11.11 Preparation Checklist for Singapore SMEs: The 8-Week Decision Window Starts Now

If you sell anything online in Singapore, your 11.11 outcome is mostly decided by three cut-offs, and the first one falls in under two weeks. Today is 14 September 2026. Double 11 lands on Wednesday 11 November — eight weeks and two days away, or roughly 42 working days once weekends are removed and before you subtract public holidays and staff leave. Working backwards from that date, the purchase order that has to clear a Chinese factory before Golden Week must be placed and paid by around 25 September; your listings, pricing and bundle structure should be frozen by mid-October; and any change to your order, stock or fulfilment systems should be live and tested by 31 October, because nothing gets deployed into a peak. Everything else — the ad spend, the creatives, the countdown posts — is genuinely last-minute work. The supply and systems decisions are not.

How many working days are actually left before 11.11?

Forty-two, give or take. That number matters because SME owners tend to think in weeks and suppliers think in working days. Eight weeks sounds comfortable. Forty-two working days, split across three teams who also have to run the existing business, close September accounts and prepare year-end filings, is not comfortable at all.

Break it into the three things that consume those days unevenly. Stock decisions consume the first ten days and then go quiet while goods move. Systems work consumes the middle four weeks. Marketing consumes the last two. If you sequence them in that order, the schedule fits. If you start with marketing — which is what most businesses do, because it is the visible part — you will discover in late October that the stock you are advertising cannot arrive in time and the system that should route the orders was never tested above normal volume.

Why does China's Golden Week decide your 11.11 stock levels?

China's National Day holiday runs 1–7 October 2026. In practice, the disruption is longer than the seven official days: factories wind down production in the final week of September as workers travel, and lines take several days to return to full output afterwards. Assume a two-week hole in manufacturing capacity, not one.

Now stack the transit time on top. Sea freight from southern China to Singapore typically runs 7–14 days port to port, plus consolidation, customs clearance and inland delivery. Air freight is 2–4 days but costs several times more per kilogram, and air capacity itself tightens in the run-up to peak. If you want goods sitting in your warehouse by 1 November — ten days of buffer before 11.11, which is the minimum that lets you catch a short shipment and still react — then production has to finish by mid-October at the latest. Production that finishes by mid-October has to be ordered and deposited before the factory closes.

That is why 25 September is a real deadline and not a conservative one. Miss it and you are not choosing between early and late; you are choosing between air freight at three to five times the cost, or simply not having the stock. This is the single most expensive mistake in the whole eight weeks, and it is made quietly, by doing nothing, in the third week of September.

What has to be locked before 30 September?

Four things, and all of them are decisions rather than tasks:

• The SKU list you are actually pushing. Not everything you sell — the eight to fifteen items that will carry the campaign. Pull last year's 11.11 and this year's 9.9 data and rank by gross margin contribution, not units sold.
• Reorder quantities and the deposit. Placed and paid, with written confirmation of the production completion date from the supplier. A verbal "no problem" in late September is worth nothing in late October.
• Your floor price per SKU. Decide the lowest price you will accept once platform commission, payment fees, marketplace subsidy participation and shipping are deducted. Write it down. Marketplace campaign managers will ask you to go lower in October, and a number decided in advance is much easier to hold.
• Fulfilment capacity. Whether that is your own pick-and-pack headcount, a 3PL booking, or marketplace-fulfilled stock that must be inbounded by a specific date. 3PL peak slots are allocated first-come; September is when they are available and October is when they are not.

Which parts of the peak should you automate now?

The useful test is not "what is manual?" but "what breaks when volume triples and the person who normally does it is on leave?" Three candidates almost always qualify for Singapore SMEs.

Stock sync across channels. If you sell on more than one marketplace plus your own site and inventory is reconciled by someone exporting spreadsheets, peak is when you oversell. Overselling costs you the order, the cancellation penalty and the listing health score that determines your visibility for the rest of the quarter. Near-real-time stock sync between channels is the highest-return automation available in this window, and it is a two-to-three week project, not a two-day one — which is precisely why it belongs in the middle four weeks and not in November.

Order-to-invoice handover. Marketplace orders that get keyed into your accounting system by hand at 200 a day will be keyed in wrongly. If you are GST-registered and moving toward InvoiceNow, the clean-up work you do here now is the same work e-invoicing readiness will demand in any case, so it counts twice.

Low-stock and exception alerts. Simple threshold alerts on your top SKUs, going to a channel someone actually reads, catch the two or three problems that would otherwise surface only when a customer complains.

What should you leave alone? Anything that touches payment capture, your accounting ledger's chart of accounts, or your core website platform. Changes there have long failure tails and no upside during a peak. Note them for January.

What breaks first when order volume triples?

In our experience with Singapore SMEs, in this order: stock accuracy, customer response time, then delivery exceptions. Stock accuracy fails first because errors compound silently — a mis-sync at 9am produces cancellations all day. Customer response time fails second because enquiry volume rises faster than order volume, roughly two to three enquiries per additional order. Delivery exceptions fail third and hurt longest, because a courier issue on 12 November becomes a marketplace dispute on 20 November, and disputes affect your account standing well into the new year.

Test for these before you need to. Take a Wednesday in late October and run a deliberate load test: process a day's worth of orders at three times normal volume using real data through your actual systems, with your actual staff. Whatever fails is the thing to fix — and you will still have two weeks to fix it.

What does the eight-week schedule look like week by week?

The shape of that table is the whole argument. The first two weeks and the last two weeks are the ones businesses instinctively plan for. The middle four — the systems weeks — are the ones that get skipped, and they are the reason a peak that should have been profitable ends up merely busy.

Frequently asked questions

Is 11.11 still worth preparing for if most of my sales are B2B or local retail?

The campaign itself may not be, but the fortnight around it is. Consumer attention, courier capacity and warehouse labour all tighten from early November through Chinese New Year. Even if you are not running 11.11 promotions, your deliveries will be slower and your suppliers busier. Plan your own November stock and dispatch around the peak rather than through it.

We missed the late-September purchase order window. What are our options?

Three, in descending order of preference. First, check what your supplier already has in finished stock — pre-holiday inventory is often available at short notice and merely needs shipping. Second, reduce the campaign to the SKUs you already hold and promote depth rather than range; a narrower, fully-stocked campaign outperforms a broad one that sells out on day one. Third, air freight a partial quantity on your highest-margin SKU only, and treat the freight premium as a marketing cost. What does not work is advertising stock you have not confirmed.

Can we build the stock sync automation ourselves in four weeks?

Often yes, if you have one capable technical person, your channels expose reliable APIs, and you scope it to one-directional stock updates rather than a full two-way integration. The realistic risk is not the build, it is the edge cases — bundles, pre-orders, reserved stock and partial cancellations — which is where most in-house syncs fail under load. Build the simple version now, test it at volume in week six, and defer the edge cases to January when a mistake costs nothing.