A new sweater PO is rarely lost to a single dramatic defect. It is lost to a small set of recurring quality risks that everyone in the supply chain has seen before — and that almost always could have been caught earlier than they were. After running OEM and private-label programs out of Dongguan since 2018, we have learned that the ten risks below cover the overwhelming majority of buyer complaints on knit programs at 3GG through 14GG. They are not exotic. They are familiar enough that a good QC plan, a complete tech pack, and the right sampling conversation can keep them from ever reaching final inspection.
This guide is written for the QC lead reviewing risk on a new sweater PO. For each risk, we cover what it looks like on the finished garment, the root cause inside the factory, when in the production timeline it tends to surface, the specific QC step that catches it, and one question you can ask the supplier during sampling to confirm the risk is being managed — not just accepted.
💬 Sourcing this for an upcoming project? Request a free quote → · MOQ 30 pcs · Sample in 7–15 days · WhatsApp us
Knitwear is a built fabric, not a cut fabric. The yarn, the gauge, the stitch density, the linking, the steaming and the finishing all interact, which means a defect at one stage can hide until two stages later. A linking problem may not look obvious until after steam-press. A shrinkage problem may not appear until after the first wash test. A dye-lot mismatch may pass on a single panel and fail across an assembled sweater. The implication for a QC plan is that inspection points must be staged, not concentrated at the end. The earlier you catch a risk, the cheaper the correction — re-knitting a panel before linking costs a fraction of re-making an assembled and trimmed sweater.
The 10 risks below are sequenced roughly by where they originate in the process, from yarn-level to packing-level. None of them are unique to one factory or one country; they are the universal failure modes of the category.
What it looks like: small fiber balls on high-friction zones — underarm, side seams, cuff folds, where a bag strap rides — appearing after the first few wears or wash cycles.
Root cause: short staple fibers, low yarn twist, soft brushed finishes, and loose stitch density all increase pill propensity. Pure soft-handle cotton or low-grade lambswool blends are more vulnerable than tight-twist worsted wool or anti-pill acrylic blends.
When it surfaces: during sample wear-testing and after wash testing on the bulk sample. Once shipped, it surfaces with the consumer in the first 30 days.
QC step that catches it: Martindale or ICI pill-box test on bulk-yarn sample (industry rating 3-4 minimum for daily wear), plus a controlled five-cycle wash-and-wear evaluation on the PP sample.
Ask the supplier during sampling: "Can you run a pill test on the bulk yarn before we approve PP, and share the rating with the wash test report?" If the yarn is borderline, agree in writing whether you will accept it, request a tighter twist, or switch to an anti-pill blend.
What it looks like: body length, sleeve length and chest measurement reducing more than the tech-pack tolerance after the first home wash. Cotton-rich blends and untreated wool are the typical offenders.
Root cause: the wash-and-block step during finishing was rushed, the steaming temperature was inconsistent, or the yarn itself was not pre-relaxed. Cotton fiber relaxes on contact with water; wool felts under heat and agitation.
When it surfaces: in the PP-sample wash test. If skipped at PP, it surfaces in bulk QC measurements after wash, or with the end consumer.
QC step that catches it: wash test on the PP sample using the actual care-label instructions, with measurements taken before and after. Tolerance is typically +/- 3% on body length and chest, +/- 4% on sleeve.
Ask the supplier during sampling: "What is the documented shrinkage rate for this yarn and gauge in your wash test, and how is the bulk finishing set to compensate for it?" A serious factory will already have a number.
What it looks like: the side seam spirals around the body instead of running straight from underarm to hem. Often visible on the wearer within the first minute, especially on chunky or oversized fits.
Root cause: uneven stitch tension between front and back panels, or asymmetric steaming and blocking after linking. Fully fashioned panels are less prone; cut-and-sew on knit fabric is more prone.
When it surfaces: at linking and again after final steam — but most clearly when the sweater is hung on a body form.
QC step that catches it: body-form inspection at finishing and again at final QC. Lay-flat measurements do not catch torsion; the garment must hang naturally.
Ask the supplier during sampling: "Are you knitting this fully fashioned or cut-and-sew, and can you show me the side-seam alignment on a dressed body form, not just the lay-flat photo?"
What it looks like: subtle shade variation between sleeves and body, between front and back, or across packed cartons. Often invisible under factory light, very visible in daylight or store lighting.
Root cause: yarn from two dye lots used in one garment, or two lots split across the PO without matching. Solid dark colors (navy, black, deep brown) and metameric grays are the highest risk.
When it surfaces: at panel inspection if the linker is alert; otherwise at final QC under daylight or D65 light box; otherwise on the buyer's retail floor.
QC step that catches it: all panels of one garment knitted from one dye lot, lot numbers recorded against bundle tickets, and a D65 light-box check at final inspection.
Ask the supplier during sampling: "How many dye lots will be used to cover the bulk yarn quantity for this color, and how will lots be segregated across the PO?" Confirm that one garment never spans two lots.
What it looks like: hangtag attached to the wrong garment location (left underarm instead of right back neck), pierced through the knit body causing a hole, or attached with the wrong string color or knot direction.
Root cause: a packing instruction that was verbal, not visual; a packing operator who has not seen the spec photo; or a last-minute hangtag change not communicated to the finishing floor.
When it surfaces: at final QC and at the buyer's DC during receiving — too late for cheap correction.
QC step that catches it: a photo-illustrated hangtag-attachment spec in the packing instruction, and during-process inspection during packing.
Ask the supplier during sampling: "Can the packing instruction include a photo of the exact hangtag location, string type, and knot, so the finishing team works from an image not a description?"
What it looks like: care label or size label sewn 5-10mm off the spec position — too far down the side seam, twisted, or facing the wrong way. Brand label drift on the inside neck is the most visible.
Root cause: labels sewn before the side seam is fully set, or sewn with no fixed reference point in the operator's jig. Drift accumulates across an order.
When it surfaces: during in-line label inspection at finishing; if missed, at final QC.
QC step that catches it: in-process inspection at the labeling station with a measurement gauge against the tech-pack reference, plus AQL inspection at final QC.
Ask the supplier during sampling: "How is the operator setting the label position — by eye, or with a jig referenced to the side seam? Can you confirm the same jig will be used on the bulk run?"
What it looks like: crew or mock-neck rib that no longer recovers — it gapes flat against the chest, or the V-neck point sags. Often appears after one or two on-and-off cycles by the wearer.
Root cause: insufficient rib density, rib yarn twist too soft, the linking at the neck attachment over-stretched the rib, or the steaming flattened the rib structure.
When it surfaces: during the fit sample and again after the wear test on the PP sample.
QC step that catches it: a stretch-and-recover test on the PP sample (stretch the neck 30% beyond resting circumference, hold 10 seconds, measure recovery within 60 seconds), plus a visual check after fitting on a mannequin head form.
Ask the supplier during sampling: "What is the rib gauge and stitch count at the neck, and can you run a stretch-recovery test on the PP sample neckline before we approve?"
What it looks like: uneven linking stitches at shoulder, neck attachment or side seam — some stitches too tight (puckered), some too loose (gaping), or a visible step in the linking line.
Root cause: mixed operators on the same PO, an unmaintained linking machine, or yarn changes between sample and bulk that the linker did not adjust for.
When it surfaces: at linking inspection if the factory runs one; otherwise at final QC and very often after the first wash, when linking stress shows.
QC step that catches it: in-line linking inspection (one inspector at the linking station), plus a seam-pull test on the PP sample (stretch the linked seam to 50% extension and check for stitch separation).
Ask the supplier during sampling: "Will the same linking operator and machine setup be used across the bulk, and can the PP sample have a seam-pull test documented?"
What it looks like: size M measuring close to L, size XL with sleeves too short, or the size jumps between L and XL being inconsistent across body, sleeve and chest. International size variance (US M vs EU 50) makes this worse.
Root cause: the tech-pack size chart was incomplete, the grading rules were assumed rather than written, or the factory pattern-maker graded only chest and applied a flat ratio to other measurements.
When it surfaces: when the fit sample is approved in M but bulk arrives across the size run.
QC step that catches it: a full-set size sample (XS through 3XL or the actual range) measured against the spec, before bulk knitting starts. Final QC must include one piece from every size in the AQL pull.
Ask the supplier during sampling: "Can we approve a size set, not just one size, before bulk yarn is committed?" The cost of a few extra samples is much lower than the cost of mis-graded bulk.
What it looks like: crease lines that will not steam out, hanger marks at the shoulder, tissue-paper transfer staining, or polybag scuffing on dark colors. Carton crushing during transit adds creases at the fold line.
Root cause: rushed folding without tissue, polybags too tight, cartons over-filled, or a fold pattern that puts the printed graphic on the outside fold where it creases.
When it surfaces: at the buyer's DC during receiving and at the retail floor when garments are unboxed.
QC step that catches it: during-packing inspection on fold pattern, tissue placement, polybag fit and carton fill, plus a drop test on the packed carton at final QC.
Ask the supplier during sampling: "Can you send a fully packed master carton sample, including polybag, tissue, hangtag, and carton mark, so we can pressure-test the packing spec before bulk?"
|
| 1 | Pilling | PP wear test / consumer | Martindale or ICI pill test + 5-cycle wash | Pill rating on bulk yarn |
| 2 | Shrinkage | PP wash test / bulk | Wash test PP sample, measure before/after | Documented shrinkage rate |
| 3 | Twisted side seam | Body-form check at finishing | Hung-form inspection (not lay-flat) | Fully fashioned vs cut-and-sew |
| 4 | Dye lot mismatch | Final QC / store floor | D65 light box, one lot per garment | Lot count and segregation plan |
| 5 | Hangtag attachment | Final QC / buyer DC | Photo packing spec + in-process check |
A QC plan that treats all ten risks equally will fail. Sequence them. Risks 1-4 are yarn-and-knitting-stage decisions — they have to be settled before bulk yarn is dyed and committed. Risks 5-8 are construction-stage controls — they need in-line inspection and a clear PP-sample sign-off. Risks 9-10 are pre-shipment controls — they need a strong AQL inspection plan and a master-carton dry run. A useful rule of thumb is that any risk caught at PP sample is roughly an order of magnitude cheaper to correct than the same risk caught at final inspection, and another order of magnitude cheaper than catching it at the buyer's DC.
We typically build the inspection plan around four checkpoints: bulk yarn approval (covers pilling and shrinkage), PP sample approval (covers risks 1-9 in sample form), during-production inspection at roughly 50-60% completion (covers risks 3, 5, 6, 8 on real bulk units), and pre-shipment AQL at 100% packed (covers risks 4, 5, 6, 9, 10). Where the program is high-risk — a new yarn, a new style block, a new factory — adding a third-party inspector at one of these points is usually worth the cost.
We work with most of our buyers under an OEM/ODM model, with catalog MOQ from 30 pcs per color, sample lead time of 7-25 days and bulk lead time of 30-45 days for typical programs in 5GG, 7GG and 12GG (3GG and 14GG when the design calls for it). The 7-25 day sample window is what makes risk-based QC realistic: it gives room for a proper PP sample with wash test, stretch-recovery test, and size-set check before bulk yarn is dyed. AQL inspection and needle/metal detection are standard at final QC. We are not unique in this — many serious OEMs in Dongguan, Guangdong work to a similar discipline. The difference between programs that ship cleanly and programs that do not, in our experience, is whether the buyer's tech pack and packing spec are detailed enough for the factory to inspect against, not just produce against.
For buyers building a new sweater program and unsure where the risk sits, the most useful starting point is usually two documents: a complete tech pack (including size chart, wash test target, yarn pill rating target, and a labeled side-by-side packing photo) and an inspection plan that names the QC step against each of the ten risks above. With those two in hand, the conversation with the factory shifts from "please make it right" to "please confirm against these checks" — which is the conversation that actually catches risk in time.
---
If you're planning a real project around any of the points above, we'd be glad to take a quick look. Send a short brief and we'll come back within one business day with a practical direction, MOQ + lead time estimate, and a sample plan if it makes sense.
→ [Request a free quote](/request-quote) · [WhatsApp us](https://wa.me/8615170244792) · info@lcgarment.cn
