A profit and loss statement that takes an afternoon to interpret is not doing its job. Ten questions below should each be answerable by looking, not by exporting anything into a spreadsheet first. If your P&L cannot answer them, the problem is usually the account structure rather than the business, and that is fixable in an afternoon of setup rather than a month of analysis.
1. What did I sell?
Gross sales before anything was deducted. Not the bank deposit, not the marketplace payout, not net revenue after fees.
This is the single most common defect in seller books. A $48,000 sales month that arrives as $31,000 in deposits gets recorded as $31,000 of revenue, and the $17,000 of fees vanishes into a number nobody looks at. Gross sales belongs on its own line, with every deduction visible underneath it.
2. What did the marketplace keep, broken out by type?
Referral fees, fulfillment fees, storage, and advertising should be four separate lines, not one “Amazon fees” bucket.
They behave differently and they are controlled differently. A referral fee is a fixed percentage you cannot change. A fulfillment fee responds to packaging and dimensional weight. Storage responds to how long stock sits. Advertising is discretionary. Collapsing them into one line makes all four look like a cost of doing business, which only one of them is.
3. What did the goods cost?
Cost of goods sold, computed from real unit costs including freight and duty, matched to the units that shipped in the period.
The test is whether the number moves when your landed cost moves. A seller who bought at $5.40 in February and $6.85 in August, and whose COGS line implies a single blended figure all year, is reading a margin that never existed in either month.
4. What is my gross margin by channel?
Amazon, Shopify, Walmart, TikTok Shop and eBay have different fee structures, different fulfillment economics and different return rates. A blended gross margin hides which of them is subsidizing the others.
This is usually the most actionable line on the whole statement, and it requires only that revenue and COGS carry a channel dimension. Most sellers discover their newest channel is their worst one, and that they have been pouring inventory into it on the strength of top-line growth.
5. Which SKUs lost money?
Not which sold least. Which sold at a negative contribution after fees, ads and returns.
Every catalog has some. The heavy, cheap, frequently returned item that looks fine on a revenue report is usually the culprit, because fulfillment and return costs scale with weight and unit count rather than with price. A P&L that stops at the category level cannot surface this, which is why SKU-level detail is worth the setup cost.
6. What did returns cost?
Three things happen when a unit comes back: the sale reverses, the cost of goods reverses if the unit is resellable, and a processing fee lands. Most books capture the first and miss at least one of the others.
Returns also should not sit inside a general expense line. As a percentage of gross sales, tracked monthly by channel, the return rate is an early warning about listing accuracy and product quality that no other metric provides.
7. What is advertising costing me as a share of revenue?
Ad spend against the revenue it plausibly drove, monthly, as a percentage. The absolute number tells you nothing without the denominator.
Watch the direction of travel rather than the level. A slowly climbing percentage at flat revenue means you are buying the same sales at a rising price, which is a much earlier signal than a decline in profit.
8. What is in inventory right now, and at what value?
This one is technically a balance sheet question, and it belongs here anyway, because the inventory balance is what makes the COGS figure above believable.
You should be able to state the number and the method that produced it in one sentence. FIFO, periodic average, batch costing. If nobody can say which method is in use, the gross margin on the P&L is an estimate. The IRS is direct about the standard: Publication 538 requires that “Your inventory practices must be consistent from year to year.”
9. Where is profit diverging from cash?
A profitable month can consume cash, and for a growing inventory business it usually does. Goods are paid for before they sell, marketplaces settle on a lag, and the gap widens exactly when the business is working.
The P&L will not answer this alone, which is the point of asking. If the profit figure looks healthy and the bank balance is falling, the answer is almost always inventory purchases, and the correct response is a cash flow view rather than a suspicion that the books are wrong.
10. How much of that cash is not mine?
Sales tax collected sits in the bank account and belongs to a state. So does any portion of a marketplace deposit that will be clawed back as a reserve or refund.
This matters more than it used to. Every state with a sales tax now has a marketplace facilitator law, a point the Tax Foundation made in November 2023 in noting that “every state that imposes a sales tax has adopted a means of taxing marketplace facilitators.” Where the marketplace collects and remits, the money never reaches you and should never appear as revenue. Where you sell direct, it does reach you and it is not income. Sellers who treat collected tax as working capital find out at filing time.
Why most P&Ls fail this test
Almost never because the numbers are wrong. Usually because the chart of accounts was set up for a service business and then asked to describe a product business.
The three structural fixes that resolve most of the ten questions above: book revenue gross with fee categories itemized underneath, give revenue and COGS a channel dimension, and load real unit costs so COGS is computed rather than estimated. Do those and the statement starts answering questions instead of raising them.
The tooling that does this work automatically for marketplace data is a small category, ConnectBooks, A2X and Bookkeep among the names sellers encounter most, and they differ mainly in whether they post summarized entries or itemized per-SKU detail into the ledger. Which one fits depends entirely on how many of these ten questions you need answered at the SKU level.
The one-minute standard
Open the statement and try the ten. Time yourself.
Any question that requires an export, a pivot table or a phone call to your bookkeeper is a question your reporting is not currently answering, and the gap between “the data exists somewhere” and “I can see it” is where most sellers lose the ability to act on their own numbers.
