AQL sounds like statistics, but in furniture sourcing it is really a negotiation tool. It decides how many units an inspector pulls, how many defects are tolerable, and whether a container ships or gets reworked.
Buyers who do not understand the tables either over-inspect small orders or accept lots that should fail, and both mistakes cost real money at retail.
This guide explains the AQL concept, walks through the sampling tables furniture importers actually use, and shows how to write the clause into a purchase order.

What AQL actually means
AQL stands for Acceptable Quality Limit: the worst defect rate, in percent, that a buyer agrees to accept as a process average across repeated lots. It comes from the ISO 2859 family of sampling standards, summarized in the reference article on acceptable quality limits. The counter-intuitive part is that an AQL inspection does not measure the exact defect rate of your lot; it makes a pass/fail decision with defined probabilities. A lot right at the AQL boundary has roughly a 90 percent chance of passing, which is the whole point: the system is designed to reject lots that are clearly worse than agreed, not to guarantee zero defects. Furniture importers typically set AQL 2.5 for major defects (functional failures, visible surface damage) and AQL 4.0 for minor defects (small cosmetic blemishes on non-visible surfaces), with critical defects at zero tolerance.
How the sampling tables work
The tables have three moving parts: lot size, inspection level and the AQL column. Under General Inspection Level II, which is the default for furniture, you find your lot size band in the first table to get a letter code, then read the letter code against the AQL column in the second table to get the sample size and the accept/reject numbers. A lot of 200 chairs gives code G and a sample of 32 units; a lot of 3,000 gives code K and 125 units; a lot of 10,000 gives code L and 200 units. Interactive calculators like the one at SQC Online automate this, but buyers should know the reference values because factories sometimes quote sample sizes from cheaper Special Levels (S-2, S-3), which cut the sample to 8 or 13 units and quietly weaken the inspection. How these random pulls differ from checking units mid-line is covered in our inline vs final inspection comparison.

The sampling numbers furniture buyers use
| Lot size (units) | Sample size (Level II) | Ac / Re at AQL 2.5 | Ac / Re at AQL 4.0 |
|---|---|---|---|
| 91 to 150 | 20 | 1 / 2 | 2 / 3 |
| 151 to 280 | 32 | 2 / 3 | 3 / 4 |
| 281 to 500 | 50 | 3 / 4 | 5 / 6 |
| 501 to 1,200 | 80 | 5 / 6 | 7 / 8 |
| 1,201 to 3,200 | 125 | 7 / 8 | 10 / 11 |
| 3,201 to 10,000 | 200 | 10 / 11 | 14 / 15 |
Reading the table: for a 3,000-unit chair order, the inspector pulls 125 random units; if 7 or fewer carry major defects the lot passes at AQL 2.5, and at 8 or more it fails and triggers rework or a second inspection. Notice the sample size grows slowly relative to lot size, which is why inspection fees stay roughly flat between a 1,000-unit and a 10,000-unit order. The three-stage framework that decides when to run this pull, earlier mid-production versus pre-shipment, is in our inspection stages guide.
Classifying furniture defects before the count
The table only works if defect classes are defined before inspection day. A practical furniture classification: critical defects include exposed staples or sharp edges, missing safety hardware on bunk structures, and any mold or odor issue, all scored zero tolerance. Major defects include visible scratches over 5 mm on finished surfaces, drawer slides that jam, joints with gaps above 2 mm, and wrong dimensions beyond 5 mm. Minor defects include glue residue on hidden surfaces, small color variation within the same batch and packaging label typos. Write these definitions with measurable thresholds into the purchase order, because “visible scratch” means something different to your team and to the factory QC lead. A real rework cycle triggered by a failed pull, including what the factory did to fix it, is documented in our mattress QC case study.
Writing the AQL clause into a purchase order
Five fields make the clause enforceable. First, the standard: name ISO 2859-1 General Inspection Level II explicitly. Second, the thresholds: AQL 0 for critical, 2.5 for major, 4.0 for minor. Third, the sampling basis: random carton selection across the full lot, at least 80 percent of units packed. Fourth, the consequence: fail means factory-paid sorting or rework plus one free re-inspection; second fail gives the buyer the right to cancel. Fifth, the witness option: the buyer or an agent such as Riwick witnesses the pull on site, which is how the clause becomes real instead of theoretical. Riwick runs AQL-based pre-shipment inspections as part of its sourcing service, with reports including photo evidence per defect, and inspection scope is quoted through the contact page.
Frequently asked questions
What AQL level is normal for furniture?
Most importers use AQL 2.5 for major defects and 4.0 for minor defects under General Inspection Level II, with critical defects at zero tolerance.
How many units get inspected in a 3,000-unit order?
125 units at Level II; the lot passes at AQL 2.5 with 7 or fewer major defects and fails at 8 or more.
Does passing AQL guarantee zero defective units?
No; AQL is a pass/fail decision system, and a lot right at the boundary still has roughly a 90 percent chance of passing, so some defective units always remain inside an accepted lot.
Why should buyers avoid Special Inspection Levels?
Special Levels cut samples to 8 or 13 units regardless of lot size, which weakens detection power; factories may suggest them because they pass more easily.
For the full wardrobe sourcing sequence, see our wardrobe sourcing roadmap.
For more detail, see our Upholstery Defects.




0 Comments