The Importer’s Calendar: When to Order to Make the Season

6 February 2027. Official holidays run about a week; the real shutdown is far longer.

What actually happens. Two to three weeks before the date, factories start winding down: workers leave for their home provinces, new orders stop being accepted, and priority goes to shipping what has already been made. Domestic Chinese logistics jam, because everyone is moving goods at once.

After the holiday, production does not switch back on in a day. Some workers never return — that is normal in China, people change factory and city over the break. The factory rehires and retrains, and reaches full capacity three to four weeks later.

Roughly six weeks disappear — from mid-January to early March.

What this means for you.

• An order placed in December still makes it.
• An order placed after 1 January will most likely ship in March.
• Watch quality at two specific points: the last week before the holiday, when the line is racing the plan, and the first weeks after, when new staff are working it. These are the only two periods of the year when we tighten inspection at intake.

Black Friday falls on 27 November 2026, followed by Christmas and New Year. For toy retail and marketplaces this is around sixty percent of annual revenue.

Counting backwards from the shelf.

Practical cut-offs for November 2026:

• Sea to the US — order now, late July through early September. After that you are paying air rates.
• Rail and road to Europe — order by mid-September.
• Air — holds until mid-October, but peak-season rates climb and capacity is booked out in advance.
• EAEU by road — take the rate and transit time for your lane from the cart and add a month for production and consolidation.

On marketplaces specifically. Amazon and other platforms publish their seasonal inventory receive-by date in late October or early November, and it is always earlier than sellers expect. Take that date from Seller Central and count back from it, not from Black Friday.

On rates. From September to November freight prices rise and capacity tightens, because everyone ships at once. It is the one time of year when transit times slip across every carrier simultaneously.

• 1–7 October, National Day. Factories close for the week — exactly when you are finishing Q4 production. Build it into the plan in advance.
• 1–5 May, Labour Day. A short stop, usually without consequences.
• Mid-Autumn Festival, September or October, date varies. A couple of days.

None compares with Chinese New Year, but combined with peak season, Golden Week costs you a full week of schedule.

A buyer who orders the same goods from the same supplier six weeks apart can pay the same money and get two completely different outcomes. The difference is not negotiation. It is where in the Chinese manufacturing year the order landed.

Chinese factories do not run at an even pace. They run in waves shaped by two things: the export peak that feeds the Western fourth quarter, and the national holidays that empty the workshops entirely. Everything else — lead time, quality consistency, willingness to take a small order, even how fast anybody answers an email — moves with those waves.

In the quiet months a factory has spare capacity and a reason to be helpful. Lead times are short, small orders are welcome, and quality is at its most consistent because nobody is rushing.

In the peak the same factory has more orders than machines. Lead times stretch, small orders go to the back of the queue behind large ones, and the pressure to hit shipping dates is exactly the pressure under which mistakes happen — the wrong colour batch, the missed inspection, the substituted component.

In the shutdown nothing happens at all. Not slower — nothing. Machines are off, workers have travelled home across the country, and the warehouse is locked.

The practical conclusion is simple and almost nobody acts on it: the cheapest improvement available to a small importer is not a better price, it is ordering six weeks earlier. It costs nothing and it moves you out of the queue.

Work from the shelf date, not the order date. Four blocks of time sit between them, and each one has to be counted separately.

Production or picking. Catalogue goods that are in stock skip this. Anything that has to be produced adds the factory's own lead time, which is the number that stretches most in peak season.

Consolidation. Goods from several suppliers converging on one warehouse, then being checked line by line. Days, not weeks — but not zero, and longer when an order touches many suppliers.

Freight. The big block, and entirely dependent on route and mode. Sea is weeks, rail and road are somewhere in the middle, air is days. The ranges by region are on the shipping page.

Customs and inland delivery. Usually short, occasionally not, and never predictable to the day.

Add them up, then add a buffer that reflects the season: a small one in the quiet months, a generous one anywhere near a shutdown or a peak. A schedule with no buffer is not a schedule, it is a hope.

Right after a shutdown. Workshops restart gradually, and the first weeks are uneven as workers return. This is a period for reordering known goods, not for launching something new that depends on a precise date.

The long quiet stretch. The best time in the year to test. Capacity is available, attention is available, and a small assortment order gets treated properly. If you are going to try twenty new models, this is when.

The run-up to the export peak. Order what you already know sells, and order it early. Every week of delay here costs more than it would at any other point in the year, because you are joining a queue that only gets longer.

Inside the peak. Expect stretched lead times and plan around them rather than against them. This is not the moment to introduce a new supplier, a new category or a tight deadline all at once.

Approaching a shutdown. The last shipping window closes earlier than people expect, because the freight forwarders fill up before the factories close. Being a fortnight early here is worth more than any discount.

Ordering into a closed factory. The single most common one. The order is accepted, the money moves, and the goods do not — because there is nobody in the building. Nothing is wrong and nothing can be fixed until the workshops reopen.

Blaming a supplier for the season. A lead time that doubles in peak is the market, not misconduct. Buyers who change supplier over it often discover the new one behaves identically, having lost a relationship for nothing.

Booking freight late. Capacity tightens before the holidays on every mode. A shipment ready on time and unbooked is still a shipment that misses the season.

Planning to a single date. Ports congest, customs inspect, weather closes routes. Plan to a range, communicate a range, and keep the buffer where you can see it.

Two fixed events dominate the Chinese manufacturing year, and their effects reach much further than the dates themselves suggest.

Chinese New Year is the larger of the two by a wide margin. It is not a public holiday in the Western sense, it is a national migration: hundreds of millions of workers travel home, factories close entirely, and many workers do not return to the same employer afterwards. The effect on production starts weeks before the official dates, as factories stop accepting orders they cannot finish, and continues for weeks after, as workshops rebuild their workforce and restore consistency. An importer who treats this as a few days off will be surprised twice: once by how early the window closes, and once by how gradually output recovers.

Golden Week at the start of October is shorter and far less disruptive, but it lands at the worst possible moment — inside the run-up to the Western fourth quarter, when every hour of capacity is already spoken for. A week lost in early October is a week that cannot be made up before the season, which is why goods intended for the year's peak selling period should be ordered well before it rather than around it.

The two events also interact with freight rather than only with production. Forwarders and carriers fill their capacity ahead of both, so the last workable booking date is meaningfully earlier than the last production date. A shipment that is finished on time and unbooked has still missed the window.

The dates published on this page are the industry's. The dates that matter are yours, and building the second from the first is a half-hour exercise you do once a year.

Start from your own selling peaks, not from China's. A shop with a December peak, a marketplace seller with a back-to-school peak and a distributor with a summer peak all need different order dates from the same factory calendar.

Subtract the transit range for your route, using the far end of the range rather than the near one. This is the single largest block of time and the one most often underestimated.

Subtract consolidation and checking. Longer when an order touches many suppliers, which good assortment orders always do.

Subtract production or picking time, and inflate it if the date falls anywhere near the export peak.

Add a buffer proportional to how close you are to a stoppage. Small in the quiet months, generous near either fixed event.

What comes out is a personal order-by date for each of your selling peaks. Write it down, put it somewhere visible, and treat it as a deadline rather than a suggestion. Almost every seasonal disappointment in this industry traces back to an order placed two weeks after a date somebody had already worked out and then ignored.

Timing is the obvious effect of the manufacturing year. Quality is the quieter one, and it moves in the same rhythm.

A workshop running at comfortable capacity has time to reject a bad moulding, mix a colour batch properly and let a supervisor look at the first cartons off a run. The same workshop at maximum capacity has orders queued behind orders and every hour of delay pushing something else late. Nothing about the factory's intentions changes; the margin for care simply disappears.

That is why the defects that reach buyers cluster in predictable weeks: substituted components when a supplier runs short, colour variation between batches produced days apart, packaging errors when boxes are assembled at speed, and inspections that happen faster than they should. None of it is malice, and all of it is foreseeable.

There is a second quality effect right after the long shutdown. Workshops rebuild their workforce, and a proportion of the workers on a line in the first weeks are new to that line. Output recovers before consistency does.

The practical response is not to demand more inspections during the peak, which nobody can deliver. It is to shift the orders that matter most into the periods when care is available, and to accept that goods bought at the busiest moment of the year carry slightly more variance — which is another reason a wide, shallow first order in a quiet month is worth more than the same money spent in a rush.

If you remember nothing else from this page, remember this: in Chinese toy manufacturing the cheapest improvement available to a buyer is not a better price or a better supplier, it is an earlier order date. Six weeks of lead time bought in advance costs nothing and buys shorter queues, better attention, more consistent production and freight capacity that still exists. Six weeks of lead time bought under pressure cannot be bought at all.