If you run a platform with a pipeline of tuck-ins, you have already had this argument with yourself.
Every target needs its numbers verified. Almost none of them can carry a full-scope quality of earnings fee. A $30,000 report on a $4 million add-on is a real share of the equity check, and doing it five times a year on deals that do not all close is a budget line nobody wants to defend.
So the practical answer has usually been to sample. Look at three months. Tie the bank to the ledger for the periods that look representative. Move on. That works until it doesn’t, and when it doesn’t, it fails in the specific way sampling always fails: the problem was in the months you didn’t look at.
Separate the two halves of the work
The way out is to stop treating “quality of earnings” as a single indivisible product. It has two halves that behave completely differently on cost.
The mechanical half is data preparation. Reconstructing cash receipts and disbursements. Matching bank activity to the general ledger. Mapping a trial balance to a standard chart of accounts. Reading leases and vendor agreements into a schedule. This work scales with record quality — how many accounts, how bad the scans, how many months — and barely at all with deal size. A $4 million business with six accounts and scanned statements can require as much preparation as one four times larger.
The judgment half is analysis. Whether an add-back is defensible. Whether a customer relationship survives a change of control. Whether management’s explanation of a margin swing holds. Whether the working capital peg is being set on a number that means anything. This work scales with complexity and with what is at stake.
Historically both halves were priced together because both were done by the same people at the same hourly rate. That is the only reason a small deal ever cost what it cost.
Scope to the decision you are making
Pre-LOI, deciding whether to proceed
You need to know whether reported earnings are real. That is a proof of cash: does the money that moved match the money they say moved. You do not need a normalized EBITDA schedule to decide whether to keep going. You need to know the base is not fiction.
Under LOI, deciding what to pay
Now you need the base and the adjustments — normalized EBITDA, working capital and operating cycle metrics, customer concentration, change-of-control review. This is where judgment earns its fee, and it should be applied to a verified base rather than to management’s schedule.
Financing, satisfying a third party
If there is SBA debt in the structure at $3 million or more of purchase price, the scope is no longer your choice. From October 1, 2026, SOP 50 10 8.1 requires a quality of earnings report including a cash proof, and the lender must use its earnings figure in the debt service coverage calculation. See the SBA quality of earnings requirement.
What we would recommend for a tuck-in program
Run the mechanical half on every target, early and cheaply. Full period, all accounts — not a sample. It is the step that finds the deals that should die, and it is worth far more at the top of the funnel than at the bottom.
Then apply the judgment half only to the targets that survive, scoped to the decision in front of you.
That inverts the usual pattern, where the cheap look is shallow and the deep look comes late. A full-period proof of cash run early is both cheaper and more diagnostic than a sampled review run under LOI, because sampling is precisely the thing that lets a bad month hide.
Our published pricing reflects that split: a rapid cash review for the pre-LOI screen, Proof-of-Cash Plus at $10,000 flat for a full reconciliation with GAAP-formatted statements, and QoE Essentials when the deal warrants the full analytical layer.
The honest caveat
Automating the mechanical half does not make the judgment half optional, and any provider suggesting otherwise is selling you a number rather than a conclusion. What it does is change the arithmetic that made verification unaffordable on small deals — so the answer to “do we need a full QoE on this add-on” can be no, without the answer to “did we verify the cash” also being no.
Talk to us about scoping a tuck-in program.