A monthly Amazon sales review that gets run is six steps long, takes about three hours by hand the first time, and ends with three written decisions. Reconcile settlements to deposits, pull units and sales by SKU, layer fees, ads, returns, and COGS onto each SKU, rank by contribution, flag what moved against last month, and write down what you will change. Reviews that skip the first step produce numbers the bank disagrees with. Reviews that skip the last step produce a spreadsheet nobody acts on.
Every Amazon deposit is a settlement, and every settlement is sales minus fees minus refunds plus reimbursements minus ad invoices minus any reserve Amazon held back. Before you analyze sales, prove the settlement ties to the bank. If it does not, something is missing from the data you are about to analyze.
Illustrative numbers. One settlement in the month:
The bank shows $27,200.15. Tied. Repeat for each settlement that closed in the month. Reading the two-week statement line by line covers what each line is and where the reserve goes.
One warning on dates. Settlements close on Amazon's cycle, not the calendar. A month can hold two deposits or three. The rest of the review runs on order date, so do not use the deposit total as "August sales."
Pull units ordered and ordered product sales by SKU for the calendar month from Business Reports, or from reconciled order data if your books carry it. Five SKUs for the example:
Total: $100,745.00. Ranked by revenue, the order is A, B, D, C, E. Hold that ranking in mind for step four.
This is the step that takes the time, because the inputs live in four places. Referral fees at 15 percent for Home and Kitchen, per Amazon's published schedule on sell.amazon.com. Fulfillment fees per unit from the settlement detail (illustrative here). Ad spend by SKU from the ad console. Refunds by SKU from the settlement. FIFO landed COGS and allocated storage from your own books.
Per-unit inputs, illustrative: fulfillment fees of $4.10, $3.95, $6.20, $5.30, and $3.30 for A through E. Landed COGS of $7.10, $3.40, $8.90, $9.60, and $2.10. Ad spend of $3,200, $1,900, $1,400, $900, and $1,600. Refund units of 28, 40, 9, 15, and 60. Storage allocations of $210, $260, $340, $480, and $90.
Take SKU-A in full. $34,986.00 minus referral $5,247.90 minus fulfillment $5,740.00 minus ads $3,200.00 minus refunds $699.72 minus COGS $9,940.00 minus storage $210.00 equals $9,948.38. Per unit, $7.11. Margin 28.4 percent.
The same arithmetic for the rest:
Catalog contribution: $20,244.77 on $100,745.00, or 20.1 percent.
By contribution the order is A, D, B, E, C. By revenue it was A, B, D, C, E. The second-biggest seller by revenue is third by profit. The fourth-biggest seller by revenue loses money on each unit. A review that stops at revenue would have called SKU-C a mid-tier product and SKU-E a small one; the contribution ranking says C is the problem and E is close behind.
Put last month beside this month, SKU by SKU, and look at the cost lines rather than the totals.
SKU-C earned $310 in July. The swing to negative $785 is $1,095. The fulfillment fee went from $5.40 to $6.20 per unit after a packaging change added a retail box; 600 units at $0.80 more is $480 of the swing. Ad spend rose $400 and units fell 70. Three causes, one product.
SKU-E's return rate went from 2.5 percent in July to 6.7 percent this month (60 of 900 units). The refund line absorbed $599.40, and at a $0.55 contribution per unit the product is one bad batch away from negative.
SKU-A and SKU-D moved within a few percent of July. SKU-B's ad spend runs at 7.0 percent of its revenue against an 18.6 percent margin; steady, and not the first fire to fight.
Write them as sentences with a number and an owner.
Three decisions. Not ten. A review that produces ten decisions produces zero, because nobody does ten things before the next review.
By hand, the six steps run about three to three and a half hours for a five-SKU catalog, and the third step scales with SKU count. At 200 SKUs, step three alone is a week, which is why most sellers run it for the top ten and guess at the rest.
The budget collapses when the first four steps are already done by the books. ConnectBooks reconciles each settlement into QuickBooks or Xero at the transaction level, so fees, refunds, storage, and ad charges post to their own accounts against the SKU that caused them, with FIFO landed COGS attached. The SKU-level profit report is step four, ready on the first of the month. What remains for you is step five and step six: reading the movers and writing the decisions. That is thirty minutes, not three hours, and thirty minutes is a review that gets run.
Step five is the part that rewards an analyst who can read all 200 SKUs at once. Crunch, the analytics AI inside ConnectBooks, takes a question such as "which products were profitable last month but are losing money this month" and runs it across reconciled data: what changed, over which periods, on which products, why, and what to do next. For the catalog above it would surface SKU-C's fee band change and SKU-E's return spike without you opening a spreadsheet. It cannot tell you that the retail box on SKU-C was a brand decision worth defending, or that SKU-E's supplier changed a mold. It flags the number and the product; you supply the reason behind the reason, and you make the call.
Running an e-commerce business comes with plenty of challenges, but ConnectBooks is here to make your life easier. With real-time insights, seamless integrations, and detailed tracking of your profitability and inventory, you can stay ahead of the game. Whether you’re selling on Amazon, Shopify, Walmart, TikTok or eBay, ConnectBooks helps you manage your finances with 100% accuracy and confidence, so you can focus on growing your business.
Ready to level up? Start making smarter, data-driven decisions every step of the way. Try ConnectBooks Free Today or Schedule a Demo
