AQL Inspection for Sportswear: An OEM Quality Control Guide
AQL Inspection for Sportswear: An OEM Quality Control Guide
"AQL 1.5 or 2.5?" — that single line on a purchase order is the one most buyers tick without thinking, and the one that decides whether a shipment survives. Many brands treat it as an inspector's preference. It is not. It is a hard trade-off between risk and cost: set it too loose and your warehouse fills with stock you end up discounting; set it too tight and rework rates and unit prices climb with it.
Last year a Canadian yoga brand placed 3,000 pairs of leggings with a factory in China for the first time. The previous factory — a competitor — had assured them before shipment: "We inspected everything. No problem." When the cartons were opened at the warehouse, one box held four different sizes, several seams had skipped stitches, and the waistband elastic was tight on one side and loose on the other. The brand cleared the whole batch at a discount and paid a penalty to the retailer. The review afterwards found that the factory had no sampling plan at all — workers simply pulled a few pieces that "looked fine."
Here is the counterintuitive truth: AQL is not a numbers game about how many pieces get checked. It is the line that decides whether a shipment is safe to send. Using a set of statistical rules, it gives you a quantifiable accept-or-reject verdict at a standard cost — instead of trusting a veteran inspector's eyes.
This is written for sourcing managers, quality leads, and founders of sportswear brands. We are a sportswear factory supplier running OEM and ODM production in Dongguan since 2014, across two facilities totalling 20,000㎡ with a monthly capacity of 500,000 pieces. What follows — how to pick the AQL level, how inspection actually runs, and how to read the report — comes from working through 1,200+ brand orders, not from a textbook. And because inspection is only half of the size story, our size chart and grading guide covers the measurement side of the same problem.
The Six Ways Brands Get AQL Wrong
- Believing "100% inspection" is safer. Inspecting every piece in a 10,000-piece run does not survive contact with labor cost. An unrealistic standard quietly becomes inspection in name only.
- Letting workers "eyeball it" instead of sampling. Without a statistical plan, today is strict and tomorrow is loose — same style, different batches, different quality.
- Filling in the AQL level casually. The supplier does not ask, so someone writes 4.0. When a problem appears, the contract says 4.0 — and you have nothing to argue with.
- Treating AQL as a zero-defect promise. AQL means acceptance quality limit, not a guarantee of no defects. Misread it and expectations go wrong on both sides.
- Running only a final inspection. By the time finished goods pile up in the warehouse and a batch-wide stitching problem surfaces, the window for cheap rework has already closed.
- Skipping defect classification. If critical, major, and minor defects are not defined, nobody can say which finding should actually stop a shipment.
Why these six traps turn directly into money: a batch-wide skipped-stitch problem caught at final inspection is not "re-sew a few pieces." It means unpacking the whole batch, reworking, re-inspecting, and re-packing — a conservative 15%–30% of the order's margin, plus air freight to recover the delay and penalties to the retailer. A QC standard written in the wrong place never shows up on the unit price. It shows up after sales.
What AQL Actually Means
AQL — Acceptance Quality Limit — is not "the maximum number of defects allowed." It is a statistical pass line based on a sampling standard (ANSI/ASQ Z1.4, equivalent to ISO 2859-1): draw the sample size for your lot, count the defects, and accept or reject depending on whether the count falls inside the acceptance range. It does not promise zero defects. What it gives you is a quantifiable answer that balances cost against risk — the same standard your third-party inspector will apply.
How to Choose Your AQL Level
This one decision shapes about 80% of your inspection outcome.
| Category / requirement | Recommended AQL (major) | Why |
|---|---|---|
| Regular sportswear / outerwear | 2.5 | The common industry line |
| Next-to-skin / yoga / kidswear / demanding orders | 1.5 | Tighter; fewer defects allowed |
| Trims / packaging / non-critical items | 4.0 | Can be relaxed |
Our outgoing default is AQL 1.5 — because on next-to-skin sportswear, loose threads, skipped stitches, and shade variation are the defects your customer sees first. Stricter levels mean fewer accepted defects but higher rework and unit cost, which is exactly why the level belongs in the contract, agreed before production — not negotiated after a problem.
How Sampling Actually Works
No feelings involved: look up the lot size in the sampling table at General Inspection Level II, draw the corresponding number of pieces, then read off the accept (Ac) and reject (Re) numbers for your AQL level. For example, a lot of 1,201–3,200 pieces falls in code letter K and samples 125 pieces; a lot of 3,201–10,000 pieces falls in code letter L and samples 200 pieces. At AQL 1.5 the allowed major-defect count is clearly tighter than at 2.5. The exact Ac/Re numbers always come from the ANSI/ASQ Z1.4 table — never from an inspector's judgment call.
Small orders are not exempt. At MOQ 300, your lot still lands in a sampling band with a proportionally higher sampling ratio — a small order is held to the same standard, not a lighter one.
In-Line and Final Inspection: Do Both
| Factor | In-line + final (our standard) | Final only |
|---|---|---|
| When problems surface | During production — early | After finishing — late |
| Rework window | Still open | Basically closed |
| Batch-wide defect control | High | Low |
| Cost | Slightly higher — saves the rework | Looks cheaper — it is a gamble |
In-line inspection samples during cutting, sewing, and pressing, so a problem stops the line while the fix is still cheap. Final inspection is the last AQL sample before packing — the gate before shipment. Factories that only run final inspection discover batch problems when it is already a mountain. We build both stages into the 8-step process below, and both sit inside the production timeline we break down in our sportswear lead time guide.
Defect Classification: Critical, Major, Minor
- Critical: safety or regulatory problems (drawcords that fail a standard, sharp points) — zero tolerance.
- Major: appearance or functional defects that affect salability — skipped stitches, holes, mixed sizes.
- Minor: small blemishes that do not matter at arm's length — an occasional loose thread end.
Classification is what settles disputes. When the report says "3 majors, 1 minor," both sides are reading the same ruler instead of arguing past each other.
Our 8-Step Quality Check, Stage by Stage
Saying "8 steps" means nothing without the detail. Here is what each step actually checks — and it doubles as the list you should walk through when you audit any factory's QC proposal:
- Incoming inspection. Fabric shade variation between dye lots, GSM, stretch recovery, and trims (zippers, woven labels) checked against the approved specification.
- Cutting inspection. Pattern matching, missing corners, miscut panels, ply-count deviation.
- In-line sewing patrol. Stitch density, skipped stitches, thread breaks, seam allowance, and symmetry (left/right sleeves, pockets) sampled during the process.
- Pressing inspection. Glazing, scorching, deformation, water stains.
- In-process sampling. Semi-finished garments sampled by ratio after sewing, to stop batch-level errors from flowing downstream.
- Final AQL inspection. Finished goods sampled to Z1.4 before packing, counted separately as critical / major / minor.
- Packing inspection. Fold method, polybag spec, hangtag content, carton quantity, shipping marks.
- Pre-shipment check. One last sample before container loading, confirming the final result and no mixed cartons.
What a Real Inspection Report Looks Like
Plenty of factories say "we provide reports." But a report the buyer cannot read is no report. Below is the format we actually use on inspection days — the figures are illustrative; real numbers differ by lot. Use it as a ruler and ask any factory: does your report carry these fields?
| Report field | What it means | Example value |
|---|---|---|
| Lot size | Total pieces in the lot | 3,000 pcs |
| Sample size | Per Z1.4, Level II (code K) | 125 pcs |
| AQL (major / minor) | Acceptance limits | 1.5 / 4.0 |
| Critical defects | Zero-tolerance count | 0 |
| Major defects | Found in sample | 2 (Ac = 3, Re = 4) |
| Minor defects | Found in sample | 7 (Ac = 10, Re = 11) |
| Result | Accept / reject | Accept |
| Defect photos | Typical defects photographed | 3 attached |
Note: the figures above demonstrate the format — which fields a report needs and how Ac/Re is read. Real sample sizes and defect counts move with the lot size, the AQL level, and the quality of that day's production, always per the Z1.4 table for the lot. We do not treat example figures as commitments.
How to read any report in three steps:
- Does the sample size match the lot band? A small lot "exempted from sampling" is a red flag.
- Is critical = 0? Any critical defect should mean an automatic reject.
- Is major ≤ Ac? Above Re, the shipment should be rejected — "close enough" is not a verdict.
Where Common Defects Get Caught
| Defect type | Usually originates | Where we intercept it |
|---|---|---|
| Shade variation / dye-lot mismatch | Fabric incoming | Incoming inspection |
| Skipped stitches / thread breaks / asymmetry | Sewing | In-line patrol + final AQL |
| Scorching / deformation | Pressing | Pressing inspection |
| Mixed sizes / missed checks | Packing | Packing inspection + pre-shipment check |
| Safety / regulatory (drawcords etc.) | Any stage | Double-gated: incoming + final, zero tolerance |
Printed defects follow the same logic at the decoration stage — which is why print methods, cracking tests, and wash fastness get locked at sampling, as we explain in our screen print vs sublimation guide.
How Your AQL Level Changes the Outcome
| Dimension | AQL 1.5 | AQL 2.5 | AQL 4.0 |
|---|---|---|---|
| Defects allowed | Fewest (strictest) | Moderate | Most (loosest) |
| Best for | Next-to-skin / yoga / kidswear | Regular sportswear | Trims / packaging |
| Rework and unit cost | Higher | Moderate | Lower |
| Complaint risk | Lowest | Moderate | Higher |
The Pre-Shipment QC Checklist to Copy Into Your Contract
Write these lines into your purchase order or inspection instruction — ten minutes of drafting beats a month of disputes afterwards, and it works just as well as a checklist to circulate to your team:
- State the AQL level: major and minor separately (next-to-skin default 1.5 / 4.0).
- State the inspection level: General Inspection Level II, or tightened and switching rules.
- Define the defect classes: what counts as critical / major / minor for your product.
- Require in-line + final inspection — not final only.
- Require the report to carry: lot size, sample size, Ac/Re, defect photos, verdict.
- Third-party re-inspection clause: whether SGS / BV is engaged, and who pays.
- Sealed sample retention: the bulk reference both sides measure against.
- No "eyeball" verdicts: every accept decision must be backed by sampling data.
The sealed-sample line matters more than it looks. A bulk run can only be judged against an approved reference — the same principle behind our tech pack guide: the spec you write before production is the spec you enforce after it.
Questions Buyers Actually Ask
What AQL should I use for sportswear?
Regular sportswear and outerwear: AQL 2.5 (major). Next-to-skin, yoga, kidswear, and demanding orders: 1.5. Our outgoing default is 1.5 — on next-to-skin garments, loose threads, skipped stitches, and shade variation are exactly what the customer notices, so we do not loosen it. Put the level in the contract; it gives you leverage that after-the-fact negotiation never does.
What is the real difference between AQL 1.5 and 2.5?
The number of defects allowed. On the same sample, AQL 1.5 tolerates fewer major defects than 2.5, so the verdict is stricter. In plain terms: 1.5 is safer but slightly higher in rework and unit cost; 2.5 is looser but complaint risk rises. For next-to-skin and demanding orders we do not recommend anything looser than 1.5.
Can I go stricter than AQL 1.5?
Yes — 100% inspection on critical zones, or AQL 1.0. But weigh it: stricter levels push rework rates, unit price, and sometimes lead time. We recommend spending the strictness on critical and major items while keeping minor items reasonable, rather than tightening everything across the board.
In-line vs final inspection — what is the difference, and do we need both?
In-line samples during cutting, sewing, and pressing: problems surface early and the line stops while rework is cheap. Final is the last Z1.4 sample before packing — the gate before shipment. We run both. A final-only factory finds batch-wide skipped stitches when they are already a mountain and the rework window has closed.
Does the factory inspect, or a third party?
Either. We self-inspect and issue inspection records; if you want independent backing, a third party (SGS / BV) can re-inspect before shipment. Reports travel with the goods — if a dispute ever comes, the sampling data and defect distribution are the evidence.
What are critical, major, and minor defects?
Critical: safety or regulatory issues such as drawcords out of standard — zero tolerance. Major: appearance or functional defects that hurt salability, like skipped stitches, holes, or mixed sizes. Minor: small blemishes invisible at arm's length, like an occasional thread end. The point of classification is a shared ruler when disputes happen, so both sides stop talking past each other.
Do you run AQL on small orders (MOQ 300)?
Yes. The smaller the lot, the relatively higher the sampling ratio — the sampling table steps by lot band, and small lots are never exempt. A MOQ 300 order is held to the same standard as a bulk run; the lower threshold does not relax quality.
The report says Reject. What happens next?
The line stops first. The report lists defect types and counts, the stage responsible, and the corrective action; the lot is then re-sampled at a tightened level and released only when it passes. A reject verdict is not the end — it is the brake that stops problem goods from shipping. It is also why in-line patrol matters: catch it early and you never reach that brake.
Can I witness the inspection?
Yes. At final inspection you can send your own QC or a third party to witness on site; remotely, we provide defect photos and live records to follow along. Book the time in advance and it slots into the normal FOB Shenzhen schedule without touching the timeline.
Send Us Your QC Standard
Send us your inspection standard — AQL level, defect classification, any special requirements — and we will return an inspection plan with the sampling schedule: which line we will hold, at which gate, before we talk about anything else. Put the level in the contract, and bulk production stops being a gamble.
As a sportswear manufacturer in Dongguan since 2014, we would rather build a long-term partnership than take one order at a time. If anything about quality control is unclear, ask — we will give it to you straight. To see our production and quality setup first, visit our OEM factory page.
