Q4 Peak Season From China: The Cutoff Dates That Decide Your Black Friday

Q4 shipping cutoff dates from China run earlier than the carrier deadlines published in December. Black Friday is 27 November 2026. Factories close 1 to 7 October for Golden Week, and Chinese domestic express absorbed 13.9 billion parcels around 11.11 last year from 21 October. Working backwards from delivery, your real decisions land in August and September.

The only cutoff your customer actually sees

Every December, carriers publish their last-post dates and every ecommerce newsletter reprints them. For Christmas 2025, USPS recommended 17 December for Ground Advantage and First-Class Mail, 18 December for Priority Mail and 20 December for Priority Mail Express in the contiguous US. Every major postal operator publishes something similar.

Those are the least important numbers on your calendar. They describe the last link in the chain: when a parcel already inside a domestic network has to be handed over. They say nothing about when it had to leave Shenzhen, when the factory had to make it, or when you had to decide it existed.

The date your customer sees is the delivery estimate on your product page. Everything upstream is arithmetic. Get it wrong and you do not get a late parcel, you get a refund, a dispute, and a customer who bought from you once. So build the calendar the other way round: start at the promise and subtract.

The fixed dates of the 2026 Q4 calendar

These do not move. Print them.

The Chinese dates come from the General Office of the State Council and were confirmed in November 2025, along with two adjusted working days, Sunday 20 September and Saturday 10 October. Thanksgiving 2026 falls on 26 November, which puts Black Friday on Friday 27 November and Cyber Monday on Monday 30 November. Chinese New Year 2027 is 6 February.

Working backwards to your real Christmas order-by date

Here is the subtraction, on our own numbers: sub-24h dispatch from the Shenzhen warehouse, and a 3 to 10 day delivery band into the 65+ countries we ship to.

Two numbers come out of that. 13 December is the arithmetic edge. 10 December is what you should publish, because December is the one month when the slow end of every band is the realistic end.

Run the same math on Black Friday. An order placed Friday 27 November dispatches by Saturday 28th and lands between 30 November and 7 December. BFCM orders are not your December risk. They arrive with two clear weeks to spare.

That is the part most Q4 planning gets backwards. Black Friday is not a delivery problem, it is a readiness problem, and its deadline sits in September. The delivery problem starts on 14 December and belongs to customers who buy late, panic, and dispute on 26 December. What a held payment does to working capital over the next 90 days is in the cash-flow math of dropshipping.

Crunch one: Golden Week, 1 to 7 October 2026

Seven consecutive rest days. Factories close, customs slows, freight offices run skeleton staff, and your supplier goes quiet.

The published dates understate it in both directions. Many suppliers wind down before 1 October and return in stages. SEKO Logistics puts the recovery window at 8 to 16 October, and notes it can run one to two weeks longer as factories, carriers, warehouses and trucking providers clear backlogs. Flexport advises booking ocean freight 3 to 4 weeks before the holiday and air freight at least a week before, and warns of peak season surcharges, general rate increases, rolled bookings, and blank sailings immediately after because there is nothing to load.

Sunday 20 September and Saturday 10 October are official make-up working days in China, added to fund the long breaks. Sorting centres and factories genuinely open on them. They are not full-strength days. Treat 8 to 16 October as a ramp, not as throughput.

The practical consequence: your pre-Golden-Week deadline is not 1 October, it is 24 September, the day before Mid-Autumn Festival. Anything needing a human decision at a factory, a sample approval, a QC re-run, a variant change, has to close by then or it slides to mid-October.

Samples take 5 to 10 days. Request them by Monday 7 September and they land 12 to 17 September, leaving a real week to inspect, reject and re-request before the country goes quiet. Request them on 25 September and you are inspecting in mid-October with 11.11 already loading. Our sourcing process runs on that calendar, and we quote 92% of requests the same day so that decision time, not quoting time, is what consumes your September.

Crunch two: 11.11, the surge that starts on 21 October

This one catches sellers outside China because it is invisible from the outside. Singles Day looks like a one-day consumer event. It is a three-week logistics load on exactly the infrastructure your goods move through.

The numbers come from China's State Post Bureau. Between 21 October and 11 November 2025, Chinese couriers handled 13.938 billion parcels, averaging 634 million a day, 1.18 times normal daily volume, with a single-day record of 777 million. Across the full year the sector processed more than 180 billion parcels, over 6,200 a second.

Eighteen percent above normal for three weeks is not a rounding error. That load falls on the same trucking, sorting hubs and line-haul that carry your goods from a Guangdong factory to a Shenzhen warehouse. Pickups slip. Queues lengthen. The leg that normally takes two days takes four.

Overlay the two crunches:

• 1 to 7 October: production stopped
• 8 to 16 October: ramping back, clearing backlog
• 21 October to 11 November: domestic logistics running 18% hot
• 12 to 26 November: clear
• 27 to 30 November: BFCM

You get roughly five clean days, 16 to 20 October, between the end of Golden Week recovery and the start of the 11.11 ramp. That is not a planning window, it is a gap.

The window that works is 12 to 26 November: two weeks of normalised domestic capacity sitting directly in front of Black Friday. One restock, one SKU launch, one packaging changeover, land it there. Everything else closes before 20 October.

The third squeeze: air capacity going into Q4 2026

Q4 air freight out of China is tight in a normal year. 2026 is not a normal year.

Xeneta upgraded its 2026 forecast in July, reversing a December 2025 prediction that long-term rates would fall 5% to 10%. It now expects them to rise 5% to 15% across 2026. The reason is supply, not demand: in the first half of 2026 capacity grew just 1% while demand grew 4%, after the escalation of the Middle East conflict on 28 February removed roughly 12% of global air cargo capacity effectively overnight. Full-year capacity growth is now forecast at the lower end of a 2% to 3% range, with the imbalance expected to keep the market tight through year-end.

For what a normal Q4 does to spot rates, last year is the reference. Freightos reported China to North America climbing from around $5.30/kg in mid-October 2025 to $6.50/kg at end of November and past $7.50/kg in early December, above the prior year's $7.30/kg peak. China to North Europe held at $3.64/kg, mainly because carriers moved freighter capacity off the transpacific onto Asia to Europe lanes.

The conclusion is narrow. A landed cost you calculated in August is not the landed cost you will pay in November. Re-run your per-SKU math in the second week of November, before you set BFCM discount depth, not after. That number decides whether a 30% off campaign is profitable or expensive, and it is what the dropshipping calculator prices: every variant, goods plus shipping plus fees in one figure. Method in how to calculate dropshipping profit.

The month-by-month decision calendar, August to December

One decision per month. If it does not close in its month, everything downstream moves.

The August row is the one people skip. It carries no external deadline, so it feels optional. It decides whether September is a week of confirmations or a month of firefighting.

Pre-buying versus per-order: the same calendar, two different risks

If your Q4 plan is to buy inventory and warehouse it locally, the calendar is far harsher, and the reason is transit time.

Flexport's Ocean Timeliness Indicator for the week to 27 July 2026 put China to North Europe at 58.2 days, US East Coast at 57 days and US West Coast at 36 days. That measures cargo-ready date at the factory through to container departure from the destination port. Inland transport, deconsolidation and receiving sit on top.

Work that backwards to stock being sellable by 20 November. Illustrative, using those transit figures alone and ignoring destination handling:

Add production lead time on top and the European pre-buy decision was made in August 2026 with zero Q4 demand data. That is not a supply chain, that is a bet. What it ties up is covered in our breakdown of dropshipping cash flow, and the European specifics in dropshipping in Europe.

Per-order changes the shape of the risk, not the need to plan:

We do not warehouse in the US or EU. Every parcel goes China-direct from one owned Shenzhen warehouse, dispatched sub-24h, delivered in 3 to 10 days on the local carrier your customer already trusts, USPS, Royal Mail, La Poste, DHL, Australia Post, with tracking branded to you. No minimum order volume, no setup fee, no storage fee on SKUs moving on normal order flow. You pay $1 per processed order, only when a parcel ships. That is why the November decision can stay open until November. More on 3PL fulfilment and shipping.

What per-order fulfilment does not fix

The limits are the useful part.

It removes the quantity guess. It does not remove the calendar. Golden Week still closes the factory. If a supplier is shut from 30 September to 12 October, no fulfilment model conjures units out of a closed building. What per-order does is stop you pre-paying for 4,000 units of the wrong variant in August to hedge against it.

It does not remove sourcing lead time. Samples still take 5 to 10 days. QC still takes a day and still fails things, which is the point of a 99.6% pass rate. A product you have not sampled by late September is a product you are launching blind into your biggest quarter.

It does not remove supplier concentration risk. One factory, one holiday schedule, one point of failure. Two verified sources per hero SKU is cheap insurance in Q4 and expensive to arrange in October. There are 30,000+ verified factories on the network for exactly this.

It does not remove the need for an honest delivery promise. If your product page says 3 to 10 days, your December cutoff has to respect the 10, not the 3.

What it changes is when you have to be certain. On a 58-day ocean lane you have to be right in August. With sub-24h dispatch and a 3 to 10 day band, you have to be ready in September and can still be responsive in November. That difference is the entire quarter.

To map this calendar against your own SKUs before Golden Week closes the window, connect your store.