Pre-Shipment Inspection for a Wiper Blade Container: The Checkpoints That Matter

The balance payment is the last real hold you have over an order. On a wiper build quoted at 15 to 20 days — the figure Guangzhou maker Zanyu Automotive states for volume production, and a normal range for the category — you get roughly one working day between the inspection report landing and the factory expecting the wire. If your wiper blade pre shipment inspection checklist and AQL terms were not agreed months earlier, that day is spent arguing about what "acceptable" means rather than deciding whether to ship.

Wipers are awkward to inspect. They are cheap per piece and shipped in large counts, so sample sizes are big. They come in a dozen near-identical lengths, so mixed-SKU errors are common and hard to spot. And the thing that decides whether a customer is happy — how the blade behaves on wet glass — is a functional test a general inspector will not perform unless you write it into the brief.

Where inspection sits in a 15–20 day build

The mechanics matter because booking late is the most common way a buyer loses the check entirely.

A third-party pre-shipment inspection is normally conducted when at least 80% of the order is produced and packed, with the balance at least completed. That is not a courtesy convention; it is what makes the sample statistically meaningful and it is what stops a factory presenting a curated pallet. On a 15–20 day production window, that puts inspection somewhere around day 14 to 18, which means:

• Book 5 to 7 working days ahead. Inspection firms schedule by region and by man-day availability, and the Guangzhou–Dongguan corridor is busy. A booking made on the day production finishes will not get you an inspector on the day you need one.
• Get the packing status in writing before the booking is confirmed. Ask for carton counts by SKU. If the factory says 80% and the inspector arrives to find 45% and a promise, you pay for an aborted visit.
• Allow a day for the report, and a day for your decision. Most firms issue a report within 24 hours of the visit. Build both days into your shipping schedule rather than into your argument.
• Fix the consequence of a fail in the purchase order, not on the day. See below.

A factory that pushes for an early inspection, offers to "finish the last cartons after the inspector leaves", or wants the inspector escorted and shown only designated pallets, is telling you something.

Sampling: how many blades actually get opened

Most inspection reports cite ANSI/ASQ Z1.4 (the sampling standard nearly every inspection firm works to; its international counterpart is numerically different but functionally the same), at General Inspection Level II, single sampling, normal severity. You choose the acceptable quality limits. The conventional set for a consumer part is critical 0, major 2.5, minor 4.0, and unless you say otherwise, that is roughly what you will get.

Here is what those numbers actually mean for a wiper order.

Critical defects are accepted at zero in almost every plan: one found, the lot fails.

Two things buyers routinely get wrong.

The sample is drawn across the lot, not across your SKUs. A 3,000-piece order split across twelve lengths gets a 125-piece sample. Left to chance — or to a factory helpfully stacking the pallets — that sample can come overwhelmingly from three SKUs. Write into the inspection brief that cartons must be drawn from every SKU and every production date code present, with a minimum number of cartons per SKU. On a wiper range, mixed or mislabelled sizes is the defect mode that costs you the most in returns, and it is invisible unless the sampling reaches every size.

AQL is a lot-level decision rule, not a quality target. Passing at 2.5 major means the inspector found few enough defects in the sample to accept the lot, not that 2.5% is fine with your customer. If your channel is a retail chain with a chargeback regime, tighten to 1.5 major and expect the factory to price it.

The wiper-specific checkpoints

This is where a generic inspection brief fails you. An inspector who is not told what a wiper defect looks like will check the carton, the barcode and the drop test, note that the blades look fine, and pass a lot with a strip that chatters.

Defect classification you can hand to an inspector

Write your own table and attach it to the purchase order. A workable starting point:

The classification is a commercial decision as much as a technical one. A barcode that will not scan is a major defect for an Amazon or retail programme and arguably a minor one for a workshop distributor. Decide it in advance, in writing, for your channel.

The functional tests to specify by name

• Wipe test on wet glass. Specify a glass panel, water applied, the blade mounted on a representative arm, a stated number of strokes, and observation for skip, streak, chatter and noise. State how many pieces per SKU get this test — it is slow, so a subset of the sample is normal, but the subset should cover every SKU. Without this line in the brief, it does not happen.
• Adapter fit and retention. The inspector needs physical arms or gauges for each connector type in your range — U-hook, push-button, bayonet, pinch-tab and whatever else your multi-fit kit claims to cover. Send them, or require the factory to supply a verified gauge set that the inspector checks against your samples. A retention or pull-off force in newtons is the right way to specify this; the value is not something to guess, so agree it with the factory at sample stage and have the inspector measure against the agreed figure.
• Length verification against the pack. Measure the blade, compare to the printed size, on every SKU in the sample. Agree a tolerance in millimetres in the purchase order; do not accept "as per sample" as the tolerance.
• Free arch and pressure distribution. Press a beam blade flat against a straight edge and look at where it lifts. A rail with an uneven arch will lift in the middle or at one end and will streak there in service. It is a two-second check that most inspectors will do if you describe it.
• Strip hardness. A durometer reading in Shore A against your agreed range. Zanyu Automotive lists a strip hardness tester, a strip microscope and a combined pressure-and-wear rig on its equipment roster for outgoing checks — useful, and worth asking for the readings as a record. A factory's own bench data is a supplement to a booked inspection, not a substitute, and the difference between the two is covered in what a wiper test bench actually measures.
• Carton drop test and cube check. One carton dropped on its faces, edges and a corner from a height matched to its weight; then carton dimensions and gross weight against the packing list. The cube number is what your freight quote was built on.
• Barcode scan on every SKU with a real scanner, plus a check that the FNSKU or retail label is the right one for the right size. On a twelve-size range this is the highest-value ten minutes of the visit.

What a pre-shipment inspection cannot tell you

Be clear-eyed about the limits, because buyers over-trust the report.

Durability claims are not testable on the factory floor. A supplier-stated figure — Zanyu, for example, describes a modified fluorosilicone strip with a graphite coating rated for −50°C to 80°C and two to three times the service life of ordinary rubber, which the company presents as its own figure — belongs in laboratory testing at sample stage, not in a pre-shipment brief. So does salt-spray corrosion on the rail and clips, ozone and UV ageing on the strip, and any low-temperature flexibility claim. Sort those out during the seven-day sample evaluation, long before a container exists. Pre-shipment inspection tells you whether this lot matches the sample you approved and the specification you wrote. That is all it is for, and it is enough — as long as the approved sample and the specification are good.

Booking, cost, and what a "fail" actually buys you

Third-party inspection is sold by the man-day, commonly quoted in the region of US$250–350 per man-day for mainland China, plus travel; confirm at booking, and expect a large mixed-SKU wiper order to need more than one day or more than one inspector. Bureau Veritas, Intertek and QIMA are the widely used firms, with a long tail of competent independents. The general mechanics of booking, scope and reporting are covered in the pre-shipment inspection guide.

The clauses that make the inspection worth paying for are in the purchase order, not the inspection contract:

1. Buyer's right to nominate the inspector, with unescorted access to the finished goods and to the packing area. Whether a given factory accepts a buyer-nominated inspector is often not published — ask before the PO, and put the answer in it.
2. Balance payment conditional on a passed report. This is the whole point. Sequence it explicitly: report issued, buyer accepts, balance released, cargo ready. Payment structures that support this are set out in the T/T and L/C payment terms guide.
3. Defined consequence on a fail. A failed lot does not automatically get you anything. Write the remedy: 100% sort and rework at the supplier's cost, followed by re-inspection at the supplier's cost, with shipment held until a pass. Without the cost allocation, a re-inspection becomes a negotiation.
4. A hold on shipment, not just a right to complain. If the goods sail before you decide, you have bought a claim, not a remedy.
5. Named acceptance criteria — your AQL levels, your defect table, your functional tests — attached as an annex, referenced in the PO body.

If you are buying through a platform rather than direct, settle one more thing before production starts: who books the inspector, who receives the report first, and who has authority to hold the shipment. ChinaMakersHub states that it will help confirm export documentation, inspection and shipping arrangements for an order, which is useful — but those three answers still need to be written down with your name against at least one of them.

A failed report is a bargaining position, and it decays fast once the booking is made and the vessel is waiting. Use it in the first 24 hours or not at all.

Common questions

Do I need a third-party inspection if the factory has its own test bench?

Yes, and the two do different jobs. In-house equipment — the hardness tester, microscope and pressure-and-wear rig on Zanyu's roster, for instance — supports the factory's own outgoing checks and gives you process data worth asking for. A booked third-party inspection gives you an independent, timed, sampled verdict on the lot you are about to pay for, taken by someone with no interest in the answer. Ask for the internal records and book the inspection. Broader context on what factory-side quality control does and does not cover is in the China manufacturing quality control guide.

Should I inspect during production instead of before shipment?

On a first order with a new supplier, do both if the value justifies it. A during-production inspection at roughly 20–40% completion catches a systemic problem — the wrong strip compound, a mis-set rail bend, the wrong adapter in every kit — while there is still time to correct it inside the 15–20 day window. A pre-shipment inspection alone catches the same problem when the only remedies left are rework or delay. On repeat orders from a stable supplier, pre-shipment alone is usually proportionate.

What if my order is under the MOQ for a meaningful sample?

The sample plan scales down, not away. A 1,000-piece order — the stated minimum at several wiper makers, Zanyu included — falls in code letter J at 80 pieces sampled, which is a real check. What gets thin at that size is per-SKU coverage: 80 pieces across twelve lengths is under seven per size. On small first orders, consider a tightened plan or a specified minimum pieces-per-SKU, and accept that you are paying a higher inspection cost per unit shipped. It is still cheaper than a container of mislabelled sizes.

What to ask the supplier next

Send this before the purchase order, not with it:

1. Do you accept a buyer-nominated third-party inspector, with unescorted access to finished goods and the packing area?
2. What AQL levels and defect classification do you work to as standard — and will you accept mine as an annex to the PO?
3. What length tolerance in millimetres do you hold on finished blades?
4. What adapter retention or pull-off force do you build to, and how is it measured?
5. Will you supply a verified arm or gauge set for every connector type in the range, for the inspector's use?
6. What outgoing checks do you run in house, on what sample rate, and will you release the records with the shipment?
7. What are the carton counts by SKU, inner and master quantities, and the packing plan the inspector should check against?
8. Do you accept a during-production inspection at 20–40% on the first order?
9. If a lot fails, who pays for the sort, the rework and the re-inspection?
10. How many days before cargo-ready will 80% of the order be packed and available for inspection?

Zanyu Automotive is a reasonable example to run that list against. The company states a 15–20 day production lead, 3–7 day sampling and a 1,000-piece MOQ, and lists a strip hardness tester, microscope and pressure-and-wear rig on its equipment roster for outgoing checks — all supplier-stated, and all worth confirming at enquiry. Its own AQL levels, defect classification and position on buyer-nominated inspectors are not published, which is exactly why questions one, two and nine belong in the first exchange rather than the last. Note too that the pack is part of the inspection scope, so the artwork, barcode and label decisions described in the piece on private-label wiper packaging formats have to be settled before the inspection brief is written, not after. Registry-checked background, and the points still flagged for verification including where production physically takes place, are set out on the Zanyu Automotive factory page; on a first order, pairing a live video walkthrough with a booked inspection is the sensible pairing, and the walkthrough agenda in the China factory audit checklist covers what to look at.