All articles
    Financial Analysis
    9 min read••Vetting Vault Team

    Quality of Earnings: How Add-Backs Move the Purchase Price

    Walk from reported to adjusted EBITDA, see which add-backs survive a quality of earnings review, and what each one is worth at the deal multiple.

    Every lower-middle-market deal is priced on a multiple of earnings, and the earnings number is rarely the one on the tax return. Between reported EBITDA and the figure in the letter of intent sits a schedule of add-backs: owner perks, one-time costs and normalizations that the seller says the next owner won't carry. A quality of earnings review is where that schedule gets tested, one line at a time.

    The stakes are simple arithmetic. At a 5x multiple, every $100,000 of add-backs is $500,000 of purchase price. An add-back schedule that doesn't survive diligence doesn't just trim the number. It reopens the price, the structure and sometimes the whole deal, usually weeks after everyone thought the hard part was over.

    What a quality of earnings review actually tests

    A quality of earnings (QoE) report isn't an audit. An auditor asks whether the financial statements follow the accounting rules. A QoE provider asks the buyer's question: how much of this profit is real, recurring and transferable to a new owner? The work usually covers four areas, and each one can move the number.

    Recast EBITDA
    Every add-back and adjustment is tested against source documents, and the provider gives a view on which ones hold. This is the part that changes the price.
    Convert to accrual
    Many small companies keep cash-basis books. Converting to accrual puts revenue and expenses in the months they were earned and incurred, which can shift profit between years.
    Proof of cash
    Reported revenue is tied to actual bank deposits month by month. Gaps here are the fastest way to find revenue that was booked but never collected.
    Net working capital
    The same monthly balance sheets set the working capital peg for the closing true-up, so the QoE often drives that negotiation too.

    The output that moves price is the adjusted EBITDA bridge: reported EBITDA, plus or minus each adjustment, with the provider's view of which ones hold up. The working capital analysis matters too. It's covered in our guide to net working capital.

    How add-backs move the purchase price

    The bridge below follows a fictional HVAC services company with $1.4 million of reported EBITDA. The seller's banker presents $2.3 million of adjusted EBITDA. Some of the add-backs are well documented; others are optimistic. Toggle each one to see what the QoE team accepted, then move the multiple.

    Interactive · Illustrative example
    The seller says $2.30M. The QoE says $1.88M.

    Reported EBITDA plus each add-back the seller claims. Accept or reject each one, then set the multiple to see what it's worth in the purchase price.

    Show
    Reported EBITDA
    $1.40M
    Owner pay above a market-rate CEO
    Payroll records show $450K; a market-rate CEO costs about $250K.
    +$200K
    Lawsuit settlement
    Signed settlement agreement; no similar claims in three years.
    +$180K
    ERP implementation consultants
    Project invoices; the engagement ended in March.
    +$90K
    Personal expenses run through the business
    Vehicle, phone and travel receipts tie to the owner's family.
    +$70K
    “One-time” marketing push
    The GL shows a similar spend in each of the last three years.
    +$120K
    Run-rate of a new service contract
    Signed in Q4 with one month billed. Pro forma, not earnings.
    +$240K
    Rent normalized to market (QoE adjustment)
    The company rents the owner's building below market. The buyer will pay market rent.
    −$60K
    Adjusted EBITDA
    $1.88M
    Seller's price
    $11.50M
    This view
    $9.40M
    Difference
    −$2.10M
    Each $100K of EBITDA is worth $500K of price at 5.00x.
    Show the numbers
    AdjustmentAmountQoE verdictWhy
    Reported EBITDA$1.40M
    Owner pay above a market-rate CEO+$200KAcceptedPayroll records show $450K; a market-rate CEO costs about $250K.
    Lawsuit settlement+$180KAcceptedSigned settlement agreement; no similar claims in three years.
    ERP implementation consultants+$90KAcceptedProject invoices; the engagement ended in March.
    Personal expenses run through the business+$70KAcceptedVehicle, phone and travel receipts tie to the owner's family.
    “One-time” marketing push+$120KRejectedThe GL shows a similar spend in each of the last three years.
    Run-rate of a new service contract+$240KRejectedSigned in Q4 with one month billed. Pro forma, not earnings.
    Rent normalized to market (QoE adjustment)−$60KAdded by QoEOwner-owned building rented below market
    The gap between the two numbers is $420K of EBITDA. At a 5.00x multiple that's $2.10M of purchase price, which is why the add-back schedule gets more scrutiny than any other page in the CIM.

    Notice the one adjustment that cuts EBITDA. QoE work runs in both directions. When the company rents its building from the owner below market, or the owner draws less salary than a hired manager would cost, the normalized EBITDA goes down. A seller-side add-back schedule almost never includes these. A good QoE always looks for them.

    The add-backs that usually survive

    The add-backs that hold up share two traits: there's a document that proves the cost happened, and there's a history showing it won't happen again under a new owner. A QoE provider isn't judging whether an expense was sensible. They're judging whether the next owner will carry it. These three categories pass that test most often when the paperwork is in order.

    Owner compensation normalized to market

    If the owner pays themselves $450,000 to run a business a hired CEO would run for $250,000, the $200,000 difference is a legitimate add-back. The key word is normalized: the adjustment replaces the owner's pay with the cost of the role, not with zero. That distinction is the line between adjusted EBITDA and seller's discretionary earnings (SDE), which adds back the whole owner and is the usual measure for smaller, owner-operated businesses.

    Documented, genuinely non-recurring costs

    A settled lawsuit, a one-time ERP implementation or a facility move can be added back when the invoices or agreements exist and the general ledger shows no similar spend in prior years. The test is history, not description. A cost labeled "one-time" that appears every year under a different name is a recurring cost.

    Personal expenses run through the business

    Family vehicles, phones and personal travel on the company card are common in owner-run businesses and usually accepted when the seller can produce statements and identify who used what. Round-number estimates with no receipts tend to get cut in half or rejected outright.

    The add-backs that usually don't

    The weakest add-backs share a pattern: they rely on the future rather than the past, or on what the buyer could do rather than what the business did. They're also where the biggest gaps between a banker's adjusted EBITDA and the QoE's tend to open up, because a single pro forma line can be worth more than every documented add-back combined.

    Pro forma run-rate
    Annualizing a contract that was just signed, a price increase that just took effect or a cost that was just cut. A contract with one month of billing is a forecast, not earnings.
    Recurring “one-time” costs
    These fail the history test. Three years of “special” marketing is a marketing budget, whatever the ledger description says.
    Buyer synergies
    “You could close the second warehouse” belongs to the buyer. Paying the seller for it means paying for your own future work.
    Rule of thumb
    From easy to prove to hard to defend

    Where common add-backs tend to land in a quality of earnings review. Documentation moves an item left; repetition and projection move it right.

    Usually acceptedNegotiatedUsually challenged
    Documented one-time cost
    Lawsuit settlement, ERP project, a move
    Owner pay normalized to market
    Supported by payroll and a comp benchmark
    Personal expenses
    Family vehicles, phones, travel on the company card
    Related-party rent or fees
    Can raise or lower EBITDA
    “One-time” costs that recur
    Annual “special” marketing, yearly repairs
    Pro forma run-rate
    New contracts, price increases, lost customers backed out
    Buyer synergies
    “You could cut the warehouse”
    Show as a table
    Add-backExampleTypical reception
    Documented one-time costLawsuit settlement, ERP project, a moveUsually accepted with support
    Owner pay normalized to marketSupported by payroll and a comp benchmarkUsually accepted with support
    Personal expensesFamily vehicles, phones, travel on the company cardNegotiated
    Related-party rent or feesCan raise or lower EBITDANegotiated
    “One-time” costs that recurAnnual “special” marketing, yearly repairsUsually challenged
    Pro forma run-rateNew contracts, price increases, lost customers backed outUsually challenged
    Buyer synergies“You could cut the warehouse”Usually challenged
    Synergies belong to the buyer, so they rarely make it into the seller's adjusted EBITDA at all. A buyer who pays for them is paying the seller for the buyer's own work.
    Pro forma items don't have to be thrown away
    A run-rate claim that doesn't belong in adjusted EBITDA can still belong in the deal. If the seller believes in a new contract, an earnout tied to that contract's revenue lets them get paid for it once it shows up in the results, without the buyer paying for it up front.

    Adjustments that lower EBITDA

    Sellers present add-backs. A thorough quality of earnings review also finds the adjustments that go the other way, and on owner-run businesses they can be just as large. They rarely appear on the seller's schedule, not because anyone is hiding them, but because the owner has never had a reason to think of a family discount or a missing accrual as a cost.

    An underpaid owner or family staff
    If the owner takes $80,000 for a job that costs $200,000 to hire, normalizing to market cuts EBITDA by $120,000.
    Below-market related-party deals
    Rent, equipment or services from the owner's other entities at a discount won't stay discounted after closing.
    Missing accruals
    Year-end bonuses, accrued vacation, warranty reserves and annual insurance true-ups often sit outside cash-basis books entirely.
    Revenue in the wrong period
    Converting to accrual can move a big December invoice into next year, or show that deposits for unfinished work were booked as revenue.

    Deferred maintenance belongs on the same list, even though a QoE may not adjust for it directly. Repairs and replacements put off in the year before a sale flatter EBITDA, and the buyer's capex budget is where they show up. For a buyer, these are the findings that justify the fee. For a seller's advisor, they're the reason to commission a sell-side review before going to market: a negative adjustment found during exclusivity costs far more than one disclosed in the CIM.

    What to request for each add-back

    The fastest QoE is the one where the evidence for every add-back is requested on day one, next to the add-back it supports. Asking for "support for adjustments" in a single line item invites a single zip file and three rounds of follow-up. Asking for specific documents per add-back gets specific answers.

    Checklist
    What to request before you believe an add-back

    Every add-back is a claim. These are the documents that turn it into a fact.

    1Payroll registers and W-2s for every owner and family member
    2A written description of what the owner actually does each week
    3A compensation benchmark for the replacement role
    Red flag

    The replacement salary is set below what the role would really cost, or a working spouse's pay is added back too.

    Show as a table
    Add-backRequestRed flag
    Owner payPayroll registers and W-2s for every owner and family member; A written description of what the owner actually does each week; A compensation benchmark for the replacement roleThe replacement salary is set below what the role would really cost, or a working spouse's pay is added back too.
    One-time costsThe invoices, settlement agreement or contract behind the cost; General ledger detail for the same account across three years; An explanation of why it won't happen againThe same “one-time” account shows up every year, just under a different description.
    Personal expensesCredit card statements and GL detail for travel, vehicles and meals; Who uses each company vehicle and phone; Which of these the new owner would keep payingRound-number estimates with no receipts, or expenses that look like ordinary sales entertainment.
    Pro formaThe signed contract, its start date and its termination terms; Invoices and cash receipts billed under it so far; The costs needed to deliver it, not just the revenueRevenue is annualized but the cost of serving it isn't, or the customer can walk away on short notice.

    The same logic applies to the proof of cash. Request twelve to thirty-six months of bank statements alongside the monthly P&Ls, so revenue can be tied to deposits month by month instead of in a scramble at the end.

    Keeping the quality of earnings off the critical path

    QoE fieldwork commonly takes several weeks, and most of the elapsed time is waiting for documents, not analyzing them. Because the QoE feeds the price, the working capital peg and the financing, a slow one delays all three at once. It's also one of the few diligence workstreams the deal team can speed up just by asking better questions earlier.

    Get the add-back schedule before the LOI
    If the price is built on adjusted EBITDA, the list of adjustments should be on the table while the price is negotiated, not discovered after exclusivity.
    Request evidence per add-back
    Make each adjustment its own request with an owner on the sell side, so it's obvious which ones are still unsupported.
    Agree the definitions early
    The QoE's adjusted EBITDA and the purchase agreement's definitions should match, or the working capital peg and any earnout will be measured on a different number than the price.
    Share the bridge with the lender
    Acquisition lenders size debt on cash flow, so the same adjusted EBITDA feeds the debt service coverage ratio. Surprises there delay the financing as well as the deal.

    Frequently asked questions

    What is a quality of earnings report?

    A quality of earnings (QoE) report is a financial due diligence analysis that tests whether a company's reported earnings are accurate and repeatable. It recasts EBITDA by testing each add-back and adjustment, usually converts cash-basis books to accrual, reconciles revenue to bank deposits, and analyzes net working capital.

    How is a quality of earnings report different from an audit?

    An audit gives an opinion on whether historical financial statements follow accounting standards. A QoE doesn't give that opinion. It asks a buyer's question instead: how much of this profit will continue under new ownership, and what is the right EBITDA to apply a multiple to?

    What counts as an add-back?

    An add-back is an expense in the historical P&L that the next owner won't incur, so it's added back to EBITDA. Common examples are owner compensation above market, documented one-time costs and personal expenses run through the business. Adjustments can also reduce EBITDA, for example when the owner is underpaid or the company rents property from the owner below market.

    Who pays for a quality of earnings report?

    In most deals the buyer commissions and pays for a buy-side QoE, often with the lender relying on it. Some sellers commission a sell-side QoE before going to market, so their add-back schedule has already been tested when buyers start asking questions.

    Tie every add-back to its evidence

    Put each add-back on its own Vetting Vault request, so the payroll records, invoices and statements that support it arrive attached to it. Then ask Deal AI to check the P&L against the bank statements, with every answer citing the files it used.

    Get Started Free