A week or two after the LOI is signed, an email arrives with a spreadsheet attached, and the spreadsheet has somewhere between one and three hundred rows. That’s the financial diligence request list, it’s the starting gun for the longest stage of the deal, and how you experience the next eight weeks depends almost entirely on whether the list surprises you. So here’s what’s on it, what the buyer’s accountants are actually doing with each pile, and which discoveries merely move the price versus the ones that end deals.
The request list, in plain categories
Strip the accounting language and the list is five asks. The statements: three to five years of P&Ls, balance sheets, and cash flows, monthly where possible, plus current-year numbers. The proof: tax returns, bank statements, and merchant records for the same period, because statements are claims and deposits are facts. The detail: general ledger exports, revenue by client by month, AR and AP agings, payroll registers, and the invoices behind your addback schedule. The paper: client agreements, vendor contracts, leases, loan documents, and insurance policies. The context: your budget versus actuals, any prior valuations, and explanations for the anomalies they will absolutely find, like the revenue spike in March 2024 (the big project) and the margin dip that winter (the bad hire).
None of this is exotic, and that’s the first comfort: the list is nearly identical across serious buyers, which means it’s preparable, and a well-built data room answers most of it before it’s asked.
What the QoE review actually does
Somewhere in this stage, many buyers commission a quality of earnings review: an accountant’s forensic pass over your P&L whose one-sentence job is to answer “is the EBITDA real, and will it recur?” The analyst rebuilds your earnings from the transaction level up, and at MSP scale a few tests dominate the work.
Proof of cash. Reported revenue gets tied to actual bank deposits, month by month. Gaps here are the fastest way to turn a friendly process cold, which is why cash-basis books with unrecorded adjustments need cleaning up long before this stage.
Revenue recognition. Did you book revenue when you earned it or when it was convenient? The MSP classics: annual agreements billed up front and recognized immediately (that’s deferred revenue, a liability, and it will be found), project revenue pulled forward into a strong year, license pass-throughs grossed up to inflate the revenue line. None of these are rare, and the honest ones get restated without drama. The pattern the analyst is really testing for is whether the last twelve months were dressed for sale.
Recurring versus one-time, both directions. Your addbacks get audited item by item, and so does the opposite category sellers never volunteer: one-time revenue. The big nonrecurring project in the trailing twelve months gets pulled out of run-rate earnings just as surely as the lawsuit expense gets added back.
Customer economics. Revenue by client by month exposes churn, concentration, repricing history, and the anchor client’s trajectory, which is why the strict version of your MRR percentage is computed here whether you computed it or not.
The output is an adjusted EBITDA figure with the analyst’s name on it, and from that moment it’s the deal’s number. If it lands near yours, diligence was a formality. The distance between your claimed EBITDA and the QoE’s is, almost mechanically, the size of your reprice.
Working capital, the fight nobody warns you about
One more thing gets settled during financial diligence, and it surprises more first-time sellers than any other single term: the working capital peg. Short version: the buyer expects the business to come with a normal operating level of working capital (receivables and prepaids, net of payables and deferred revenue), the way a house comes with its furnace. Diligence is where “normal” gets computed from your trailing twelve months, and at close the price adjusts up or down against that peg. Sellers who don’t watch this get squeezed twice: aggressive peg-setting during diligence, then a true-up after close computed by the buyer’s accountants. It’s a big enough topic to get its own post later in the series. For now, the diligence-stage rule: when the buyer’s team starts asking detailed AR and deferred-revenue questions, the peg is being built, and your CPA should be in that conversation.
Surprises that reprice versus surprises that kill
Every seller’s fear in this stage is the discovered problem, so let’s sort the discoveries honestly.
Things that reprice: softer-than-claimed margins, addbacks that don’t survive scrutiny, a big client’s revenue drifting down, deferred revenue nobody had booked, project revenue mix higher than advertised. These change the number, sting, and close anyway. Deals absorb bad facts constantly.
Things that kill: discoveries that change what the buyer believes about you. Revenue that can’t be tied to cash. A second set of books, even a benign-sounding one. Material problems you clearly knew about and didn’t disclose: the anchor client’s termination notice, the tax lien, the lawsuit. The pattern is trust, not size. A $200k problem disclosed in week one gets structured around; a $50k problem concealed until week seven poisons every other number you’ve ever presented, because diligence can verify facts but it can only sample honesty.
Which yields the operating rule for the whole stage: disclose early, disclose yourself, and attach your explanation before someone else attaches theirs. Sellers who run this play consistently report the strange experience of diligence building the buyer’s confidence, and it shows up where you’d want it to: fewer holdbacks, lighter earnouts, faster closings.
The last word is preparation, as it always is in this series. Every document on that request list can be assembled now, this quarter, unhurried, and sellers who do so turn the scariest stage of the process into its most boring one. Boring is the goal. Boring closes.