lender would fund
Run the coverage arithmetic before you agree a number, not after. A price that fails the lender's test has to be renegotiated anyway, and it is cheaper to discover that before an LOI than during underwriting.
Questions buyers ask
How much is a small business usually worth?
Usually some multiple of what it earns for a working owner. In the IBBA and M&A Source Market Pulse survey for the second quarter of 2026, which covers businesses valued up to $2 million and draws on 255 advisors reporting 181 closed deals, reported multiples ran from about 2× to 3× SDE across the size bands. That is a range across every industry in the cohort, not a price for a particular business: the same earnings can be worth meaningfully more or less depending on how well documented they are, how concentrated the customers are, and whether the business runs without its owner.
How do you figure out how much a small business is worth?
Establish the earnings first, then the multiple. Rebuild Seller's Discretionary Earnings from the tax returns rather than accepting the seller's figure, check each add-back against a document, and only then ask what multiple is justified by the risks you actually found. Doing it in the other order — starting from an industry multiple — produces a confident price built on a number nobody has verified.
How many times revenue is a small business worth?
Revenue multiples are a rough sanity check rather than a valuation method for owner-operated businesses. Two businesses with the same revenue and different cost structures are not worth the same amount, and a revenue rule of thumb cannot tell them apart. Main Street pricing is based on earnings to a working owner; revenue is useful mainly for noticing when a claimed profit looks implausible against the sales that produced it.
Is the multiple applied to EBITDA or SDE when buying a small business?
For owner-operated businesses it is normally SDE, which includes one owner's compensation. EBITDA is used further up the market, where a buyer assumes a hired manager. Because SDE contains a salary that EBITDA does not, the two produce different earnings figures for the same business and carry different multiples — so a multiple quoted without saying which basis it uses cannot be compared to anything.
Can a small business valuation calculator tell me what to pay?
No, because it takes the earnings figure from you and you took it from the seller. The arithmetic is fine; the input is the unverified part. A calculator is useful in reverse — divide the asking price by the claimed earnings to see what multiple is being asked, then go and find out whether the earnings are real.
Sources
- IBBA / M&A Source, Market Pulse Q2 2026 (255 advisors, 181 closed deals)
- IRS — Get transcript: how a taxpayer obtains a return transcript
Read next
- What documents to ask a seller for — and what each one has to prove
- What a Quality of Earnings report looks like — and what it costs
- Seller's Discretionary Earnings (SDE)
- Add-backs
- What a broker's opinion of value does not cover
- Working capital (and the peg)
Have the documents already? A DealLoupe pre-screen reads what the seller gave you and reports the red flags, the evidence gaps and the questions to ask — before you spend anything on due diligence. See what it costs →
Last updated: 2026-09-25. Educational information, not financial or legal advice — and not a substitute for a CPA, an attorney or a formal Quality-of-Earnings review.