Recommended starting spec buyers can copy into a PO
Use one construction name throughout the PO: 4-side double-turn hem, single-needle lockstitch, contrast thread. Keep the PO focused on finished-product acceptance. Leave machine settings, needle brand, lubrication system, folder style, and sewing sequence as supplier controls unless you are correcting a known repeat failure from earlier orders.
For a standard value-retail fleece blanket, the specification below is commercially workable. Tighten only the items that change claim risk, shelf appearance, or retail compliance.
Decision table: what belongs in the PO versus what stays supplier-controlled
The buying mistake is to freeze too many machine-level variables. A PO should define the result. The supplier should control how to achieve it, unless previous failures justify locking a process detail at PPS.
Contract-ready PO hierarchy
The quickest way to reduce disputes is to structure the spec in the same order a lab, PPS reviewer, and final inspector will use it. Keep contractual items grouped by product identity, dimensions, hem geometry, durability, appearance, and inspection.
Fabric stability and hem geometry
Edge problems often start upstream in the fleece base. Uneven brushing, weak heat-setting, broad widthwise tension variation, or unstable anti-pill finishing can create roping and edge draw-in even when sewing is technically sound. That is why a hem article still needs base-fabric discipline.
For 230gsm fleece, 230gsm +/-5% is a practical commercial range. Tighter tolerances can improve hand consistency, but they usually cost more than they save on entry-price programs. Put control effort first into shade continuity, pilling, and hemming appearance before chasing very tight GSM.
A double-turn hem is the correct base construction here. Around 12mm finished hem depth is usually the best balance for 230gsm fleece. Around 10mm can look cleaner on smaller throws but reduces tolerance for cutting variation. Around 14-15mm increases fold mass and can worsen edge roping if feed balance is not corrected.
Freeze one corner style at PPS. A squared folded corner is faster and cheaper. A mitred corner usually packs flatter and looks sharper, but it exposes cutting accuracy more clearly. The buyer-facing requirement should be symmetry, bulk, and durability, not the exact trimming motion used by one sewing line.
SPI protocol, stitch appearance, and seam performance
SPI is useful only if the count method is fixed. For lockstitch hems, count stitches over 25.4mm on straight runs, not at the corner, not within 100mm of the corner turn, and not across the overlap or start-stop zone. Take five readings per blanket: one per side plus one extra on the longest side. Accept on average 7-9 SPI, with no single reading below 6.5 or above 9.5. That avoids rejecting a blanket for a single local variation while still controlling appearance.
SPI does not prove seam durability. Buyers usually expect at least a simple extension route. For a hem claim article, the practical option is a low-load seam opening check rather than a full structural seam test on every lot. Apply 10N load to a 100mm hem section and measure the maximum opening at the stitch line. Accept max 1.0mm before wash and max 3.0mm after 3 wash cycles. This is more reproducible than 'light hand extension' or 'no seam opening over 5mm'.
If your retail channel has a higher abuse risk, add a seam-strength test by agreement. A simple route is grab or seam strength on a representative sewn strip, but many value-retail blanket buyers do not contract that unless prior failures justify it. If omitted, say so explicitly: the PO controls seam appearance and wash durability rather than a minimum destructive tensile value.
Model wash route buyers can write into the PO
Saying 'ISO 6330 and ISO 5077' is not enough. The laundering route needs the actual program, temperature, drying method, and cycle count. A commercially useful model route for retail fleece blankets is shown below. Adapt it only if the retailer has a house manual.
Model route: ISO 6330 domestic wash, front-loading program at 40C, normal agitation, ballast load to the machine mass required by the selected method using clean cotton ballast, standard reference detergent without optical brightener where required by the method, followed by line dry or tumble dry low exactly as agreed. Run 3 cycles for durability review. Recondition for at least 24h before remeasurement. Assess dimensional change to ISO 5077.
For dark fleece with light contrast thread, add post-wash visual review under D65 or equivalent standard lighting. The question is not whether the thread changed at all; the question is whether the thread now creates commercially unacceptable contrast shift, staining halo, or patchy appearance versus the approved control sample. If you want to lock that visually, state that fleece and thread are assessed separately.
Colourfastness and contrast-thread assessment
Contrast hems create two separate colourfastness questions: the fleece face and the sewing thread. They should not be merged into one vague line item. For the fleece, use a standard wash-fastness requirement such as ISO 105-C06 minimum grade 4 colour change and 3-4 staining on adjacent fabric, unless the channel accepts lower on very dark shades. For contrast risk, add ISO 105-X12 dry 4, wet 3-4 on the fleece face for saturated shades.
For the thread, most buyers do not run a separate laboratory grade on loose thread unless the contrast is critical. A practical contractual route is visual assessment against the approved control after the agreed wash cycles, under standard lighting, with no commercially obvious contamination of the fleece ground. If thread performance is high risk, specify a sewn-panel assessment and minimum pass/fail language at PPS.
Do not assume thread is exempt. If the buyer wants the contrast effect to remain clean, write that fleece and thread are reviewed separately: fleece by laboratory grade, thread by visual match and staining impact.
Separating size change from sewing distortion
A blanket can meet shrinkage tolerance and still look wrong because hemming pulled the panel out of square. Keep these as separate acceptance lines. Dimensional change tracks overall size loss after laundering. Skew, bow, and edge draw-in track panel distortion caused by cutting, feed balance, or hem tension.
Before wash, measure finished length and width on a conditioned blanket laid flat without tension. After the agreed wash route, recondition and remeasure for ISO 5077 dimensional change. Separately assess skew or bow as a percentage of panel dimension, and visually review whether the blanket lies flat without edge corkscrewing or rope-like waviness.
For value-retail fleece, a workable starting point is max 2.0% skew and max 2.0% bow. If your shelf presentation is folded, also check that opposite edges stack reasonably straight when folded in the approved retail pack orientation.
Defect classification examples for AQL inspection
AQL only works if defect mapping is agreed before inspection. The examples below are typical for retail fleece blankets; your retailer manual may override them. The point is to classify defects by customer impact, not by how annoying they were to sew.
PPS checklist: approve the result, not every machine setting
The PPS stage is where buyers should stop process drift without turning the PO into a sewing manual. Approve the visual result, measured hem geometry, wash route performance, and one sealed corner sample. Ask the supplier to record their machine setup internally, but do not contract every setting unless the style has a known failure history.
What to leave out of the PO unless there is a failure history
Some details belong in supplier SOPs, not in the commercial contract. Exact needle brand, lubrication level, perimeter sewing sequence, and whether the operation uses one overlap join or another acceptable start-stop method are process controls. They matter to the factory, but they do not always improve the buyer's position when written into the PO.
A good rule: if the buyer can inspect the outcome easily, write the outcome. If the buyer cannot verify the process detail without standing on the sewing floor, leave it as supplier-controlled unless you are solving a known repeat defect. This keeps the contract enforceable and leaves the factory room to optimise production without changing the finished standard.
Frequently asked
Should buyers specify exact needle size and thread brand in the PO? Usually no. Buyers should lock the finished requirement: hem geometry, SPI acceptance, appearance, wash durability, and thread fibre/ticket range such as continuous-filament polyester Tex 27-30. Exact needle size, point style, and thread brand are normally supplier process controls. Freeze them at PPS only if previous runs showed repeat defects such as skipped stitches, needle cuts, oil marks, or colour migration.
How should seam opening be specified for a fleece blanket hem? Use a measurable route. A practical starting point is a 100mm hem section loaded to 10N, then measure the maximum opening at the stitch line. Before wash, keep seam opening within 1.0mm. After 3 agreed ISO 6330 wash cycles, keep it within 3.0mm and require no stitch breakage, no raw-edge exposure, and no local seam failure. Avoid vague language such as 'light hand extension' or 'no seam opening over 5mm' without load and length.
Is 7-9 SPI enough to control a contrast lockstitch hem? No. SPI controls stitch density, but not by itself seam appearance or durability. Count over 25.4mm on straight runs only, take five readings per blanket, exclude corners, and accept on average 7-9 SPI with no single reading below 6.5 or above 9.5. Then add stitch appearance checks for skipped stitches, looping, and stitch breakage, plus a simple seam-opening test under load.
How should the wash route be written into the PO? Write the full route, not just ISO numbers. A workable model is ISO 6330 domestic front-load wash at 40C with standard reference detergent, cotton ballast to the required load, then one agreed drying method such as line dry or tumble low, for 3 cycles. Recondition for at least 24 hours and assess dimensional change to ISO 5077. If care labeling is part of the program, keep the test route aligned with the final care instruction set.
How do we assess contrast-thread colour change without over-specifying it? Separate fleece from thread. For fleece, set laboratory grades such as ISO 105-C06 minimum grade 4 for colour change and 3-4 staining, plus ISO 105-X12 dry 4 and wet 3-4 on dark shades where rubbing risk exists. For the contrast thread, many buyers use visual assessment after the agreed wash cycles under standard lighting against the approved control sample, with no commercially obvious contamination of the fleece ground. If the contrast effect is critical, add a sewn-panel PPS review or a separate thread test route.
What is the simplest way to control corner bulk reproducibly? Use compressed thickness at a fixed point rather than subjective folded-ply wording. A practical starting point is maximum 4.5mm compressed thickness measured 15mm from the corner point under 2kPa pressure, or no more than the approved PPS sealed sample plus 0.5mm. Also keep one corner construction only per order and check opposite corner projection difference within 5mm.
Have a project in mind? Send us your spec — we'll reply within one business day with indicative pricing and a sample plan.