Pre-Shipment Inspection for Wood Flooring: A Usable Checklist

A wood flooring pre-shipment inspection in China fails for boring reasons far more often than dramatic ones. Not counterfeit species, not a swapped finish — a carton count that is short by eleven, a moisture reading taken on the surface instead of the core, or a container with a patched roof that nobody looked at. The generic inspection checklist an agency runs on housewares does not catch any of those, because wood is a hygroscopic product that keeps changing after the report is signed. What follows is a checklist you can hand to SGS, Bureau Veritas, Intertek, TÜV or your own inspector, written specifically for parquet, solid plank and engineered flooring lots, plus the contract wording that makes the report mean something when you release the balance payment.

Set the terms before the inspector books a date

Three decisions have to be made in the purchase order, not in an email the week before shipment.

When the inspection happens. The industry convention is that goods are 100% produced and at least 80% packed before an inspector attends. For flooring, push for 100% packed, for one specific reason: cartons that are still open are cartons that can be re-packed after the inspector leaves. If the mill wants the inspector earlier to protect the sailing date, the compromise is 100% produced and 100% packed with a defined percentage of cartons opened for inspection and re-sealed with tamper-evident tape in the inspector's presence.

What sampling standard applies. Name it: ISO 2859-1, single sampling, general inspection level II, with acceptance numbers stated for critical, major and minor defects. Without a named standard, "we inspected 50 boards" is not a result — it is an anecdote. Whether a given mill will contract to a specific AQL is worth confirming before you assume it; ask, and get the answer in writing.

Who pays for the re-inspection. Fail-then-rework-then-re-inspect is a normal outcome, not a catastrophe. What ruins the relationship is arguing about the re-inspection fee at the moment the sailing is at risk. Settle it in the PO.

What the report gates. The natural gate for a T/T balance transfer is a passed inspection report, and the wording matters: "balance payable against a passed final random inspection report issued by [agency], goods 100% produced and packed" is contractable. "Subject to satisfactory inspection" is not. The mechanics of tying money to milestones are covered in the payment terms guide.

Section 1 — Moisture content across the lot, not one carton

This is the highest-consequence measurement in the whole inspection and the one most often done badly.

• Meter type and settings. A pin-type meter reading at core depth is preferable to a pinless meter for packed flooring, because pinless meters read a shallow zone and are sensitive to board density and backing. Whichever is used, the meter must be set to the correct species density / specific gravity correction — a meter left on the oak default will misread teak, walnut and Sonokeling. Record the species setting on the report.
• Depth, not surface. For solid stock, insulated pins driven to roughly mid-thickness give the core reading. A surface reading on an oiled board tells you about the last two days of warehouse air, not about the board.
• Spread of samples. Take readings from cartons drawn from different pallets and different production dates, not from one accessible stack near the door. Minimum practice: at least three boards from each of at least ten cartons spread across the lot, with carton IDs recorded.
• Ambient conditions on the report. Temperature and relative humidity in the warehouse at the time of measurement must be written down. A lot reading 8% in a humid Guangdong warehouse in August is a different fact from a lot reading 8% in a conditioned room, and your destination climate decides which one matters. The spec side of this is in moisture content and acclimatisation.
• Range, not average. Report the minimum, maximum and spread — not the mean. An average of 9% made of boards at 6% and 12% is a floor that will move against itself after installation.
• Engineered boards read differently. On three-layer and multi-layer construction, take the reading on the wear layer and note that adhesive lines interfere with pin readings; the spread across the lot still matters more than any single figure.

Section 2 — Dry lay: the test that finds what calipers miss

Dimensional measurement with calipers is necessary and insufficient. A dry lay is the only way to find fit problems that live in the interaction between boards.

Have the inspector lay 15–20 boards, or four to six full panels, on a flat surface — a levelled workbench or a sheet of MDF, not a warehouse floor — drawn from at least five different cartons, and check:

• Tongue-and-groove seating. The tongue should seat under hand pressure or light taps with a block. If a mallet is required, the machining is tight or the profile is off. If the joint closes with no resistance at all, it is loose and will open.
• Face gap at the joint. Measure with feeler gauges at the seam, at several points along the board length, not just at the middle.
• Lipping (height offset) between adjacent boards. Run a fingernail and a straightedge across the joint. On a hand-sanded, oil-finished floor, lipping shows more than it does under a thick lacquer film.
• End-match fit on end-jointed boards, checked separately from the long edges.
• Bow, crook and cup on individual boards: straightedge along the length and across the width, gap measured.
• Panel squareness. For assembled parquet panels, measure both diagonals — they should agree — and check each corner with a framing square. A panel out of square by a millimetre multiplies across a room.
• Panel-to-panel module consistency. Lay four panels in a 2×2 block and look at the cross joint where all four meet. That intersection is where cumulative module error becomes visible.
• Back face and substrate. Turn boards over. On engineered and laminated substrate boards, look for delamination at the edges, voids, core gaps and glue starvation along the perimeter.

Section 3 — Finish, cure and appearance

Hand-applied multi-coat oil finishes are the highest-variance element in a handmade flooring lot, which makes this the section where an inspector earns their fee.

• Cure state. Oil finishes cure by oxidation over days, not by passing under a lamp. Check for tackiness with a firm thumb press held for a few seconds — a print that remains is undercured. A clean white cotton cloth rubbed firmly on the face should not pick up colour or oil residue.
• Blocking and print-through. Open cartons and look at where board faces contact each other or the interleaving. Face-to-face marking is the classic symptom of goods packed before cure.
• Coverage and sheen uniformity. Look across the board at a low angle under raking light, then compare boards from different cartons side by side. Hand-rubbed oil penetrates the grain rather than filming over it, so uneven application shows as patchy depth of colour rather than as a visible film edge.
• Appearance against the approved sample set. Compare against the signed sample range, not a single reference board — the whole point of a range is that natural colour variation is expected between the two bounds and rejected outside them. Grading definitions belong in the contract; see writing a grading spec.
• Machining and carving quality on relief and inlay work. Check for tear-out at grain reversal, burn marks from dull tooling, sanding scratches across the grain, and — on inlay and marquetry — gaps at the species boundaries and filler used to hide them.
• Grain-matched components inspected together. This is the detail almost every generic checklist misses. Where stair treads, risers, nosings, skirting and mouldings are supposed to come from the same timber lot as the floor, the inspector must lay a trim piece and a stair part beside a floor board from the same lot and judge them together under the same light. Inspecting the trim carton in isolation proves nothing about matching, which is the entire reason you bought them from the same mill.

Section 4 — Carton count, marking and packing

The single most common written-up discrepancy in flooring shipments is arithmetic.

• Full carton count against the packing list, by SKU and by pallet — physically counted, not accepted from a tally sheet.
• Boards per carton and m² per carton verified on opened cartons, and multiplied out to confirm total lot area against the invoice. Short-count by one board per carton across a container is a large number that nobody notices.
• Carton markings carry, at minimum: species, structure, grade, nominal dimensions, m² per carton, lot or batch number, production date, and the buyer's PO and shipping marks. Confirm they match the packing list character for character.
• Lot/batch numbers recorded on the report so the same-lot claim for trim and stair components can be checked at destination.
• Carton construction — board grade, edge protection at panel corners, interleaving between finished faces, and whether cartons are strapped or shrink-wrapped to the pallet.
• Pallet condition and ISPM 15 mark on wood packaging, legible and correctly formatted. A missing or illegible mark is a destination-clearance problem, not a factory problem, until it delays your container.
• Photographs of markings, of an opened representative carton, and of the palletised stacks.

Section 5 — The container is part of the inspection

Loading supervision is frequently sold as a separate service. For wood flooring, buy it. A perfect lot loaded into a wet container arrives as a claim.

• Close the doors and look for daylight from inside — pinholes in the roof and side panels are how rainwater enters.
• Roof condition checked from outside: dents, patches, standing-water marks.
• Floor condition and previous cargo residue or odour. Chemical or fish odour transfers into oiled wood and is unrecoverable.
• Door gaskets intact; CSC safety approval plate present and in date.
• Desiccant. Confirm the type (calcium chloride hanging bags absorb far more than silica gel by weight and are the norm for long ocean transits) and the quantity used, and that units are hung or placed so they cannot fall against the cargo. Record quantity and placement on the report.
• Container liner or floor protection where specified, and dunnage or airbags to stop the stacks shifting.
• Loading pattern photographed at start, mid-load and doors-closed, with the seal number visible in the final frame.
• Seal number recorded and cross-checked against the bill of lading.

Defect classification: agree it before, not after

The classification is what makes an AQL number meaningful. "AQL 2.5" with no defect definitions transfers the whole argument to the moment of failure. Agree the table first, and attach it to the purchase order alongside the tolerance table. The general framing for inspection stages and what each catches is set out in product inspection before shipment and quality control in China manufacturing.

Common questions

Should I use the factory's own QC records or an independent inspector?

Both, for different jobs. In-house records — moisture logs, dimension logs, coat and cure logs — tell you whether the mill controls its process, and asking to review them is a reasonable request that a serious plant can meet. An independent report tells you whether this lot conforms. Ask which records exist and in what form before you assume either.

Does a hand-finished floor need a different inspection from a UV-coated one?

Yes, in one specific way. Machine-coated boards are consistent within a lot and their failure modes are film defects — orange peel, craters, edge build. Hand-rubbed multi-coat oil varies board to board by design, so the inspection question shifts from "is it identical" to "is the variation inside the approved range, and is it cured." Weight the sampling toward finish and cure accordingly, and bring the signed sample range to the warehouse.

How much does this cost, and is it worth it on a small order?

A standard man-day plus loading supervision is a small fraction of a container of precious hardwood flooring. The asymmetry is what decides it: the cost of inspecting is knowable in advance, and the cost of installing a wet or mismatched floor in a finished building is not.

What if the mill has never hosted a third-party inspection?

That is a question to ask directly rather than assume the answer to. A plant that has hosted SGS, BV or Intertek before knows the routine and the day runs smoothly. A plant that has not is not disqualified, but budget more time, send the checklist in Chinese and English in advance, and confirm who at the plant will accompany the inspector.

What to ask the supplier next

Before the goods-ready date, send the mill this list and ask for written answers:

• Will you accept ISO 2859-1, general inspection level II, at the AQL we specify, with the attached defect classification table?
• Which in-house QC records can the inspector review on the day — moisture, dimension, coat and cure logs?
• What is the carton and pallet specification, the m² per carton, and the cartons per 20ft and 40ft container?
• Will stair parts, skirting and mouldings from the same timber lot be staged beside the floor lot so they can be inspected together?
• What desiccant type and quantity do you load for a sea shipment of oiled wood flooring, and will you allow loading supervision?
• Have you hosted third-party pre-shipment inspections before, and for which agencies?

A plant that runs quality inspection as a named step in its own chain — as Anrantabu Flooring lists it alongside substrate preparation, parquet assembly and hand-rubbed oil finishing at its Huizhou base — should be able to answer the records question from a file rather than from memory. That, not the eventual pass or fail, is the signal worth reading. If your production date is still soft, the lead-time breakdown explains why the goods-ready date moves and how to pin it down.