Quality of Earnings: Red Flags Buyers Look For

When a buyer hires us to run a quality of earnings analysis, they are not asking us to re-audit the financial statements. They are asking a narrower, sharper question: of the profit you reported, how much is real, repeatable, and likely to show up again next year under the same management. A quality of earnings study, usually shortened to QoE, exists to separate durable earning power from the accounting that flatters it.

That distinction drives the price. A buyer pays a multiple of adjusted EBITDA, so every dollar of earnings that survives our scrutiny is worth several dollars of purchase price, and every dollar we strike out costs the seller the same multiple on the way down. This is why the QoE, not the audit, is where deals actually get made or repriced. We spend our time on the seams between what the numbers say and what the business does.

I have sat in this chair on both buy-side and sell-side engagements, and the pattern repeats. The companies that clear diligence cleanly are rarely the ones with the highest margins. They are the ones whose earnings tell a consistent story across the bank statements, the ledger, the contracts, and the sales pipeline. When those four disagree, we start pulling threads. Below are the threads we pull most often.

Revenue is the first place we look, because it sits at the top of every calculation and because it is the easiest number to bend without technically breaking a rule. The classic move is timing. A company pulls next quarter's shipments forward, books the revenue early, and posts a growth curve that looks better than the business underneath it. We catch this by tracing revenue back to delivery dates, signed contracts, and the point where the customer actually owed the money, not the point where someone entered an invoice.

Under current standards, revenue belongs in the period you satisfy the performance obligation. So we test that. If you sell annual software subscriptions and recognize the full year on the day of sale rather than ratably over the term, your reported revenue is running ahead of the economics, and a buyer will restate it. The same applies to long-term service contracts booked at signing, bundled deals with the discount buried in one line, and channel stuffing where product moves to a distributor who has not yet sold anything through.

None of this requires fraud. Aggressive but defensible policies still get normalized in diligence, because the buyer is pricing what the business earns, not what a generous reading of the rules permits. If your recognition policy is more forward-leaning than your peers, expect a downward adjustment, and expect it to compound at the multiple.

The mirror image of hiding bad news is parading good news that will not repeat. A large one-off order from a customer who bought once and left. A pandemic-era demand spike. A government grant, an insurance recovery, a legal settlement in your favor, a gain on selling a building. Each of these is real money that hit the income statement, and none of it tells a buyer what next year looks like.

Our job is to strip these out of the run-rate so the buyer prices a normal year rather than a lucky one. We flag them when a revenue line appears once and never again, when a customer's spend balloons for two quarters and collapses, or when margins jump in a period with no operational explanation. Sellers sometimes resist this, because it lowers the headline number. The honest framing is that a buyer will find these items anyway, and a company that has already identified and explained them reads as disciplined rather than evasive.

Concentration is not an accounting error, so it never shows up as a restatement. It shows up as risk, and risk moves the price and the deal structure. When one customer drives 40 percent of revenue, the buyer is not really buying a diversified business. They are buying a single relationship with a lot of overhead attached, and they will price the chance that the relationship walks after close.

We build the customer concentration picture early: revenue by customer over 3 years, the length and terms of the top contracts, whether those contracts survive a change of control, and how retention has trended. A top customer with a month-to-month arrangement and no switching cost is a very different asset than one locked into a multi-year agreement with penalties for leaving. The same 40 percent means two different valuations depending on how sticky it is.

Concentration on the supply side and in your people counts too. A single vendor you cannot replace, or a founder who personally holds every major relationship, both narrow the pool of buyers and invite earn-outs that keep the seller on the hook after the deal closes.

Working capital is the quiet part of diligence, and it is where a surprising amount of value quietly changes hands. Most deals set a working-capital target, or peg, and the seller delivers the business with a normal level of working capital at close. If the seller drains it in the months before closing, the buyer has to refill the tank on day one, and that is real money out of the buyer's pocket that the headline price never mentioned.

So we watch the timing. Stretching payables so vendors go unpaid at close. Pushing hard on collections to shrink receivables right before the measurement date. Letting inventory run down so shelves are thin the day the buyer takes over. Each move flatters cash today and creates a hole tomorrow. We normalize working capital across a trailing 12-month average precisely so these end-of-period maneuvers wash out and the peg reflects how the business actually runs.

The other half of this is quality of the balances themselves. Receivables that are aging past due and unlikely to collect, inventory that is obsolete but still carried at cost, and prepaid balances that are really deferred losses all inflate working capital in a way that reverses after close. We test the balances, not just the totals.

Private companies run on relationships, and some of those relationships are with the owners themselves. A building leased from the founder. A management fee paid to a holding company. A relative on payroll. A supplier owned by the seller's brother-in-law. None of these is wrong, but all of them distort earnings, because the prices were never set at arm's length.

We identify every related-party arrangement and re-price it to market. If the owner pays himself a below-market salary, reported earnings are overstated, and we add a market-rate cost back in. If the company pays above-market rent to a landlord who happens to be the owner, earnings are understated, and we adjust the other way. Either direction, the buyer needs to know what the business costs to run once the family arrangements end and normal market terms apply. Undisclosed related-party revenue is the most damaging version, because it means some of your growth was really the owner buying from the owner.

Every sell-side package arrives with an adjusted EBITDA schedule, and every schedule carries add-backs. Some are legitimate: a genuine one-time legal cost, the owner's excess compensation, a discontinued product line. The discipline is knowing where legitimate ends and wishful begins. A buyer's QoE exists in large part to test that line, and a schedule stuffed with soft add-backs does more harm than good, because it makes us question the credible ones too.

The add-backs that draw a red pen are the ones that describe ordinary business as if it were exceptional. "Pro forma" savings from synergies that have not happened. Marketing the company chose to cut. Bad-debt write-offs treated as non-recurring when they recur every year. Owner perks reclassified with no support. Restructuring charges that appear in three consecutive annual "one-time" columns. When we see the same adjustment year after year, it is not an adjustment, it is a cost of doing business, and it goes back into earnings.

The sellers who clear diligence at full price are almost always the ones who ran the analysis on themselves first. A sell-side quality of earnings study, commissioned before you go to market, does two things. It finds the problems while you still have time to fix them or frame them, and it hands a serious buyer a credible starting point instead of a blank page they will fill with their own worst assumptions.

Timing matters. Working-capital cleanup, aged-receivable collection, and revenue-policy documentation all take months, not days, and none of them can be done convincingly in the week before a data room opens. Starting 6 to 12 months ahead lets you clean up balances honestly rather than cosmetically, so the trailing averages a buyer will study already look the way you want them to.

The mindset that carries a seller through diligence is straightforward: assume every number will be tested, and get there first. When we run a sell-side QoE, we are effectively the buyer's toughest analyst, working for you, so that nothing in the buy-side report is a surprise. Deals survive on trust, and the fastest way to build it is to hand over a set of numbers that hold up the harder anyone looks.