Start with the job, not the cloth name
Microfiber is not one product family. A 250 GSM coral fleece drying towel, a 220 GSM short-pile glass cloth, and a 320 GSM warp-knit all use polyester/polyamide yarns, but they behave differently in wiping drag, particle pickup, and wash recovery. If your RFQ only says "microfiber cleaning cloth" the factory will fill the gap with whatever construction meets your target price.
For this reason, we ask buyers to define the surface and the soil before we quote. Stainless steel cleaner, eyewear wipe, gym equipment disinfectant, and car interior detailing each need a different pile height and edge treatment. This is also where many importers accidentally compare unlike offers from two mills and assume one is simply cheaper.
| Use case | Recommended construction | Typical GSM | Main risk if underspecified |
|---|---|---|---|
| Glass and mirror | short pile warp knit, tight density | 190-230 | streaking from too much pile drag |
| General household wipe | terry microfiber, medium pile | 250-300 | low pickup and quick flattening |
| Auto interior and screens | suede or short-pile knit | 180-240 | scratch perception from hard edge or trapped debris |
| Water absorbent drying | twisted or coral fleece | 350-480 | slow drying and seam distortion if edge is heavy |
- Ask for the actual knitting type: warp knit, weft knit, suede, coral fleece, twisted loop
- State the target wash temperature: 40°C domestic, 60°C commercial, or higher trial condition
- List the chemical contact: alcohol spray, quat disinfectant, glass cleaner, wax residue
- Name the surface finish sensitivity: piano black trim, coated lens, chrome, ceramic glass
The construction details that separate usable stock from returns
The two most common material mixes are 80/20 and 70/30 polyester/polyamide. The higher polyamide share usually improves capillary pickup and handfeel, but not every application needs it. For routine household or janitorial wipes, 80/20 is often enough. For glass and detailing, 70/30 can justify the extra cost because it leaves less residual moisture on the pass.
Blend ratio alone does not decide performance. Split fiber quality matters more. A mill can quote 80/20 on paper, yet if the splitting is shallow or inconsistent, the cloth will feel slick instead of grabby. We check this during development by wiping a black acrylic panel and a polished stainless coupon, then reviewing under angled light for haze and lint transfer.
| Spec line | Commercial range | What to lock before sampling |
|---|---|---|
| Blend | 80/20 or 70/30 | keep ratio tolerance within ±3 percentage points |
| Finished size | 30x30 cm to 40x40 cm | state tolerance; we use ±1.0 cm after finishing |
| Finished weight | 18 g to 60 g per piece | approve per-piece weight, not only GSM |
| Shrinkage after 3 washes | 2.5% to 6.0% | set max by size-sensitive application |
| Edge finish | overlock, hidden edge, ultrasonic cut | match edge to scratch and cost requirement |
Edge finish is one of the easiest places to save a few cents and create a return problem. On cloths used against glossy surfaces, a coarse overlock seam can become the complaint even when the pile itself is fine. Ultrasonic cutting avoids thread edges, but the cut line must be stable; weak sealing can open after repeated wash cycles, especially on lower-density fabric.
A real RFQ comparison: why two similar offers landed 21% apart
Last quarter we quoted a European cleaning brand for a 38x38 cm blue cloth packed 10 pieces per polybag, 200 pieces per carton. They had another offer that was lower by USD 0.052 per piece, but the sample from that source came in at lighter actual weight and with a different edge build than the quote implied.
| Line item | Our sample | Competing sample | Commercial effect |
|---|---|---|---|
| Actual finished weight | 31.4 g | 27.9 g | lighter cloth loses body after wash |
| Blend declaration | 80/20 tested in-house by supplier paperwork | 80/20 declared only | lower verification confidence |
| Edge | ultrasonic sealed all four sides | 3-thread overlock | higher scratch complaint risk on screens |
| Dimensional stability after 3x 60°C wash | -3.1% length / -2.7% width | -6.4% length / -5.9% width | pack count and fit variance |
| Lint test on black glass | trace, acceptable | visible transfer | reclean needed |
The cheaper offer looked workable until the buyer recalculated use life. Their target was at least 80 wash turns in a commercial cleaning program. If one cloth survives only 45 to 55 turns before edge distortion and lint complaints, the invoice saving disappears quickly. In that case the better-built cloth reduced annual replacement demand by roughly 28,000 pieces across their service network.
For the best microfiber cleaning cloths sourcing playbook, edge finish deserves its own checkpoint
We rarely see returns caused by shade variation alone in this category. More often the return starts with edge behavior: roping after wash, seam puckering, hard seam feel, corner curl, or a cut edge that opens. If the cloth is sold for glass, electronics, or auto detailing, edge selection should be written into the purchase order with a sample reference code.
- 3-thread overlock: lowest conversion cost, suitable for utility wipes, but not our first choice for delicate surfaces
- Covered edge / binding: cleaner retail appearance, though heavy binding can reduce flat contact on smaller cloths
- Ultrasonic cut: soft hand around perimeter, good for edgeless programs, but requires stable base fabric density
- Laser cut plus hem: uncommon in value programs; used when branding and cut precision matter more than speed
Our acceptance point for seam or cut integrity on bulk inspection is practical rather than cosmetic perfection. On a general AQL 2.5 inspection, we separate critical, major, and minor issues. Open seam longer than 15 mm, cut-edge split that can propagate by hand, or fused edge scorching visible from 50 cm goes into major defect territory. A slight corner wave under 6 mm lift on a single cloth may remain minor if function is unaffected.
Wash life is a spec, not a promise
Buyers ask for "machine washable 200 times" too casually. Without wash conditions, the claim means little. We prefer to write a benchmark like this: 40°C or 60°C domestic/commercial wash, neutral detergent, no fabric softener, tumble low or line dry, evaluated at cycle 1, 10, 25, and 50 for shrinkage, edge stability, absorbency, and linting. If bleach contact is possible, test that separately because many microfiber programs should avoid chlorine bleach altogether.
For internal development we use a simple bulk control sequence: precondition samples 24 hours, record finished dimensions and mass, run repeat wash/dry cycles, then inspect against the approved hand sample. We also rub the washed cloth across a mirror panel and a dark polycarbonate sheet because some defects only appear on reflective surfaces. This is more useful than a generic soft-hand comment on the lab report.
| Checkpoint after wash | Preferred benchmark for mid-grade cloth | Fail signal |
|---|---|---|
| Shrinkage after 3 washes at 60°C | within 4.5% each direction | over 6.0% or skewed distortion |
| Mass loss after 10 washes | under 3.0% | noticeable thinning and weak body |
| Edge integrity | no opening over 10 mm | split edge or seam run |
| Colorfastness to washing | grade 4 minimum | grade 3 or staining on adjacent fabric |
| Lint transfer on black glass | no visible lint at arm's length | reclean required after one pass |
Colorfastness standards should be named in the tech pack. We commonly reference ISO 105-C06 for colorfastness to domestic and commercial laundering, ISO 105-X12 for rubbing, and dimensional stability methods aligned with the buyer's lab protocol. For chemical compliance, our programs can be aligned with OEKO-TEX Standard 100 Class I where required, and factory system control follows BSCI and ISO 9001 procedures.
The QC thresholds we put on paper before bulk starts
A cloth category looks simple until inspection day. If your spec sheet lacks numeric tolerances, the conversation becomes subjective. We write tolerance bands directly into the PI and production card so merchandiser, knitting, cutting, and packing are working to the same limits.
- Finished size tolerance: usually ±1.0 cm for 30x30 cm to 40x40 cm cloths; tighter only if the application truly needs it
- Finished weight tolerance: ±6% per piece on regular bulk lots; for retail count packs we often tighten the carton average range
- Shade variation: against approved standard under D65 light source, no obvious lot mix within one sales pack
- Skew/bow: keep under 3% on check or directional constructions so folded presentation stays square
- Major defect threshold: no oil marks, no hard contamination, no holes, no seam opening above agreed limit
On microfiber, trapped contamination is a real issue. A tiny hard filament, needle tip fragment, or fused polymer bead can create scratch complaints out of proportion to its size. That is why we add a hand-feel and glide check before final packing on sensitive-surface programs. The operator runs the cloth across a black acrylic plate and rejects any piece with drag points.
Price bands in 2026: what changes the FOB number
For practical buying, piece weight and edge conversion move price more than color alone. Packaging also matters more than some buyers expect, especially when each cloth is barcoded or retail-carded.
| Construction example | MOQ | FOB China price band | Typical lead time |
|---|---|---|---|
| 30x30 cm, 220 GSM short-pile, overlock edge | 5,000 pcs | USD 0.18-0.26/pc | 18-24 days |
| 38x38 cm, 260 GSM terry microfiber, ultrasonic cut | 5,000 pcs | USD 0.29-0.41/pc | 22-30 days |
| 40x40 cm, 320 GSM edgeless detailing cloth, 70/30 blend | 10,000 pcs | USD 0.44-0.63/pc | 25-35 days |
| 40x60 cm, 400 GSM coral fleece drying cloth | 8,000 pcs | USD 0.72-1.05/pc | 28-38 days |
Our standard MOQ for custom towel and cloth programs is 500 pieces per design per color, but microfiber cleaning cloth economics become more stable once you move above 5,000 pieces because knitting efficiency, dye lot utilization, and pack setup all improve. If you want three colors split across a small run, we can do it, though the price per piece usually rises by USD 0.02 to 0.06 depending on carton complexity and label setup.
A recent janitorial-private-label order illustrates this. The buyer requested 12,000 pieces in four colors, each 36x36 cm, 240 GSM, printed belly band, 6-pack retail set. The cloth itself represented about three quarters of FOB value. Packaging added roughly 11%, barcode sticker and insert another 4%, and extra manual sorting by color ratio added just under 3%. That is why we prefer to quote fabric and pack-out as separate lines.
Sampling path that prevents the usual argument at bulk stage
For this category, one approval sample is not enough. We recommend three signoff points: first hand-feel and construction swatch, then pre-production sample in bulk color, then one washed reference retained by both sides. The washed reference matters because microfiber can change noticeably after the first laundering; a buyer who only signs an unwashed sample may reject bulk for a change that is actually normal finishing relaxation.
- Approve fabric hand and pile direction before discussing retail packaging
- Approve edge finish by physical sample, not by wording only
- Keep one washed approval standard with cycle details written on the tag
- Confirm carton count and barcode placement before the final PI
Related reads: if you are still building the RFQ, start with build-towel-tech-pack-that-mills-can-quote and microfiber-vs-cotton-towel-comparison. If your program is for automotive channels, auto-detailing-microfiber-towel-program goes deeper on pile selection and packaging formats.
Compliance and factory questions worth asking before deposit
Microfiber cleaning cloth buyers sometimes focus on unit cost and skip compliance verification because the product looks low risk. We do not recommend that. If the cloth will be used in homes, gyms, or retail channels, at minimum ask for current OEKO-TEX Standard 100 scope where claimed, social compliance such as BSCI, and process control under ISO 9001. Then ask a practical question: who actually knits, dyes, cuts, and packs the cloth, and which of those steps are subcontracted.
Subcontracting is not automatically a problem, but it changes lead-time reliability and defect tracing. A cloth that is knitted in one region, dyed elsewhere, and edge-finished by a third workshop may be fine on a stable repeat order. It becomes riskier on a first run with tight color tolerance or urgent replenishment.
Related reads: for compliance review, see how-to-read-oeko-tex-certificate. For quantity planning and MOQ trade-offs, negotiate-towel-moq-without-killing-margin is useful. For freight timing once goods are packed, container-vs-air-freight-towel-orders helps frame the shipment decision.
The short version buyers can put into the PO
If you want clean quoting and fewer sample loops, the PO should state these lines clearly: construction type, blend ratio, finished size tolerance, piece weight or GSM with tolerance, edge finish, target wash condition, acceptance criteria for shrinkage and seam integrity, packaging format, carton count, and inspection level. That gives the mill something objective to run.
A microfiber cloth fails less often because of one dramatic mistake than because six small unspecified details were left for the factory to guess.
We manufacture custom towel and microfiber programs in China with a 220-person team, annual output around 2.4 million pieces across categories, and MOQ from 500 pieces per design per color. For microfiber cloth programs, the faster route is to send your target use case, dimensions, edge preference, wash condition, and planned pack-out first. We can then tell you whether the spec is commercially balanced before sampling starts.
Build your microfiber cloth spec with us
Send the use case, target dimensions, wash condition, packaging plan, and expected order volume. We will reply with a workable spec, MOQ, FOB range, and production timing. WhatsApp: +86 13205717266 | Email: [email protected]
Request a quote →