A quality of earnings review is an outside check on whether a business's profit is real and will repeat. For an Amazon and Shopify brand, it ties revenue to the cash that was paid out, moves refunds and fees into the right month, tests product cost and stock, adds late ad charges, and questions every one-off add-back. The result is the profit a buyer will actually pay for.
Meet Priya. She runs a made-up travel mug brand.
On a Monday in March, a buyer emails her. They love the brand.
They love the growth. They love the reviews. They love the $1.2 million profit on her summary sheet.
They offer $4.8 million. Then they hire an accounting firm to check the books.
Six weeks later, her phone rings. It is her own accountant.
The firm found $330,000 less profit. The new offer is about $3.5 million.
Nobody lied. Nobody hid anything. So where did $1.3 million go?
A quality of earnings review is an outside check on whether a business’s profit is real and will repeat. For an online brand, it often finds less profit than the seller reported.
That is not because sellers lie. Amazon and Shopify pay you in a way that makes profit look better than it is. That lasts until someone lines every dollar up with its month.
This post covers what the review is and what it checks for a brand on Amazon and Shopify. Then we walk through Priya’s numbers, line by line, and how you can get ready.
What a quality of earnings review is
A quality of earnings review, often shortened to “QoE,” is a deep check of a company’s profit. An accounting firm usually does it. A buyer or a lender usually pays for it.
It answers two questions.
- Is the profit real? Does every dollar of revenue show up as cash in the bank? Is every cost counted?
- Will it repeat? Is this profit what the business makes in a normal year? Or did a one-off event push it up?
The number it lands on is usually called adjusted EBITDA. EBITDA means earnings before interest, taxes, depreciation and amortization. In plain words, it is operating profit before the costs of debt, tax and wear on assets. “Adjusted” means the reviewer has fixed it for timing errors and one-off items. The SEC’s staff says a measure built differently from plain EBITDA should carry a different name, such as “Adjusted EBITDA” (SEC staff guidance on non-GAAP measures). That guidance is written for public companies, but private deals use the same label.
A QoE is not an audit. An audit asks whether the books follow the rules. A QoE asks whether the profit is the right number to price a business on.
Why it matters so much: many small brand sales are priced as a multiple of that adjusted profit. If you want the long version, see how consumer brands are valued. The short version is that every dollar the review removes from yearly profit can remove several dollars from the price.
What a QoE checks for an Amazon and Shopify brand
Every business gets checked for the basics. An online brand has five places where reviewers spend most of their time.
| What they check | The question they ask | Where it often goes wrong |
|---|---|---|
| Revenue tied to payouts | Does reported revenue match the cash that reached the bank? | Revenue taken from order reports, not from what was paid |
| Refunds and fees timing | Is each refund and fee in the same month as its sale? | Booked when paid, so they land a month late |
| Inventory and product cost | Is the cost per unit right, and is the stock really there? | An old unit cost, or stock that is lost, damaged or will not sell |
| Ad spend true-ups | Are all ad charges for the month in that month? | Late invoices, credits and agency fees booked later |
| One-off add-backs | Is each cost the seller wants added back truly one-off? | “One-off” costs that happen every year |
Revenue tied to payouts
Amazon’s orders report shows what customers ordered. The settlement report shows what Amazon actually paid you, after fees, refunds and holds. These are different numbers. The reviewer will start from the cash side, because cash is hard to argue with.
Amazon’s settlement report lists every charge and fee, the dates the settlement covers, and the deposit date (Amazon’s developer docs). Shopify’s payments activity report shows your transactions, fees and payouts for any date range (Shopify Help Center). Both were true as of October 2026.
The reviewer will take each Amazon settlement and each Shopify payout and match it to a bank deposit. Then they will ask why your reported revenue is higher or lower. If you have not done this yourself, read Amazon settlement report vs orders report first. It walks from $100,000 ordered to the cash that actually lands.
A “source of record” is the one system you agree is the truth for a number. For revenue in a QoE, the source of record is the bank, backed by the payout reports.
Refunds and fees timing
This is where online brands lose the most profit on paper.
Say a customer buys in December and returns the item in January. Many brands book the sale in December and the refund in January. December looks great. January looks a little worse. Over a full year it may even out, but the months that matter for the deal may not.
The same thing happens with fees. Amazon storage fees for one month are often charged in the next. Payment fees on Shopify come out of each payout, not out of each order. A reviewer will move each refund and fee back to the month of the sale it belongs to. This is called accrual timing: costs sit in the month they were earned, not the month they were paid. The IRS puts it the same way. Under an accrual method, income counts when earned and expenses when incurred, so the two match (IRS Publication 538).
Inventory and product cost
Your profit depends on what each unit cost you. A reviewer will test that number.
They will ask for landed cost: the factory price plus freight, duty and the cost to get it into a warehouse. If you used one old cost all year while freight went up, your profit was too high. Our guide to calculating COGS and landed cost covers how to build it.
They will also check the stock itself. Is it really in the warehouse? Is some of it lost or damaged at Amazon? Is some of it slow stock that will never sell at full price? Stock that cannot be sold gets written down, which means its value is cut and the cut becomes a cost. Tax rules point the same way. The IRS asks for a physical count at reasonable intervals, with the books adjusted to match (IRS Publication 538).
Ad spend true-ups
A “true-up” is a late correction that makes a number match the final bill.
Ad platforms bill on their own schedule. An agency may invoice two weeks after the month ends. Influencer fees may be paid on delivery, not when the post went live. So the month of the big sale often carries too little ad cost. The reviewer will move those charges back.
One-off add-backs
An add-back is a cost you argue a new owner will not have. Common ones are a one-time legal bill, a failed product launch, or the owner’s personal expenses run through the company.
Here is where it gets expensive.
Sellers often list a lot of add-backs. Reviewers accept the ones with proof and reject the rest. The SEC’s staff makes a similar point for public companies. It warns that a profit measure leaving out normal, recurring, cash operating costs could mislead (SEC staff guidance). The usual reasons for rejection:
- The cost happens every year. A “one-off” photo shoot that happens every spring is a normal cost.
- There is no receipt or contract to show what it was.
- A new owner would have to replace it. If the founder works for no salary, a buyer will have to pay someone. That is a cost, not an add-back.
A worked example: $1.2 million that becomes $870,000
Back to Priya. Illustrative numbers only. This is a made-up travel mug brand, not a real business.
She reports adjusted EBITDA of $1,200,000 for the last twelve months. About 70% of sales come from Amazon and 30% from the brand’s own Shopify store. A buyer’s reviewer goes through the books.
| Step | Change | Adjusted EBITDA |
|---|---|---|
| Seller’s number | $1,200,000 | |
| Revenue tied to payouts: orders that were later cancelled were still counted | −$45,000 | $1,155,000 |
| Refunds moved to the month of the sale (December returns booked in January) | −$38,000 | $1,117,000 |
| Amazon storage and long-term fees moved back a month | −$22,000 | $1,095,000 |
| Landed cost updated for higher freight from March on | −$96,000 | $999,000 |
| Slow stock and lost Amazon units written down | −$41,000 | $958,000 |
| Agency and influencer invoices moved back to the month of the work | −$34,000 | $924,000 |
| Add-backs rejected: yearly “one-off” shoot, founder salary set to zero | −$84,000 | $840,000 |
| Add-back accepted that the seller missed: one-time legal bill | +$30,000 | $870,000 |
| Reviewed number | −$330,000 | $870,000 |
Profit fell by 27.5%.
Now put that through a price. Say the buyer pays four times adjusted EBITDA. That multiple is made up for this example, not a market figure. At $1,200,000, the price would be $4.8 million. At $870,000, it is about $3.5 million. The gap is $1.32 million.
Notice what drove it. None of these items is fraud. Most are timing. The biggest single one was a unit cost that nobody updated when freight went up. And one item went the other way. Reviews do not only take away. A clean set of books can surface add-backs the seller never claimed.
Where it goes wrong
Beyond Priya’s list, a few more patterns make a reviewer slow down.
The summary sheet and the books disagree
The founder tracks profit in a spreadsheet built from Amazon and Shopify dashboards. The bookkeeper tracks it in the accounting system. They have never been matched. The reviewer will find the gap on day one, and then trust neither.
Months that were never closed
A month is “closed” when every payout is matched, every cost is booked, and nobody changes it again. If last March still moves every time someone opens it, a reviewer has to rebuild it. That costs time and fees, and it raises doubts. Our monthly close checklist lays out the steps.
One revenue word, three meanings
Ordered sales, shipped sales and settled sales are all called “revenue” somewhere. If your reports mix them, the trend you show a buyer may be built from two different things. Pick one basis per report, and say which one it is.
A big month you cannot explain
A one-time deal event or a viral video can lift a year. If you cannot show why it happened and what it cost, expect a reviewer to treat it as a one-off.
Costs that sit outside the books
Product samples sent to creators. Free stock given away. A freight bill the founder paid on a personal card. All are real costs. If they are not in the books, the reviewer will add them, and your profit falls.
How to prepare for a quality of earnings review
The good news: almost all of this can be fixed before a buyer ever calls.
Start about a year before you plan to sell or raise. Most of the work is boring, and that is the point. A buyer pays more for a business that is boring to check.
- Close every month within a few weeks of month end. Then do not reopen it without a written reason.
- Tie every Amazon settlement and Shopify payout to a bank deposit. Keep the match in one file you can hand over.
- Book refunds, fees and chargebacks in the month of the sale. Ask your bookkeeper to switch if they do not do this today.
- Keep a dated landed cost for each product. When freight or the factory price changes, record the new cost and the date it started.
- Count your stock at least once a quarter. Include stock at Amazon, in your warehouse and in transit. Write down lost and damaged units.
- Accrue ad and agency spend. At month end, estimate any bill that has not arrived yet, then correct it when it does.
- List every add-back with proof. For each one, keep the receipt and one line on why it will not repeat.
- Set a market salary for every role the owner fills. Better to cut your own profit now than have the buyer do it later.
- Think about a sell-side review. Paying for your own QoE first lets you fix or explain problems before a buyer finds them.
For a wider list that covers contracts, brand rights and team, see the brand sale readiness checklist.
Where software helps, and where it stops
Most of the work above is matching. Payout to bank. Refund to sale month. Cost to date. Software is good at matching.
Synthesis connects Amazon, Shopify, ad accounts and books like NetSuite, and checks numbers against their source. Amazon sales tie to settlements. Ad spend ties to each account. That makes the matching faster and keeps one definition for each number.
It does not replace the review. A quality of earnings report is an outside firm’s opinion, and its value comes from being independent. Software also cannot decide whether a cost is truly one-off, or count boxes in a warehouse. Those calls are yours and your accountant’s.
The short version
A quality of earnings review asks whether your profit is real and whether it will repeat. For an Amazon and Shopify brand, the answer turns on timing. Revenue, refunds, fees, product cost and ad spend all need to sit in the month they belong to, tied to the cash that reached the bank.
Most of the profit a review removes is not hidden. It is just in the wrong month. Close your months, tie your payouts, date your costs and prove your add-backs. Then the number a buyer finds is the number you already knew.
This is general guidance, not accounting or legal advice. Ask your accountant how it applies to your business.
Questions people ask
What is a quality of earnings report?
It is a report, usually written by an accounting firm, that tests whether a company’s stated profit is accurate and likely to repeat. Buyers and lenders use it to set a price or a loan size. It often lowers the profit figure the seller started with.
How long does a quality of earnings review take?
It depends on how clean the books are. A brand with closed months, tied payouts and a clear cost record moves faster. A brand that has to rebuild past months from exports will take much longer, and pay more in fees.
Who pays for a quality of earnings report?
Usually the buyer or lender orders one as part of their checks. Some sellers order their own first, called a sell-side review, so they find the problems before a buyer does. Practice varies by deal.
What is an add-back in a quality of earnings review?
An add-back is a cost the seller says will not happen again under a new owner, so it should be added back to profit. Examples are a one-time lawsuit or the owner’s personal car. Reviewers accept add-backs only with proof, and reject costs that keep coming back.
Why does profit often go down after a quality of earnings review?
Because the review moves costs into the month they belong to, adds charges that arrived late, and rejects weak add-backs. Ecommerce brands often book revenue when orders are placed but fees and refunds when they are paid. That gap flatters profit until someone lines it up.
How do I prepare my ecommerce brand for a quality of earnings review?
Close every month on time. Tie marketplace and store payouts to the bank. Book refunds and fees in the month of the sale. Keep a dated landed cost for each product. Count stock. List every add-back with a receipt. Start a year before you plan to sell.