How Do You Estimate the Shipment ETD of a 500-Piece Bag Order?

A delivery date is not a number a vendor gives a buyer. It is a number the buyer can build, and then check the vendor answer against. This piece is the seven-step sum, written so that it can be done on one page with a calendar, using a real 500-piece order as the worked case. Where a step has a range rather than a figure, the range is stated and the reason for it is given.

What Are the Seven Steps in the Sum?

1. Sampling — 6-10 working days from a signed-off tech pack, 12-15 for a complex build.
2. Bulk production — 35-50 days from written sample approval plus receipt of the deposit.
3. Inspection — 3-5 days, including the re-work and re-inspection branch.
4. Booking — 3-4 days to a container yard cut-off, longer in a peak.
5. Holiday overlay — any public holiday at the production location inside the run.
6. Change buffer — the days a realistic change request consumes, if any is expected.
7. Transit — outside the FOB promise, but inside the date the buyer actually cares about.

The last step is the one that separates an ETD from a delivery date the buyer can plan around. ETD is the vessel departure. Under FOB Xiamen the promise ends at the ship rail, so transit, customs entry at the destination and inland delivery are the buyer legs. Add them and the number becomes a warehouse arrival rather than a ship departure.

How Do You Put Real Numbers Against Each Step?

Read the two columns and the point of publishing a range becomes clear: the difference between the short and the long case is around thirty-five days, and almost all of it is decided by choices the buyer makes in the first week. Fabric, hardware, holiday timing and whether the specification is actually finished are four decisions, and they are worth more than any negotiation conducted later.

How Does the Worked Example Read?

A 500-piece insulated cooler style, dyed shell, stock zipper, welded liner, retail-ready packing. Sampling 8 working days. Bulk 42 days, because the dyed shell makes fabric the critical thread. Inspection 5 days with the re-check branch reserved. Booking 4 days. No holiday inside the run. No change expected. Transit is the buyer leg and is not counted in the ETD.

The sum: 8 working days of sampling, which is about 12 calendar days, then 42 days of bulk, then 5 days of inspection, then 4 days of booking. Sample approval on 1 March gives an ETD of roughly 28 April. That is the number to plan around, and every part of it can be checked against a document as the run proceeds.

How Does the Sum Read on Three Different Programmes?

Three worked cases make the range concrete, and they differ by far more than quantity. A drawstring style on a stock fabric with stock hardware lands at the short end because three of the seven rows collapse. An insulated cooler style with a dyed shell and a welded liner lands in the middle. A protective equipment case with a tooled fitting and a custom foam insert lands at the long end because two dependencies are front-loaded.

The lesson in that table is the one worth carrying into every estimate: quantity is the variable buyers watch and the one that matters least. All three are 500-piece orders and they are thirty-three days apart, and every one of those days is decided by a specification choice rather than by a size. A buyer who wants a shorter number should edit the specification row, not argue about the quantity row.

How Do Holidays Enter the Sum?

As an overlay rather than as a step, because a holiday does not add work — it removes working days from the middle of the run. Count the closure, add the ramp-down and the ramp-up around it where the shutdown is a long one, and add a week either side for the port congestion if the run matures near the closure.

The official public holiday arrangement is published annually by the General Office of the State Council and should be read from that publication rather than from anyone recollection. Where the run crosses a long shutdown, the overlay is four to six weeks rather than the length of the holiday, and pretending otherwise is the single most common reason a Q1 estimate is wrong.

One more caution on the worked example, because it is the mistake the sum invites. The 42 days in the example is a bulk window, not a total, and it does not start until the sample is approved in writing. A buyer who marks 1 March as the order date rather than the approval date has already lost the sampling window before the arithmetic starts, and the resulting ETD is wrong by the length of it.

Why Does the Inspection Step Carry a Branch?

Because a first submission that fails is re-worked and re-inspected with a fresh sample drawn under the same code letter. On a 500-piece lot at General Inspection Level II under ISO 2859-1 the draw is code letter H, 50 units, accept 3 and reject 4 at AQL 2.5 for major defects. If those 50 produce four majors the lot is rejected, and the re-inspection is a second full inspection rather than a glance at repaired units.

Reserving three days assumes the first submission passes. Reserving five assumes it might not, and the two extra days are the least expensive insurance in the estimate. The what happens when an order fails AQL post sets out what actually happens in the rejected case, and the calendar it consumes is the reason for the branch.

What Change Buffer Is Realistic?

Zero if the specification is genuinely frozen, and most first runs are not. The realistic figure for a first programme with a brand-owned shade is five to seven days: one round of dips, one artwork correction, or one hardware decision that was left open at approval. Naming the buffer in the estimate is not pessimism; it is the difference between an estimate and a hope.

A useful way to size the buffer is to count the decisions still open at the point the estimate is written. Every unanswered question about a dimension, a hardware reference, a print position, a carton mark or a barcode is a change waiting to happen, and each one costs a round of working days in whichever thread it touches. Five open questions is a five to seven day buffer, and closing them before the order is a free reduction.

The buffer is also the only row in the sum the buyer fully controls. A tech pack that is complete, a dip answered within a day, and a hardware reference decided before sampling rather than after it remove the row entirely. That is the practical answer to a buyer who wants a shorter estimate and does not want to change the specification.

The recomputation is also the moment to ask the only question that matters once a run is underway: which thread is now critical. At the start of a dyed programme it is fabric; after the cloth is in the store it becomes the line slot; at the end it becomes the inspection branch. A buyer who knows which thread is critical at any given moment knows where pressure will help and where it will only cost goodwill.

How Is the Estimate Checked as the Run Proceeds?

By re-running it with actuals in place of plans. At each milestone, replace the planned duration with the actual one and recompute the ETD. This takes five minutes and it is the only way to know, in week four rather than week seven, that the date has moved. A buyer who does this after each milestone is never surprised; a buyer who does not is surprised once.

The recomputation also tells you where to spend effort. If sampling overran by four days and everything else is on plan, the recovery has to come from the rows with float — booking, and possibly inspection if the first submission is clean. If the bulk window has overrun, there is no recovery available and the conversation is about the specification rather than the schedule.

There is one more row that belongs in the sum and is almost always left out: the buyer own approval cycle. A dip that sits unanswered for four days, a sample approval that waits for a monthly meeting, and an artwork correction that needs three internal signatures are days on the critical path, and they are days the vendor cannot see or recover. Add them explicitly and the estimate stops blaming the wrong party.

A practical way to handle that row is to set a response standard before the order is placed: one working day for a dip, two for a sample approval, and a named deputy who can answer when the named person is away. Those three commitments are worth more to the calendar than any concession a buyer can negotiate out of the production side, and they cost nothing.

What Should the Estimate Be Written On?

Order

Style, quantity 500 pcs, colourway count

Trigger

Written sample approval plus deposit, dated

Sampling

6-10 working days, or 12-15 for a complex build

Bulk

35-50 days, with the critical thread named

Inspection

3-5 days including the re-check branch

Booking

3-4 days, more in a peak season

Overlays

Holiday closure, ramp-down and ramp-up where they apply

Result

ETD as a date, plus the buyer-side transit leg

One further caution on the worksheet is worth stating, because it is where estimates quietly go wrong. An ETD is a departure date, not an arrival, and the buyer side of the calendar is longer than most plans allow. Transit, customs entry at the destination, any examination hold, and inland delivery to the warehouse or the fulfilment centre are four legs, and the second and third of them are outside anyone control.

The honest way to present that is as a separate line on the same page rather than as a footnote. One date for on-board, one for arrival at port, one for delivery. Buyers who plan against three dates are rarely surprised; buyers who plan against one and treat the other two as administrative are surprised regularly, and the surprise always lands in the same week.

Eight lines and the estimate is a document rather than a conversation. We do not promise the fastest 20 days. We promise a delivery date you can calculate. The worksheet is the calculation, and the re-run at each milestone is what keeps it true.

Quantity deserves one closing word, because it is the variable buyers negotiate hardest and the one that matters least. Five of the ten stages are quantity-independent, so doubling an order moves the calendar by days rather than by a multiple. The variables that move it by weeks are the fabric decision, the hardware decision, the holiday timing and the number of decisions left open at approval.

Frequently Asked Questions

People Also Ask