Most online sellers set the holiday return window in a hurry, copy last year’s dates, and discover the cost in late January. The window is a pricing decision disguised as a policy decision — and this year the dates matter more than usual, because three separate carrier price changes land inside the exact period your extended returns will come back.

Decide it now, in October, while you can still choose. By December 26 the window is whatever your site said it was.

Three dates that sit inside your return window

The 2026–27 calendar puts a cliff in the middle of January:

The USPS holiday prices have been in effect since October 4, 2026 and apply an 8% transportation surcharge to Priority Mail Express, Priority Mail, Ground Advantage and Parcel Select. That surcharge is on your return labels too, right through the peak return period, and it comes off at midnight on January 17.

So USPS returns get cheaper mid-January while FedEx returns get more expensive at the start of it. If your reverse logistics runs on one carrier by default, this is the year to check whether that default still holds in January.

UPS had not published a 2027 general rate increase at the time of writing. Figures above are list values and move with your own carrier agreement and discounts.

Where the window actually falls

Return windows are measured from delivery, not from purchase, which pushes them later than most sellers assume:

Read that against the carrier calendar. A 30-day window on a December 15 delivery closes on January 14 — three days before the USPS surcharge expires, meaning essentially every label in that cohort carries the 8%. Push the same policy to 45 days and a large share of returns arrive after January 17, on the cheaper rate card.

That is the counterintuitive part: a longer window can produce a lower cost per return label this year, because it moves volume past the surcharge expiry. It also raises the share of items that come back too late to resell at full price, which is the trade-off to weigh — not whether a longer window « costs more » in the abstract.

What a return actually costs you

Work it with your own numbers rather than an industry average. The structure looks like this for a store with 1,200 December orders:

• Return rate. The NRF and Happy Returns put online returns at 19.3% of online sales by dollar value. At 20% of orders, that is 240 returns.
• Paid labels. If you fund returns on 80% of them, 192 labels.
• Label cost. At $8.00 per label, $1,536 in postage alone.
• Surcharge effect. Removing the 8% USPS surcharge takes an $8.00 label to about $7.41. Moving 77 of those labels past January 17 saves roughly $45.
• FedEx effect. A 5.9% increase adds about $0.47 per label — near $90 across 192 labels.

The postage swing is small in isolation. The reason to run it anyway is that it points at the much larger number sitting behind it: processing, restocking, markdown on anything that comes back after the season, and the share you cannot resell at all. Postage is usually 15 to 25% of the fully loaded cost of a return.

Carrier-side details worth knowing before January

USPS return services carry $100 of insurance included, with coverage available up to $5,000 for a fee, and handle items up to 70 lb. Transit runs 1–3 days on Priority Mail Express Return, 2–3 days on Priority Mail Return, and 2–5 days on Ground Advantage Return.

Two fees are easy to avoid. USPS Label Delivery — where the carrier prints and delivers a label to a customer who cannot print one — costs $1.65 per label. And a noncompliant package quality charge of $0.25 per label applies to labels that do not meet print standards, which is a quiet tax on sellers generating labels from a poorly configured template.

On the FedEx side, the $25 Paper Document Fee starting January 18 is aimed squarely at anyone still attaching printed commercial documents to international shipments. If any part of your returns flow is cross-border and still paper-based, that is a per-shipment fee on work a document upload eliminates.

Setting the window: three workable policies

Standard 30 days from delivery. Simplest, lowest late-return risk, and the cheapest on markdown. Nearly all returns land before January 17, so budget them at surcharged rates. Best for perishable-relevance categories and fast-moving fashion.

Extended to January 31 for gifts bought after November 1. The most common retail pattern and the easiest to communicate. It straddles the surcharge expiry, so roughly half your labels land on each rate card, and it removes the « bought early for Christmas » complaint entirely.

Extended to February 15. Strong conversion lift on gifting, and almost all labels land after January 17. It is also the version most exposed to resale loss — an item back in mid-February from a seasonal line is usually a markdown, and the $0.59 saved per label does not cover a 30% price cut.

Three things to do this week

Pick the window date explicitly and publish it on the product page, not only in the policy footer — gift buyers check before they buy, not after.

Price your return labels for both sides of January 17 and tell whoever manages carrier accounts which cohort lands where, so nobody is surprised by a January invoice.

Check whether your return label generation is on a carrier that gets cheaper in January or one that gets more expensive, and whether switching the reverse leg alone is worth doing. Outbound and returns do not have to be on the same carrier.

Want the return window, label costs and January rate changes modeled against your own order volume and lanes? HereWeShip runs the numbers with sellers before peak, while the policy is still a choice.