After Q4: The January Plan for Marketplace Sellers

The fourth quarter ends on 31 December and the work it makes does not. The returns from December arrive in January. The peak surcharge on Amazon runs to 14 January. The stock that did not sell starts paying storage in the first week of the year, and on Amazon the long term storage fee assessment falls in the middle of every month. And the decisions that shape the next fourth quarter, what to order, what to drop, what to fix, are made from figures that are only readable in January.

This is the plan for the eight weeks after the holiday, for a seller on any of the marketplaces, in the order the work arrives. It is the half of the Q4 plan that most sellers never write, and it is where the quarter's profit is either kept or given back.

January is for closing the fourth quarter properly, in the order the work becomes possible. The first two weeks belong to returns: inspect every one, refund inside each marketplace's window, count returns by product and by reason, and on Amazon reconcile the season's reimbursements while the claim windows are still open, because peak receiving and holiday refunds are where the errors happen.

Then the stock that did not sell, decided product by product from its months of cover at January's rate of sale, its storage cost and what it would carry into next season. Then reviews, feedback and the account's health. The real fourth quarter margin by product can only be read once the returns are in and Amazon's peak surcharge has ended, on 14 January 2027 for holiday 2026, which makes the second half of the month the time for the numbers.

And the last weeks start next year's fourth quarter: what to order, when, and what this one taught that the plan should change.

Holiday returns arrive in the first fortnight of January in the volume of a small sales month, and every marketplace extends its returns window across the season, so the customer who bought in November is inside it. The desk has three jobs. Inspect, so a unit that came back sealed is relisted and one that did not is graded. Refund inside the platform's window, because a late refund is a case, and a case is a metric. And count, by product and by reason, because the return reason is the cheapest product research a seller gets all year.

On Amazon the returns also generate the reimbursements the season owes: units returned to the wrong condition, refunds issued without the return arriving, inventory lost or damaged in the peak receiving queue. The claim windows are finite and the peak volume is where the errors happen, so the January reconciliation is the year's most valuable one. FBA reimbursements: what Amazon owes you is the method; on Walmart the equivalent is in the Walmart reimbursements article.

January is when the fourth quarter's cash arrives and the next quarter's stock has to be paid for, often in the same weeks. Planning the two together avoids being short of money in a month when the sales figures look strong.

When December's sales pay out. Amazon holds a seller's proceeds until seven days after each order's delivery date as its standard reserve, and pays out what is available on the account's disbursement schedule. Late December orders delivered in the new year therefore pay out in January, and returns refunded in January come out of the same balance. Read the payout forecast against the returns, not the December sales total.

The Lunar New Year break. Lunar New Year falls on 6 February 2027. Factories in China close around it, and suppliers commonly lose two to four weeks of production, more for small ones, with freight space tight in the weeks before as every importer ships at once. Purchase orders for spring stock have to be placed and paid for in January to leave before the break.

Put the two on one sheet. Expected payouts by week, refunds expected from returns, supplier deposits and balances due, freight and duty. The week the sheet goes negative is the week a decision is needed, and January is the month it usually happens.

Ask each supplier for its last production day and restart date before the break, in writing, rather than relying on a general estimate.

Every unit still in a fulfilment centre on 1 January is a unit the season did not want at the price it was offered. The question per product is not whether to clear it but at what price and by when, because the alternatives all have dates: Amazon's long term storage fee falls on the monthly assessment day, the capacity limit for the next quarter is set from the Inventory Performance Index that this stock is lowering, and the January buyer is spending gift money and is the cheapest buyer to reach.

The decision is a table: units, months of cover at the January rate of sale, storage cost per month, the price that clears it in six weeks, and the price it would carry if held to next season. For seasonal stock the second number is usually the honest one and the storage bill is the argument. For everything else, the January price is the one to take. The mechanics of that price on Amazon are in the inventory management article, and the Walmart and eBay equivalents apply the same table with different fee lines.

The season leaves reviews behind it, and the January ones are written by the customers who had a problem, because the ones who did not have gone back to their lives. Read every one. A review that names a defect is a listing to fix; a review that names a mismatch between the page and the product is a page to fix; a review that names a delivery problem is a carrier to change. What you can do about a bad Amazon review covers which of them can be addressed and how.

The account health metrics of December are read in January, and a late shipment rate or a cancellation rate that rose in the last fortnight of the year is a warning that arrives before the action does. On Amazon the routine in the account health article applies; the point is to read the December figures before Amazon's systems do.

Campaigns built for November and December keep running into January unless somebody changes them, and the settings that suited the peak rarely suit the weeks after it.

Reset what the season raised. Budgets lifted for the holiday, bids raised to hold position against heavier competition, and placement adjustments added for deal days should all come back to their normal levels in the first week of the year. Budget rules scheduled for the season should have an end date; check that they did.

Turn off the gift searches. Keywords about gifts, stocking fillers and Christmas stop converting once the season ends, and campaigns targeting them keep spending. Pause them and add the January versions of the same products' searches instead.

Follow the new year's demand. January has its own buyers: fitness, organisation, planning, self improvement and the categories tied to resolutions. A catalogue with products in those categories should move budget toward them rather than spreading it evenly.

Read costs against the season, not against December. The cost of a click usually falls after the peak, and a cost of sale that looks better in January is often the market, not the campaigns. Compare January with last January where the data exists.

The weekly PPC routine is the right habit to restart in January, on data that is no longer distorted by the season.

The fourth quarter's figures are only readable once the returns are in and the surcharge has ended, which is the second half of January. Read them by product and by marketplace, and read five things: units by week against the forecast, margin after every fee including the surcharge and the deal fees, advertising cost per unit in the event weeks against the build weeks, the return rate, and the stock left over. The product that sold to forecast, held margin, returned little and left nothing is the one to order more of; the product that did none of those is the one that does not go back into the next plan.

The comparison across marketplaces is the one most sellers skip. A product that sold on Walmart in November at a better margin than on Amazon is telling the seller where its next unit of stock should go, and the ROAS by platform, read the way the ROAS standards article describes, is what says which advertising to do again.

A listing dressed for the holidays reads oddly by the middle of January, and some of what was added for the season works against it afterwards.

Take the season out of the words. Gift wording in titles, Item Highlights and bullets, holiday terms in the hidden search field, and seasonal claims in A+ content were right for December. In January they take space from terms that sell now, so put back the everyday wording, or the new year's, product by product.

Change the images that date it. Gallery images with festive scenes, gift wrapping or holiday props make a listing look out of date to a January shopper. Swap them for the standard set, and keep the seasonal images for next year rather than deleting them.

Close the promotions. Coupons, deals and promotional prices set for the season should end when it does. A promotion left running in January discounts sales that would have happened anyway, and on some marketplaces it changes the reference price shoppers see later.

Record what was changed. A note of each seasonal change and its date, kept with the product, turns next year's preparation into a reversal of this year's, and makes the fourth quarter's results easier to read against what the listings actually said.

The purchase orders for next October are placed in the summer, and the summer's decisions are made from January's table. So the plan for the next fourth quarter is written now, while the reasons are fresh: the products that go back in and the quantities, the deals that earned their fee and the ones that did not, the marketplace that surprised, the inbound date that was missed and why, and the one thing that would have made the most difference if it had been done in September. Written in February, that document is the difference between a fourth quarter planned and one repeated.

For the brands we run, this is the January work in marketplace management, and it is the work we would put ahead of anything else in the calendar: the returns cleared, the reimbursements claimed, the stock decided, the reviews read, the figures by product and marketplace, and next year's plan started before this year's is forgotten.