Pre-Shipment Inspection for Bags: When to Book It and What It Covers
Book it against packing progress, not against a date — normally when the order is at least 80% packed and fully finished. It runs after the bulk window, before loading, and covers appearance, dimensions, materials, function and packing. It does not cover anything the designed duty test has to establish.

The pre-shipment inspection sits after the bulk window and before loading. Trigger it at 80% packed, give the inspector the tech pack in advance, and expect the report within one working day of the visit. Everything else is detail.
Two mistakes account for most unhappy inspection days, and both are booking mistakes rather than quality problems. The first is booking against a calendar date agreed weeks earlier, which usually means either inspecting part of an order or cancelling and starting again. The second is booking too close to loading, so that a failed lot has no schedule left in which to be fixed.
What Does a Pre-Shipment Inspection Actually Cover?
Six areas, in roughly the order an inspector works through them. Appearance and workmanship against the counter-sample. Dimensions against the tech pack. Materials — the fabric, its stated construction, the lining, the hardware. Function of every closure, strap and moving part. Packing, including carton count, weights and marks. And finally the marking and labelling, which is where most compliance failures sit.
What holds it together is the reference set the inspector is given. A verdict on appearance without an approved counter-sample is an opinion; a dimension without a stated reference point is a guess. The two documents that turn an inspection into a measurement are the signed-off sample and the tech pack, and both must be present in the room.
| Area | What is checked | Reference it is read against |
|---|---|---|
| Appearance and workmanship | Shell, panels, prints, seams, thread ends, cleanliness | The approved counter-sample and the defect classification |
| Dimensions | Body, handle drop, strap length, volume-critical internal dimensions | The tech pack drawing, stated flat and measured to a named reference |
| Materials | Shell fabric construction, lining, foam or padding, hardware finish | The specification sheet, sometimes confirmed by weight or laboratory test |
| Function | Every closure, buckle, slider and strap adjustment cycled | The duty stated in the specification and the cycle count agreed |
| Packing | Carton count, units per carton, weights, marks, mixed cartons | The packing instruction and the shipping documents |
| Marking and labelling | Care labels, composition, country of origin, tracking labels | The destination-market requirement named on the order |
When in the Programme Should It Be Booked?
After bulk production and before loading. Sampling runs 6-10 working days from a signed-off tech pack and bulk runs 35-50 days from written sample approval; inspection is the event that closes the second window. Booking it inside the bulk window assumes everything passes first time, which is the assumption that turns a quality finding into a missed shipping date.
The prudent approach builds two dates at quotation stage: the expected inspection date, and the fallback date that applies if the first visit fails. The gap between them is the re-work cycle plus a second inspection, and it should be visible on the programme calendar even if nobody expects to use it. A plan with no room for failure is a plan that will be renegotiated at the worst possible moment.
How Much Packing Progress Do You Need Before It Can Run?
The usual threshold is 80% packed, with the remainder expected within the week. That threshold exists because inspectors must draw a random sample from across the whole lot: with only a third of the cartons available, the sample taken represents the cartons that exist rather than the order. The report should always state how many units were finished and packed at the time of the visit.
Below about 80% the honest options are limited. You can accept a conditional verdict covering what exists, understanding that the remainder is unchecked; you can wait; or you can ask for the incomplete cartons to be held out of the lot and inspected separately later. The third is often the best commercial answer and it costs one extra visit.
It also helps to know what the 80% figure is protecting. A sampling plan assumes the units drawn represent the units that will ship. When half the cartons do not exist yet, the sample can only speak for the half that does, and the parts made last — usually under the most schedule pressure — are exactly the ones missing. The threshold is not bureaucracy; it is the condition under which a random draw means anything.
| Point in the programme | What is happening | Inspection action |
|---|---|---|
| Sample approval | Written approval of the reference sample | Confirm the counter-sample is retained and labelled for later use |
| Start of bulk window | Materials in, first units off the line | Book against the expected packing date rather than against today |
| Around 70% packed | Most cartons exist, forecast firming up | Confirm the date and send the four reference documents |
| 80% to 100% packed | Lot presentable for random selection | Inspection runs; report within one working day of the visit |
| Report issued | Verdict known, decision pending | Release, re-work, sort or waiver — whichever the order specified |
| Before loading | Container booked | Confirm the loaded quantities match the report and the packing list |
What Information Does the Inspector Need Before Arriving?
Four documents, sent at least two working days ahead: the tech pack with dimensions and tolerances, the defect classification, the packing instruction, and the approved reference — either the counter-sample itself or photographs of it if it cannot travel. Without those, the inspector is working from general practice rather than from your specification, which is precisely what the visit is supposed to prevent. A fuller document set also shortens disputes: when a finding is challenged six weeks later, the answer is usually one line in one of these four documents.
Add two administrative items that save more time than they deserve. Confirm who attends and who can authorise anything on the day, and confirm the address and access arrangements for the goods. Visits have been lost to a container already being loaded, and to nobody on site having authority to open cartons.
What Happens on the Day?
The inspector arrives, establishes the lot from the packing list, selects cartons at random, draws units, and works through checklist area by area: appearance first, then measurements, then function, then packing. Each finding is recorded with a location and a photograph. Where a laboratory property is involved — fabric weight, coating adhesion, colour fastness — samples are drawn for a separate test rather than judged on the bench.
Multi-colourway lots deserve their own booking note: they take longer than buyers expect. Each colourway needs its own appearance check against its own reference, its own shade-band comparison, and often its own hardware check, while the sample is still shared across the whole draw. A lot in four colourways can take most of two days for one inspector.
The order of work also matters more than it appears. Appearance comes first because units get handled afterwards and handling changes appearance. Measurements follow, taken flat on a bench rather than on a knee. Function runs last on the units that passed the earlier stages, because a closure that has been cycled should not later be judged visually against the others.
Expect the inspector to stop early in one specific circumstance: when the count reaches the reject number. Most agencies stop there and write the report, since further counting changes nothing about the verdict. Some will continue for a fuller picture for an agreed fee, and asking for that is often the right call after a failure, because the fuller list is what the re-work conversation needs.
How Long Should There Be Between Inspection and Loading?
At least enough time to act, which in practice means two or three working days between the report and the container. That window covers the only outcomes worth planning for: re-work and resubmission, a sort and partial release, or a waiver decision taken in writing by somebody senior enough to own it. None of those can be arranged in the two hours before a truck arrives.
The result is a release decision made on evidence rather than on hope. Where loading is booked for the day after the inspection, the inspection has already lost most of its commercial value: it becomes paperwork rather than a release decision, and everybody in the room understands that.
What Does It Not Cover?
Three things regularly get assumed into the scope and then discovered missing when needed. It does not establish durability: cycling a zipper fifty times during a visit approximates the duty rather than proving it. It does not establish any measured performance property — insulation performance, colour fastness, coating adhesion — those belong to laboratory tests with their own standards. And it does not certify compliance with destination rules beyond what is visible on the unit.
It also does not tell you what the remaining units are like, beyond the estimate the sample supports. A passed lot is evidence that the sample complied, nothing stronger. Buyers who read a passed report as a warranty about every carton set themselves up for a disagreement later, usually at the arrival port where leverage is worth least.
One question always follows and deserves an answer before it is asked: is the pre-shipment inspection worth running at all on a small or low-risk lot? Yes, and the reason has little to do with the sample. Its real value is that somebody independent opens the cartons and writes down what they found before payment is released — an event that changes behaviour on the line long before it changes any verdict.
That behavioural effect is strongest when everybody knows the inspection is booked and the report will be read. Programs where reports are filed and never opened tend to drift, and the drift shows up first in exactly the place the standing of the plan is weakest: in the minor column, in the packing counts, and in whatever the classification did not anticipate.
What Are the Common Reasons a Report Cannot Be Issued?
Six recurring ones, all avoidable with a day of preparation. The goods are not finished. The counter-sample is missing. No packing list exists, so units cannot be selected at random. Nobody with authority is present. The specification conflicts with what was actually made, which is a real finding but needs a decision before it becomes a verdict. Or the lot was split and nobody knows which units belong to which shipment.
Every one of those costs a return visit and a week, and every one is visible twenty-four hours before the visit if somebody checks. A short confirmation call the day before — finished quantity, samples present, packing list ready, authority on site — prevents almost all of them.
What Happens When an Order Has Several Shipping Dates?
Each shipment is its own lot and needs its own inspection, unless the whole order is presented at once. That is uncomfortable when the first shipment is 500 pieces of a 5,000-piece order, because it costs five visits rather than one. There is no shortcut available that preserves the plan's meaning: inspecting the first shipment and releasing the remainder without inspection is a commercial decision to self-insure, not a sampling decision.
Where repeat shipments genuinely carry low risk — a long clean history, an unchanged specification, steady output — the switching rules offer the legitimate route. Reduced inspection draws roughly 40% of the normal sample once a sustained run of accepted lots has been recorded, which lowers cost while keeping some evidence on every lot.
There is one arrangement worth naming because it is frequently proposed and rarely understood: inspecting the first shipment and treating it as covering the rest. That is not sampling. It converts four lots into self-insurance, and while it can be a reasonable commercial decision for a trusted programme, it should be taken deliberately and recorded as a waiver rather than described as an inspection.
The better structure for genuinely repetitive programmes is a step-down plan: full pre-shipment inspection on the first two lots, reduced inspection once the history supports it, and a return to normal immediately after any rejection. Written down, that gives everybody a rule rather than a negotiation before every shipment.
What Should Be Written Into the Booking Instruction?
A single page, sent with the order and repeated at booking. It removes every decision that otherwise gets made in a hurry on the day.
Ten lines for the booking instruction
- The lot identity: purchase order, style, colourway and the units this inspection covers.
- The standard and edition, the inspection level, and single or double sampling.
- The AQL per defect class, with the classification annexed.
- The reference set present in the room: counter-sample, tech pack, packing instruction.
- Who attends and who can authorise a decision on site.
- The trigger: how much of the order must be finished and packed.
- The report requirement: what it contains, and when it arrives.
- The consequence of a failed lot, agreed before the visit rather than after it.
- What happens to units opened or destroyed during testing.
- The confirmation call: who checks readiness the day before.
Agreeing all of that once means later visits need only a date. The two reference points worth reading alongside are the dispute route for when a verdict is contested, and the packaging inspection detail for what the packing section should contain.
Book the inspection against progress, not a date
Send the order quantity, duty and shipping schedule. We coordinate production and inspection around the same calendar.


Frequently Asked Questions
When should I book a pre-shipment inspection?
Against packing progress, usually at 80% packed with the rest expected within the week. Booking against a fixed date risks inspecting part of the order or making a wasted visit.
Can it run while production is still going?
Partly. An inline or mid-production check can run during production, but the pre-shipment inspection needs finished and packed goods because the sample is drawn from cartons across the whole lot.
What minimum packing level is needed?
Around 80% is the practical threshold. Below that the sample represents only the cartons that exist, so the report should state the quantity finished and packed at the time of the visit.
How long before loading should it be scheduled?
Leave two or three working days between the report and the container. That is enough time to arrange re-work, a sort and partial release, or a written waiver if any of them become necessary.
What documents does the inspector need in advance?
The tech pack with dimensions and tolerances, the defect classification, the packing instruction, and either the counter-sample or photographs of it. Send them at least two working days ahead.
How long does the visit take?
Most of a working day for one inspector on a standard general-purpose bag at the usual sample sizes. Several colourways in one lot takes considerably longer, because each needs its own reference check.
Does it cover durability or performance?
No. Function checking during a visit approximates duty rather than proving it, and measured properties such as insulation performance or colour fastness belong to laboratory tests with their own standards.
What happens if the goods are not ready?
The visit is rescheduled and usually charged. A readiness check the day before — finished quantity, sample present, packing list in hand — prevents almost every instance.
Can several shipping dates share one inspection?
No. Each shipment is its own lot and each needs its own inspection. With a long clean history the programme may qualify for reduced inspection, which draws roughly 40% of the normal sample.
Who decides when a lot fails?
The pre-agreed consequence in the purchase order decides: re-work and resubmission, a signed waiver, or rejection. Deciding after the failure turns a quality finding into a schedule argument.
What if nobody with authority is on site?
Nothing can be agreed and the value of the visit collapses. Confirm in the booking instruction who attends and who can authorise decisions.
Is a passed inspection a guarantee about every unit?
It is evidence the sample complied. The remaining units carry the risk implied by the sampling plan, which is why repeat offenders should be discussed as a process problem rather than a lot problem.
People Also Ask
What is a pre-shipment inspection?
A check carried out on finished and packed goods before loading, covering appearance, workmanship, dimensions, materials, function, packing and marking, with a verdict taken against a sampling plan.
How long before shipping should inspection be done?
Far enough ahead to act on the result — in practice two or three working days between the report and the container, triggered by the order being around 80% packed.
What is the difference between inline and pre-shipment inspection?
Inline runs during production on partly finished goods to catch faults early; pre-shipment runs after packing to decide whether the lot releases. They answer different questions and are not substitutes.
Can inspection be waived for repeat orders?
Not entirely, though programmes with a clean history can move to reduced inspection at roughly 40% of the normal sample. Some evidence on every lot is still required.
What happens if the inspection fails?
The pre-agreed route applies: re-work and resubmission, a sort and partial release, or a written waiver signed by the buyer, each with its own cost and schedule effect.
Sources and further reading
Every figure on this page is either our own programme data or attributed below. Open the originals before you write them into a specification.
- ISO 2859-1:1999 Sampling procedures for inspection by attributesInternational Organization for Standardization
- ANSI/ASQ Z1.4 — Sampling procedures and tables for inspection by attributesAmerican Society for Quality
- ISTA test procedures — packaged-product performance testingInternational Safe Transit Association