Quality of Earnings: What a QoE Report Actually Tests (and Why the Seller's Number Isn't It)

A buyer signs a letter of intent on a landscaping business at a price built on $640,000 of seller's discretionary earnings. The broker's packet is tidy. The add-back schedule is itemized. The trailing-twelve-month revenue is up. Everyone at the table is treating that $640,000 as the number, the figure you multiply to get a price and divide into a loan payment to prove the deal services its debt.

Then someone actually reads the records. One of the "one-time" legal expenses has appeared in four consecutive years. The owner's full $190,000 draw was added back with nothing subtracted for the manager who would have to replace him. And $120,000 of the trailing year's revenue came from a single construction job that has already wrapped and will never recur. The $640,000 the buyer was about to pay a multiple on was never $640,000. It was closer to $470,000. The difference, at a 3x multiple, is roughly half a million dollars of price.

This is the gap a quality of earnings analysis exists to close. It is the most-cited and least-understood document in small-business acquisitions, and the buyers who get hurt are almost always the ones who treated the seller's earnings figure as the answer instead of as a claim to be tested.

What quality of earnings actually means

A quality of earnings report (QoE) is an independent analysis that tests whether a company's reported earnings are real, sustainable, and transferable to a new owner.

Notice what that sentence does not say. It does not say the QoE checks whether the books are correctly recorded. That is an audit, and it answers a different question. An audit asks: do these financial statements comply with the applicable accounting standard? A QoE asks: is the earnings number we are using to value this business the right number to use?

Those questions diverge more than most first-time buyers expect. A set of books can be perfectly compliant and still bury everything that matters to a buyer. Customer concentration is not a GAAP violation. Adding back an owner's full salary is not a GAAP violation. Booking a one-time settlement above the line is not a GAAP violation. An audit is not built to flag any of them, because none of them break the rules of recording. A QoE is built to flag all of them, because each one changes what the earnings are worth.

The three tests inside that one sentence are worth pulling apart, because they map to the three things a buyer is really buying:

• Real. Did the business actually earn this? Does booked revenue tie to money that landed in the bank? Are the expenses complete, or has something the business genuinely needs been quietly removed to make the number look bigger?
• Sustainable. Will it earn this again next year? Or is the trailing figure inflated by a one-time project, a pulled-forward contract, a customer that prepaid, or a cost that was deferred and is about to come due?
• Transferable. Will it still earn this after the seller is gone? Are the customers loyal to the business, or to the owner who is leaving? Does the number assume free labor from a founder who won't be there on Monday?

Hold those three words. Everything a QoE examines is in service of one of them.

What a QoE actually examines

The work of a quality of earnings analysis is a series of specific tests, each aimed at a specific way an earnings number can be true on paper and false in practice. Here is what gets examined, and why each one matters.

Owner-compensation normalization

In an owner-operated business, the owner's pay is set by what the business can afford and what the tax return rewards, not by what the role is worth. So the QoE normalizes it. The owner's compensation comes out, and the market-rate cost of replacing that owner's labor goes back in.

That second half is where buyers get cooked. Adding back a full owner salary without subtracting a manager's replacement cost is the single most common way an earnings figure gets inflated. If the owner draws $190,000 and runs the business, and a working general manager in that market costs $110,000, the real add-back is $80,000, not $190,000. The other $110,000 is a job someone has to do and pay for. A QoE that adds back the full draw without netting the replacement is not normalizing earnings; it is hiding a salary. (We cover this in depth in does SDE include owner salary.)

Add-backs: which survive, which don't

Every add-back is the seller's assertion that an expense won't continue under new ownership. Some of those assertions are true. Many are optimistic. A few are simply wrong. The QoE gives each one an opinion: accepted, partially accepted, rejected, or unverifiable.

This is where my own work lives more than anywhere else, because it is where a defensible number and a hopeful one separate. The pattern repeats across deals: the personal auto that the business doesn't need, the "consulting fees" paid to a spouse with no work product, the meals and entertainment treated as if a new owner will never eat. And the one I see most: the "one-time" cost that shows up four years running. A cost that recurs is not one-time, no matter what the schedule calls it. The QoE checks each add-back against three questions: is it actually in the tax return, will it genuinely not recur, and can it be documented? An item that fails any of the three comes out. This is exactly the test an SBA underwriter runs, which is why the QoE's bridge and the lender's adjusted EBITDA tend to converge; for the lender's view specifically, see how to analyze add-backs and what SBA lenders actually accept, and for the seller-side mechanics, how brokers inflate SDE.

One-time versus recurring items

Beyond the add-back schedule, the QoE sweeps the income statement for items that distort the run rate in either direction. A non-recurring legal settlement, a forgivable loan booked as income, a gain on the sale of equipment, a single unusually large project: these inflate the trailing figure and need to be pulled out before a multiple touches it. The mirror also matters. A cost that was deferred, a one-time benefit that flattered margins, an expense the seller skipped this year that a new owner cannot skip: these understate the true run rate and have to be normalized too. The goal is a number that reflects what the business earns in a normal year, not the best or worst twelve months the seller chose to present.

Revenue quality and concentration

Two businesses with identical earnings can be entirely different assets depending on where the revenue comes from. The QoE tests durability: how much revenue is recurring versus one-time, how concentrated it is (the top customers as a percentage of the total), how well it has been retained, and what the contracts actually say.

A business with 60% of its revenue in one customer is structurally more fragile than one with the same earnings spread across hundreds of accounts, and the QoE quantifies that rather than asserting it. It also tests something the customer list alone can't show: whether those customers are loyal to the business or to the departing owner. A 25% concentration in two accounts under multi-year contracts is a manageable risk. The same 25% in two accounts that do business only because they golf with the founder is a different deal entirely. (The full framework is in customer concentration risk in SMB deals.)

Working-capital normalization

This is the exhibit buyers most often skip and most often regret skipping. The QoE builds the trailing working-capital schedule (receivables plus inventory minus payables, normalized to a trailing average) and proposes the peg: the normal level of working capital the business needs to operate, which the buyer should receive at closing.

The peg matters because it quietly moves cash between the two parties at the closing table. If the seller's broker sets it, it tends to be set low, which lets the seller keep more cash and receivables on the way out and leaves the buyer to inject working capital just to keep the doors open. A poorly set peg is the single most common source of a closing-table re-trade. The full mechanism is in working capital adjustment at closing.

SDE versus EBITDA

One framing point underneath all of this: smaller owner-operated deals are usually valued on SDE (seller's discretionary earnings), which adds back one working owner's full compensation on the theory that a single owner-operator will step in. Larger deals, and any deal where management is a hired team rather than the owner, are valued on adjusted EBITDA, which assumes management is a real, ongoing cost and does not add the owner's pay back. The QoE works in whichever frame the deal is being priced in, but it is precise about which one, because the two produce different numbers from the same business and a multiple meant for one does not transfer to the other. If you want the underlying mechanics, start with how to calculate SDE.

Buy-side versus sell-side

A QoE can be commissioned by either party, and the direction changes its purpose.

A buy-side QoE is ordered by the buyer, after the LOI, to test the seller's number before paying for it. It is investigative by design. Its findings are leverage: every rejected add-back, every isolated one-time item, every gap between booked revenue and bank deposits is a reason to adjust price or structure while there is still exclusivity to do it in.

A sell-side QoE is ordered by the seller, before going to market, for the opposite reason: to find and fix the weak spots first, then present buyers with a clean, pre-validated earnings figure. A good sell-side QoE smooths a process and supports the asking price. It does not, however, replace the buyer's own work. A sell-side report was paid for by the party with an interest in the answer, and a serious buyer reads it as a useful starting point, not as independent confirmation. When real outside money is at the table, lenders and equity partners generally want a buy-side report from a firm they chose.

What's inside the report, what it costs, and when you get it

I won't re-walk the report section by section here, because that is its own piece: a quality of earnings report example, exhibit by exhibit lays out each part of the document and what to challenge in it. The short version is that a QoE for an SMB deal is usually a 30-to-80-page document combining narrative findings with supporting exhibits, built around the adjusted-earnings bridge. The companion guide on when a QoE is worth the spend covers the thresholds and the decision in detail.

On cost, the honest answer is a range, not a figure. A QoE for an SMB acquisition typically runs $15,000 to $50,000. The market has split into roughly three tiers: tech-enabled providers in the $10,000 to $25,000 range for deals under about $5M, boutique transaction-CPA firms in the $25,000 to $50,000 range, and Big 4 transaction advisory rarely below $10M in deal size and starting around $75,000. Treat all of those as typical ranges; any real quote turns on scope, the number of entities, and how clean the seller's records are. Below the point where a full QoE earns its keep, a lender-grade SDE schedule reviewed by a transaction CPA covers most of the same ground for something closer to $3,000 to $8,000.

On timing, two things matter: when you order it and how long it takes. A buy-side QoE is commissioned right after the LOI is signed, in the first days of the exclusivity period. That placement is deliberate. Before the LOI, the seller usually won't grant the general-ledger and bank-statement access a real QoE requires. After exclusivity is well underway, the window to act on the findings is closing. Order it early in diligence, and its results can still move price or structure while there is leverage to renegotiate. A sell-side QoE runs earlier still, before the business is listed. Either way, typical turnaround once it kicks off is three to six weeks.

A worked example: walking $640,000 down to $470,000

Return to the landscaping business from the top. Asking price built on $640,000 of presented SDE at a 3x multiple, which is to say a roughly $1.9M price. SBA financing, single owner-operator planning to exit within six months of close. The buyer commissions a QoE in week one of diligence. Here is what the bridge looks like when the analyst is done.

Now read it back the way the three tests would. The owner-comp line is transferability: that $110,000 is a job the buyer has to pay for, and pretending otherwise just hid a salary. The non-recurring project revenue is sustainability: a real $120,000 of revenue that genuinely happened and genuinely will not happen again. The legal fee, the auto, and the unverifiable consulting fees are reality: costs the business actually carries, or claims it can't support.

The presented figure was $640,000. The defensible figure is closer to $440,000 to $470,000, depending on how generously you treat a couple of the partial items. Call it $470,000 to be fair to the seller. At the seller's own 3x multiple, the supportable price falls from roughly $1.9M to roughly $1.4M, a swing of about half a million dollars. The QoE that surfaced it cost, say, $18,000. That is the leverage, and it is why the buyers who skip this step are the ones who overpay.

A QoE does not always produce a finding this large; many produce a modest revision in the $20,000 to $50,000 range, and some confirm the seller's number is sound, which is itself worth knowing. But the asymmetry is the point. The worst case is a clean read that lets you proceed with confidence. The best case pays for the engagement many times over.

Where it fits in your diligence

A quality of earnings analysis is the financial core of due diligence, but it is not the whole of it. It sits alongside the legal review, the operational review, and the lease and contract work that make up financial due diligence more broadly. It is also not a substitute for being able to read a messy set of books yourself; when the records are a patchwork of broken spreadsheets and missing months, the problem shows up long before the QoE, and messy financials in due diligence covers how to spot that early. The QoE is the instrument that turns a seller's earnings claim into a number you can defend to a lender, to an investor, and to yourself.

What it ultimately gives you is permission to stop trusting a figure on faith. The seller's number is the seller's number: a starting point, assembled by the party with the most reason to make it large. Quality of earnings is the discipline of testing it, line by line, against the records that either support it or don't.

The takeaway

Quality of earnings is not about whether the books are correct. It is about whether the earnings are real, whether they will last, and whether they survive the seller walking out the door. A QoE normalizes owner compensation, weighs each add-back on its merits, isolates the one-time from the recurring, tests the durability and concentration of the revenue, and sets a working-capital peg that decides who keeps the cash at closing. The result is the one number that should drive your price and your loan, with every adjustment tied to its evidence.

Test the earnings before you sign a price. Independently when the deal size and the people at the table call for it, with a disciplined SDE schedule when they don't, but always before you commit. The number the seller hands you is where the analysis starts, never where it ends.

This article is general information, not financial, legal, tax, or investment advice. QoE cost, scope, and findings vary by deal, provider, and the quality of the seller's records. Consult a qualified CPA, transaction advisor, or attorney before relying on any earnings analysis.