Ecommerce Accounting Standards: 36 Checks for a Seller's Books

I run an ecommerce business on SKU.io, and I build SKU.io. In November 2026 I scored my own books against the list on this page. Two things failed that I had assumed were fine. Sales on our payout channels were being booked twice in SKU.io’s own ledger, once from the order and once from the payout. And we had not yet closed a month on a written close checklist with a second person signing off. The first is fixed. The second is the next thing on my list.

These 36 checks are what ecommerce accountants expect to find when they take over a seller’s books. They come from public talks, webinars and podcasts by ecommerce accountants and consultants, checked against our own work fixing our books. They are working practice, not accounting law. Nothing here replaces generally accepted accounting principles (GAAP), your tax adviser or your accountant.

The list works on any system. Each check says what it is, why accountants insist on it, and how to test your own books in an afternoon. Where SKU.io does something for that check and we have a live guide for it, there is one line on how. Where SKU.io does nothing for a check, the line is left out.

If you fix only one group, fix the first one. In my experience, posting marketplace income the wrong way causes more broken ecommerce books than anything else on this page.

Score your own books: the ecommerce accounting standards checklist has all 36 checks as rows in a spreadsheet, in the same order as this page.

The eight groups:

1. Payout and settlement accounting, checks 1 to 5
2. Landed cost and true COGS, checks 6 to 9
3. Opening inventory and trusted counts, checks 10 and 11
4. Go-live readiness, checks 12 to 15
5. The ecommerce P&L and chart of accounts, checks 16 to 23
6. Month-end close, checks 24 to 29
7. Shrink, returns and losses, checks 30 and 31
8. Tax and revenue timing, checks 32 to 36

Payout and settlement accounting

A marketplace deposit is not a sale. It is sales, minus fees, minus refunds, minus whatever the marketplace holds back, paid on the marketplace’s schedule. These five checks decide whether income in the ledger can be tied to the bank. The full method is in the pillar guide, how to account for marketplace payouts with a clearing account.

1. Post marketplace income as summaries, not order by order

What it is. Income reaches the ledger as one summary document per payout or per day, not as one invoice per order.

Why accountants insist on it. Xero and QuickBooks were not built to hold thousands of order invoices a month. Orders and payments do not match one to one, so every order invoice leaves an open receivable that someone has to clear by hand. No rule says the ledger must repeat every order. The order detail belongs in the system that sold it.

Check your books. Open a recent month in your ledger and count the sales invoices. If the count is close to your order count, you are posting per order. Then look at accounts receivable for a marketplace channel. A balance that grows every month, on a channel that pays you automatically, usually means orders were posted and never cleared.

In SKU.io: a channel set to Per payout books each sale once, from the payout. The order’s invoice stays on the order but posts nothing and is not sent to Xero or QuickBooks. See the payout channels and order invoices guide.

2. One gross document per settlement, fees separate, matched to the deposit

What it is. Each marketplace settlement posts at its gross value. Fees, refunds and reserves are separate lines, not netted off the sale. The document adds up to the amount that reached the bank, to the cent. A settlement that runs across month end is split, so each month carries its own part.

Why accountants insist on it. Netted income hides what the marketplace costs you. If a settlement posts as one net figure, your revenue is understated and your fees vanish. Matching to the deposit is what makes the bank reconciliation a tick instead of a search. The month-end split keeps the books on an accrual basis.

Check your books. Pick three settlements from one marketplace. For each, find the document in the ledger, the deposit on the bank statement and the settlement report from the marketplace. All three should agree to the cent, and the fee lines should appear on their own. Then pick a settlement that started in one month and paid out in the next. See which month got the revenue.

In SKU.io: Walmart settlements and TikTok Shop payouts each post as one sales invoice and one fee bill that add up to the deposit to the cent (Walmart guide, TikTok Shop guide). A settlement that spans two months posts one part per month, and the parts still add up to the one deposit. See the month-end split guide.

3. Shopify per payout, and a clearing account for every other processor

What it is. Shopify Payments income posts once per payout, fees included, matching the deposit. Every other way a customer pays you (PayPal, buy-now-pay-later providers, a second card processor) gets its own clearing account in the ledger. Money goes in when the sale happens and comes out when the processor pays you.

Why accountants insist on it. A clearing account shows at a glance what a processor still owes you. When it does not return to zero, something is missing: a fee nobody posted, a refund, a payout that never arrived. Without one, those gaps hide inside revenue.

Check your books. List every way your customers can pay. For each one, find its clearing account in the chart of accounts. Run a balance on each at the last month end. Any balance that is more than a few days of sales, or that has grown for three months, needs an explanation.

In SKU.io: Shopify stores connected now start on Per payout, which books each sale once, from the payout (guide). Stripe and PayPal payouts from a SKU.io Storefront have their own Storefront payouts guide. The month-end close includes a check that payment clearing accounts are settled (close checks reference).

4. Do not let any tool post marketplace income by calendar day

What it is. Whatever posts marketplace income should post it by settlement, not as a daily total of orders.

Why accountants insist on it. A daily order summary never matches a deposit. The marketplace pays on its own cycle, takes fees on its own schedule and holds reserves. A calendar-day total leaves the accountant to rebuild each settlement by hand. When two tools both post income (an inventory system and a settlement app, say), the same sale lands in the ledger twice.

Check your books. For each sales channel, write down which tool posts its income to the ledger. There should be exactly one per channel. If your inventory system and a settlement tool both post for the same marketplace, turn one off before anything else.

In SKU.io: on Amazon, Shopify, eBay, Walmart, TikTok Shop and Temu, a channel set to Per payout books revenue from the payout only. Amazon channels start on Per payout. See the payout channels guide.

5. Wholesale and EDI orders get one invoice per order

What it is. Business-to-business orders are the exception to check 1. Each wholesale or EDI order gets its own invoice in the ledger, and its payment status follows it.

Why accountants insist on it. A wholesale customer pays an invoice, not a settlement. Terms, partial payments, short pays and chargebacks all attach to that invoice. A receivables aging report only means something if each invoice is there.

Check your books. Run your receivables aging report. Every wholesale customer who owes you money should appear with the invoices they owe, not a single lump balance. Pick one paid invoice and confirm the payment is applied to it, not sitting unapplied.

Landed cost and true COGS

Gross margin is only as good as the cost behind each unit sold. These four checks cover what goes into that cost, when it moves to the P&L, and how purchases are controlled. The method and the worked examples are in the pillar guide, landed cost: what it is, how to calculate it, and what belongs in a unit’s cost.

6. Capitalize landed cost into inventory

What it is. A unit’s cost is its purchase price plus freight, insurance, duty and tariffs, and anything else needed to get it ready to sell. That full cost goes onto the balance sheet as inventory. It reaches the P&L only when the unit sells. When items in one shipment differ a lot in size, freight is allocated by weight or volume, not by value.

Why accountants insist on it. Expensing freight and duty in the month you pay them makes import-heavy months look like losses and later months look too good. It also makes every SKU look more profitable than it is. Allocating freight by value loads heavy, cheap items with too little freight and light, expensive items with too much.

Check your books. Find last quarter’s freight-in and duty bills. See where they were posted. If they went straight to an expense account, your inventory and gross margin are both off. Then take one product and compare its cost in your inventory system with its purchase price. If they are the same, landed cost is not in it.

In SKU.io: you can import a customs broker’s 7501 entry or an ACE ES-003 report and split the broker’s duty bill by the duty charged on each product, instead of by cost (customs entries guide). Inbound fees Amazon takes from a payout become a landed-cost bill that is part of the shipment’s cost (guide). The SKU.io team also runs a cost check against your supplier documents as a service, with the price agreed up front.

7. Post COGS by ship date, per channel

What it is. Cost of goods sold (COGS) is posted from the inventory system, dated when each order ships, as a regular journal per channel. It is not a monthly estimate or a fixed percentage of sales.

Why accountants insist on it. A fixed COGS percentage hides losing products and losing channels. A monthly snapshot of cost times quantity drifts from what was really shipped. Dating COGS by shipment ties it to the stock that actually left.

Check your books. Divide COGS by revenue for each of the last six months. A number that never moves is almost always a percentage, not a calculation. Then pull the COGS journal for one day and find the shipments behind it.

In SKU.io: COGS posts from each shipment. With the optional fulfillment basis, revenue moves to Sales in the same entry as its COGS (on payout channels: Amazon, eBay, Shopify and Walmart, not TikTok Shop or Temu). See the revenue recognition at fulfillment guide.

8. A purchase order makes the commitment visible before the bill arrives

What it is. Every inventory purchase starts with a purchase order (PO). Once issued, the PO shows up in what you owe, so payables and cash commitments are visible before the supplier’s invoice arrives.

Why accountants insist on it. Supplier invoices often come weeks after the goods ship, and deposits are paid before that. If nothing is recorded until the invoice comes, cash looks better than it is and the month’s inventory is incomplete.

Check your books. Ask your buyer for every open PO. Add up what is still owed on them. Compare that figure with what your ledger and your cash plan show as committed. The gap is money you have spent without the books knowing.

9. Approve bills against the purchase order and keep every supplier invoice

What it is. A supplier bill is paid only after it is checked against its PO and its receipt, and approved by someone other than the person who entered it. The supplier’s invoice is kept with the bill.

Why accountants insist on it. Paying from the invoice alone is how overbilling, double billing and freight charged twice get through. Buyers, auditors and tax examiners all ask for the supplier invoice behind a cost.

Check your books. Pick five supplier payments from last month. For each, find the PO, the receipt and the invoice. Note who approved the payment. If any of the four is missing, so is the control. The three-way match explainer covers how supplier deposits fit in.

In SKU.io: bill payment approval policies apply to bills, and with Mercury connected each bill payment is approved before it goes out (Mercury guide). Bills for costs with no PO, such as rent or 3PL fees, have a duplicate check (expense bills guide).

Opening inventory and trusted counts

The numbers you start a system with decide every margin that follows. When you switch systems, the quantity is usually known and the cost usually is not. The pillar guide, opening inventory: how to set your beginning inventory value when you switch systems, covers both.

10. The inventory system is the source of truth, and every SKU has a cost

What it is. POs, costs and quantities live in one inventory system, for every location including FBA. Every purchase has a PO. Every SKU has a cost, and that cost has a document behind it.

Why accountants insist on it. A SKU with no cost posts zero COGS when it sells, which inflates margin on every sale. Two systems that each hold part of the cost give two answers to one question.

Check your books. Export your product list with costs. Count the SKUs with a zero or blank cost. Then pick ten products and ask where each cost came from: a supplier invoice, a PO price, an estimate or a guess.

In SKU.io: before accounting is first turned on, the go-live readiness check blocks on missing product costs. Cost trust shows what share of stock value has a documented cost, weighted by source, and labels it High, Medium, Low or Estimate. It does not include FBA or AWD stock yet (cost and count trust guide).

11. Verify your 3PL’s and FBA’s counts

What it is. Counts reported by a third-party warehouse (3PL) or by Amazon FBA receiving are checked against what you expected to arrive and against your own physical counts.

Why accountants insist on it. A 3PL’s report is a claim, not a count. Short receipts, units received under the wrong SKU and stock that drifts away over months all start as small gaps in someone else’s system. They end as a write-off in yours.

Check your books. For your last three inbound shipments, compare units shipped, units the warehouse says it received, and units now on hand. Then ask when each warehouse was last physically counted by someone you trust.

In SKU.io: count trust is worked out per warehouse per month from physical counts, and each 3PL gets an audit trail rating. These are warnings, never blocks, and FBA and AWD stock is not included yet (cost and count trust guide).

Go-live readiness

These four checks are about the moment you connect an inventory system to the ledger, and the foundations under it. The pillar guide is inventory audit before go-live: are your costs and counts clean enough for the ledger?

12. Connect the inventory system to the ledger only after the data is clean

What it is. The inventory system starts posting to the ledger only after it is fully set up, the data has been cleaned and your accountant has reviewed it.

Why accountants insist on it. Bad costs and wrong counts, once posted, become journal entries that someone must find and reverse. For years many ecommerce accountants refused to connect inventory tools to the ledger at all. The ones who do now set conditions first.

Check your books. Before switching on any sync, write down three things: the start date, who signed off on the opening counts and costs, and what your accountant reviewed. If any of the three is blank, you are not ready.

In SKU.io: the go-live readiness check runs before accounting is first turned on. It blocks on product costs, payout codes with no account and the inventory tie-out to the ledger, and warns on stale POs, unfinished receipts and foreign-currency bills. The books health check compares SKU.io’s figures with your Xero or QuickBooks trial balance month by month, read-only.

13. A separate entity, bank account and card for the business

What it is. The business has its own legal entity, tax ID, bank account and card. Personal and business money never mix.

Why accountants insist on it. Mixed spending turns every month’s bookkeeping into a sorting job. It weakens the liability protection the entity gives you, and it makes the books much harder to sell or borrow against.

Check your books. Scan last month’s business card and bank statements. Count the personal charges. Then scan your personal account for business costs. The target for both is zero.

14. Every process runs from a written checklist

What it is. Recurring finance work (posting, reconciling, closing, paying bills) follows a written checklist, a standard operating procedure (SOP).

Why accountants insist on it. A process that lives in one person’s head stops when they are away. It also cannot be automated, because nobody can say what the steps are.

Check your books. List the finance tasks done every month. For each, find the written steps. If you cannot hand a task to a new hire with only the document, it is not written down.

15. Pick Xero or QuickBooks, and choose tools by fit, not by name

What it is. One general ledger, Xero or QuickBooks Online, not both. Settlement tools, an inventory system and later an ERP are added as volume needs them. The person who sets the system up matters as much as the software.

Why accountants insist on it. Both ledgers work for ecommerce, and migrating between them rarely pays back. Buying an ERP sized for a much larger company adds cost and setup time without fixing the books. A poor setup on a good tool still produces poor books.

Check your books. Write down every finance tool you pay for and what it posts to the ledger. Any tool that posts nothing and that nobody opens can likely go. Two tools posting the same thing is a problem (see check 4).

In SKU.io: SKU.io sends its accounting to Xero or QuickBooks Online. It does not replace either one, or your accountant.

The ecommerce P&L and chart of accounts

These eight checks are about the reports you run the business from. The pillar guide is ecommerce chart of accounts and P&L: a short chart that reports by month and by channel.

16. A short chart of accounts, with channel tracking

What it is. Income is a handful of accounts: sales, shipping income, discounts, refunds and reimbursements. Marketplace fees have their own lines. Channels are a tracking category (Xero) or classes (QuickBooks), not a separate income account per channel.

Why accountants insist on it. A chart with a dozen income accounts per channel is hard to read and harder to keep consistent. A short chart with a channel tag gives the same answer by channel and a clean total.

Check your books. Count your income accounts and compare them with the five above. For each extra one, see whether it exists only to split out a channel. Check whether your ledger has a channel tracking category or class at all.

In SKU.io: new accounts start with an ecommerce chart. Existing accounts can apply the ecommerce standard from the nominal code mappings settings after a full preview. It only adds what is missing and never remaps, renames or deletes what is already set (release note). Revenue, COGS, marketplace fees and FBA inventory can carry a channel tracking category in Xero or classes in QuickBooks (channel tracking guide). Account rules by category, brand, supplier or product type send new products to the right accounts (account rules guide).

17. The ledger holds dollars, and per-SKU profit lives elsewhere

What it is. The general ledger records totals. Profit by SKU and by order is worked out in the inventory system or a reporting tool, not in Xero or QuickBooks.

Why accountants insist on it. Forcing product detail into the ledger means thousands of accounts or items, a slow file and a chart nobody can read. The ledger’s job is a correct total. Product profit is a different question.

Check your books. Ask where you would go to see last month’s profit on your top ten SKUs. If the answer is “a spreadsheet someone builds from the ledger”, it lives in the wrong place.

In SKU.io: on payout channels, Xero or QuickBooks receives the payout documents, not each order. The ecommerce P&L reports by month and by channel from SKU.io’s own records.

18. Management accounts on an accrual basis

What it is. The P&L you run the business from is on an accrual basis: income in the month it was earned, costs in the month they were incurred. This holds even if your tax return is filed on a cash basis.

Why accountants insist on it. A cash-basis P&L moves with when you paid suppliers and when marketplaces paid you, not with how the business did. Buyers of a business value the accrual P&L.

Check your books. Look at a month when you paid for a large inventory order. If the P&L shows a loss that month and a big profit the next, it is cash-based.

In SKU.io: a settlement or payout that spans two months posts one part per month, dated within each month, and the parts still add up to the deposit (month-end split guide).

19. A weekly cash projection

What it is. A cash projection at least four to six weeks out, updated weekly, with a bad case beside the expected one. It is kept separate from the P&L.

Why accountants insist on it. A profitable seller can still run out of cash, because inventory is paid for months before it sells. The P&L will not warn you. A projection will.

Check your books. Ask what your bank balance will be six weeks from today, and what that figure is based on. If nobody can answer in five minutes, you do not have a projection.

In SKU.io: the cash-flow calendar shows cash by month from the starting cash you enter, expected sales payouts, and supplier deposits and balances due. It is a purchasing view, not a cash forecast: it has no payroll, operating expense or tax lines, so the weekly projection still belongs in your own model.

20. Contribution margin per channel, in one or two dashboards

What it is. Once there are several channels, each one is measured by contribution margin: revenue, minus COGS, minus that channel’s fees, shipping, ads and other direct costs. One or two dashboards show it. The owner owns the data.

Why accountants insist on it. Gross margin makes most channels look fine. Contribution margin shows which ones pay for themselves after fees and ads. Twenty dashboards nobody trusts are worse than one that ties to the books.

Check your books. For each channel, write down last month’s contribution margin in dollars. If you cannot, or if you need a week to work it out, this check fails.

In SKU.io: the ecommerce P&L’s By channel by month tab shows revenue, COGS, gross profit, order costs, allocated indirect costs and contribution margin. It exports as one Excel workbook.

21. Books ready for a sale or a lender, years ahead

What it is. Accrual books, a steady COGS percentage you can explain, contribution margin per channel, a list of add-backs, every supplier invoice kept, and a note on each unusual month. You start keeping them this way two to three years before you might sell.

Why accountants insist on it. A buyer’s diligence team rebuilds your numbers from source. Every month they cannot rebuild lowers the price or kills the deal.

Check your books. Pick a month from two years ago. Try to produce its P&L, the supplier invoices behind its COGS, and an explanation of anything odd in it. Time how long it takes.

22. Recalibrate dashboards against the closed books

What it is. Your sales dashboards and analytics tools are compared with the closed books each month. They should agree within about 5%. Voided and canceled orders are not revenue.

Why accountants insist on it. Dashboards count orders. Books count what was earned. When the two drift apart, decisions get made on the wrong number.

Check your books. Take last month’s revenue from your main dashboard and from your closed P&L. Work out the difference as a percentage. Anything over 5% needs an explanation, and canceled orders are the usual first suspect.

In SKU.io: the books health check compares SKU.io’s figures with the Xero or QuickBooks trial balance month by month and grades each difference against bands, with a close match at 5% or less by default. The ecommerce P&L labels each month Operational, Ledger, History or Estimate, so you can see which numbers came from where.

23. Watch ad spend against a planned end date

What it is. Ad spend for a launch or a test has a budget and a date it is meant to end. Someone reviews it every month against that plan.

Why accountants insist on it. Launch spend that never ends is one of the most common ways a growing seller loses money while sales go up. The person reviewing the books is often the first to ask when it stops.

Check your books. List every ad campaign over a set monthly amount. For each, write down why it started and when it is meant to end. Campaigns with no answer to either are where to look first.

Month-end close

A close is the point where a month’s numbers stop moving and someone signs for them. These six checks cover the tie-outs, the checklist and the review. The pillar guide is month-end close for ecommerce: the checklist, the review and the inventory tie-out.

24. Tie inventory to the ledger every month

What it is. At each month end, the inventory value in your inventory system is compared with the inventory balance in the ledger. The difference is explained and adjusted. Inbound freight is the usual cause.

Why accountants insist on it. Inventory is often the largest asset on an ecommerce balance sheet. If it does not tie, either COGS or the balance sheet is wrong, and usually both.

Check your books. Take the inventory valuation report at last month end and the ledger’s inventory account balance on the same date. Write down the difference. Do the same for the two months before it. A difference that keeps growing points to a posting that is missing every month.

In SKU.io: the books health check compares inventory, in-transit stock and vendor deposits with your trial balance month by month. The ecommerce P&L has a balance sheet tab with inventory at each month end beside the books. At close, the inventory-to-ledger check is a warning by default, and making it block the close is opt-in.

25. Reconcile every account to its statement

What it is. Every bank, card and merchant account is reconciled to its statement each month. A connected bank feed is not a reconciliation. When the books and the bank disagree, the bank wins.

Why accountants insist on it. A feed brings in transactions. It does not prove the closing balance matches. Missing, duplicated and miscoded transactions only show up when someone ties the balance to the statement.

Check your books. For each bank, card and processor account, compare the ledger’s balance at last month end with the statement’s closing balance. Any account that does not match to the cent is not reconciled, whatever the feed says.

In SKU.io: bank reconciliation stays in Xero or QuickBooks. For Amazon, each settlement shows its Fund Transfer status (deposited, balance owed, awaiting, not paid out and others), and the month-end close checks it (deposit check guide).

26. A written month-end close checklist

What it is. The close follows a checklist: interest, depreciation, COGS, vendor deposits, inbound inventory, POs closed, stock counts and write-offs. Each account is reviewed and closed, and then the statements are read as a whole.

Why accountants insist on it. Without a list, the same items get missed in the same months. The list is also what lets a second person check the work (see check 28).

Check your books. Ask for last month’s close checklist with each item ticked and dated. If there is no document, the close is a memory.

In SKU.io: the close checklist grades the month. Blocking checks stop the close until they are fixed or overridden with a reason, a sign-off records what is still outstanding, and a saved report pack becomes the month’s numbers of record. Months close oldest first. One hard close date locks the books, and moving it is recorded with the person and the reason (hard close date guide).

27. Close on a fixed date

What it is. Each month has a date by which the previous month is closed and reported, and the business measures itself against it.

Why accountants insist on it. Numbers that arrive two months late cannot steer anything. A fixed date also brings problems forward: a payout that has not arrived, a bill nobody entered.

Check your books. Write down the date each of the last three months was actually closed. If you cannot find the dates, or they drift later each month, set a date and hold to it.

28. A second person reviews the close

What it is. The person who prepares the close is not the one who signs it off. In a larger team there are layers of review, with someone at the top with controller-level judgment. Automation does not remove the need for review.

Why accountants insist on it. People do not catch their own mistakes, and automated postings fail quietly. A reviewer who did not do the work is the cheapest control there is.

Check your books. For each of the last three closes, write down who prepared it and who approved it. If the names are the same, or there is no approval, the review is missing.

In SKU.io: with two or more people doing accounting, the preparer and anyone who ticked an item or recorded an override cannot close the month, administrators included. Overrides are shown to the reviewer by name (review guide). The rule works between users in one SKU.io account.

29. Have the books done by an ecommerce specialist

What it is. The accounting is done, or at least reviewed, by a bookkeeper or accountant who works mainly with ecommerce sellers.

Why accountants insist on it. Marketplace settlements, landed cost, facilitator tax and FBA stock are specific problems. A generalist may not have seen them, and a tax preparer is not a bookkeeper.

Check your books. Ask whoever does your books how they post a marketplace settlement and how they handle landed cost. If either answer is vague, compare it with checks 2 and 6.

Shrink, returns and losses

Lost, damaged and returned stock has to land somewhere on the P&L. These two checks decide whether it lands where you can see it. The pillar guide is inventory shrinkage accounting: where losses, damage and returns belong on your P&L.

30. Post count and write-off adjustments from the inventory system

What it is. Stock counts and write-offs are recorded in the inventory system, which posts the adjustments to the ledger. Count physically at go-live and then on a schedule. The year-end count includes FBA stock, both sellable and unsellable.

Why accountants insist on it. Losses buried in COGS make margin look worse for no visible reason, and nobody can tell theft from damage from a counting mistake. Separate accounts by reason show where stock is going.

Check your books. Find last year’s write-offs in your ledger. See whether they sit in COGS or in their own accounts. Then find the date of your last full physical count, and whether it included FBA.

In SKU.io: losses go to accounts by reason: shrinkage, lost and stolen to Inventory Shrinkage, damaged to Inventory Damage Write-Offs, expired to Expired Stock Write-Off, and counting mistakes to COGS, for losses recorded after you map the accounts (inventory losses guide). An opening count is one initial stock take per warehouse, dated to the inventory start date (initial inventory guide).

31. FBA reimbursements, removals and unsellable stock

What it is. Amazon’s reimbursements for lost or damaged FBA stock are matched to the SKU they came from. The removals report is run on the same day each month. Unsellable stock is watched, not left to grow.

Why accountants insist on it. A reimbursement cannot correct COGS without knowing which SKU it covers. Removal orders that are booked as COGS when they are really transfers to another warehouse overstate costs. Unsellable stock is a write-off waiting to happen.

Check your books. Pull the last three months of FBA reimbursements and see how they were posted. Then find last month’s removal orders and see whether the returned units were added back to stock or expensed.

In SKU.io: FBA return-type removal orders move through Inventory In Transit, not COGS, and units Amazon discards go to the FBA inventory loss account. Disposal and liquidation orders still go to COGS (FBA removal orders guide).

Tax and revenue timing

These five checks cover when revenue counts, whose sales tax is whose, and the year-end calendar. The pillar guide is marketplace facilitator tax: how to account for sales tax the marketplace collected.

32. Pick a revenue date, and defer revenue only when orders wait to ship

What it is. Revenue is recognized either when the order is placed or when it ships, consistently. If orders ship within a few days, the difference is small, and deferring revenue adds work without changing much. Pre-orders and long lead times are different.

Why accountants insist on it. Deferred revenue is real work every month. It is worth it when a meaningful share of orders is paid for well before it ships. Otherwise it is complexity for its own sake.

Check your books. Work out the average days between order and shipment for last month. Then check how much revenue at month end was for orders not yet shipped. If it is small, order-date revenue is fine. If it is large, talk to your accountant about fulfillment-date revenue.

In SKU.io: revenue is recognized at the order date by default. For sellers who need revenue to wait for shipment, such as pre-orders or long lead times, an optional fulfillment basis moves revenue to Sales when each shipment goes out. On payout channels it covers Amazon, eBay, Shopify and Walmart, not TikTok Shop or Temu. You preview it month by month on your own recent orders before you switch (fulfillment revenue guide).

33. Sales tax you collect is a liability, not income

What it is. Sales tax you collect is held in trust for the state. It sits in a liability account and is kept out of cash you plan to spend. Every order from every channel goes into your tax data. Filings are outsourced, every notice is answered, and exemption certificates are kept for tax-exempt buyers.

Why accountants insist on it. Tax counted as revenue inflates sales and then shows up as a surprise bill. Tax data that misses a channel means filings that are wrong from the start.

Check your books. Find the sales tax liability account and compare its balance with what your last returns said you owed. Then check that every sales channel, wholesale included, feeds your tax tool.

In SKU.io: buyer-paid tax on eBay and Shopify payout lines goes to the Sales Tax account, never to revenue (payout mappings guide). If you use Numeral, SKU.io sends it every sale and credit from every channel, and Numeral does the registering, filing and paying (Numeral guide).

34. Tax the marketplace collected and remitted is not yours

What it is. Under marketplace facilitator laws, the marketplace collects and pays the sales tax on its sales. That tax is not the seller’s liability, even though it passes through the settlement. You may still have to file returns that report those sales.

Why accountants insist on it. If marketplace-collected tax lands in your sales tax liability, your books say you owe tax you do not owe. If it lands in revenue, your sales are overstated.

Check your books. Find the tax lines on a marketplace settlement and follow them into the ledger. They should net to zero in their own account, not add to sales tax payable or to revenue.

In SKU.io: tax a marketplace collects and remits posts to a separate Marketplace Facilitator Tax account twice, once collected and once remitted, so it nets to zero and stays out of Sales Tax. Tax you collect yourself still posts to Sales Tax. The month-end close checks it (facilitator tax guide).

35. 1099s: W-9s at onboarding, forms by January 31

What it is. A W-9 is collected from every US contractor before their first payment. Form 1099-NEC goes out by January 31. Payments made through card processors and marketplaces are reported to you on 1099-K, and each 1099-K is reconciled to the revenue you booked for that channel.

Why accountants insist on it. Chasing W-9s in January is slow and often fails. A 1099-K that does not match your books is what a tax examiner asks about first.

Check your books. List everyone you paid as a contractor last year and see whether you have a W-9 for each. Then put each marketplace’s 1099-K beside that channel’s booked revenue and explain the difference.

36. Plan taxes in November or December

What it is. Meet with your tax preparer before the year ends, while there is still time to act. Set cash aside through the year for the spring tax bill.

Why accountants insist on it. By the time the return is being prepared, every choice that could have lowered the bill is gone. An ecommerce seller who reinvests every dollar into stock can owe tax on profit that is sitting on a shelf.

Check your books. Check whether a planning meeting is on the calendar before December 31, and how much cash is set aside for tax today.

Who this list is for

The checks apply to any seller who holds inventory, and you can work through them on any system. SKU.io itself is built for sellers on several channels, with stock in more than one place and an accountant or bookkeeper in the picture. If you sell on one channel, under about $1M a year, a simpler inventory app such as Zoho Inventory or inFlow beside Xero or QuickBooks is likely a better fit than SKU.io. Accountants are welcome to send this page to clients as it is; what SKU.io posts and how you check it is written for you.

Frequently asked questions

What are ecommerce accounting standards?

They are the working practices ecommerce accountants apply to a seller’s books: how marketplace income is posted, what goes into the cost of a unit, how inventory ties to the ledger and how the month is closed. They are not a rulebook from a standards body. Your financial statements still follow GAAP or your country’s rules, and your accountant decides how those apply to you.

Do I need an inventory system to meet these checks?

Not for all of them. Checks 13, 14, 25 and 33 to 36 work with a ledger and good habits. Checks 6, 7, 10, 24 and 30 depend on knowing the cost of each unit and how many you have, which is very hard to keep right in a spreadsheet once you have more than one warehouse or channel.

Should my inventory system post every order to Xero or QuickBooks?

No, not for marketplace and Shopify sales. Post those as one document per payout or settlement that matches the deposit (checks 1 to 4). Wholesale and EDI orders are the exception and get one invoice each (check 5).

Where should I start if my books fail many of these?

Start with how income is posted (checks 1 to 4), because every other number depends on it. Then fix costs (checks 6 and 10). Then set up a close with a second reviewer (checks 26 and 28). The rest can follow month by month.

How often should I run this checklist?

Once in full now, and again before any system change. After that, checks 24 to 28 belong in every month’s close, and the rest once a year or when something changes, such as a new channel, warehouse or supplier country.

Does SKU.io replace my accountant or my ledger?

No. SKU.io sends its accounting to Xero or QuickBooks Online, and your accountant still owns the books, the close and the tax returns. Several checks on this page, such as 13, 21, 27, 29, 35 and 36, have nothing to do with software at all.

Can I score my own books against this list?

Yes. The ecommerce accounting standards checklist has the 36 checks as rows in the order of this page, with a column to mark each one met, partly met or not met.

If you want to see how SKU.io handles the checks it covers on your own data, book a walkthrough. The accounting overview covers the same features in one place, and what to post to QuickBooks Online goes deeper on check 1.