Define the SKU correctly before you compare FOB points
For sherpa throws, 320gsm should be written as a finished composite article specification, not left as a loose fabric claim. Some mills use GSM for the face fabric only; others use it for the complete two-layer article after bonding or sewing. For buying control, state: finished size with tolerance; finished composite mass per square metre; finished piece weight tolerance; fibre content; face and reverse construction; edge finish; fold method; pack method; and carton pack. A tighter example is: 150x200cm finished size ±3cm; finished composite weight 320gsm ±4%; 100% polyester; flannel face 190±10gsm; sherpa reverse 130±10gsm; 3-thread overlock edge or 10-12mm turned hem as approved; one-piece polybag; 10pcs/carton.
Keep finished composite GSM, finished piece weight, and finished dimensions as separate acceptance points. They are linked but not interchangeable. A 150x200cm throw has an area of 3.00m², so a nominal 320gsm article gives a target textile mass of about 960g before polybag, insert, belly band, carton, and tape. In production, trims, sewing thread consumption, hem construction, brushing loss, and normal moisture regain can move the complete piece weight by several grams to a few tens of grams. If the PO only says 320gsm, suppliers can still deliver noticeably different warmth and handfeel if size, seam build, or component split drifts.
Use a method, a sample count, and conditioning language. A practical buyer clause is: condition samples for at least 24 hours in a standard textile atmosphere of approximately 20±2°C and 65±4% RH; measure finished dimensions on 10 pcs per lot after relaxation without tension; weigh 10 complete finished pieces excluding outer carton but including standard sewn-in labels and agreed retail band if shipped that way; and verify component areal density on pre-sewing fabric or approved in-process panels using a GSM cutter method. Where wash performance is claimed, test dimensional change to ISO 6330 home laundering protocols and write the wash cycle in the PO, not just the standard name.
If you need adjacent construction guidance, 300gsm sherpa-to-coral fleece blanket programs and fleece-weight throw blanket programs are useful references for how mills separate component weights, finished article mass, and pack assumptions. The key point is simple: define the article first, then compare ports. Otherwise the freight model is built on a moving SKU.
Write FOB accurately under Incoterms 2020 and know when FCA is cleaner
Under FOB Incoterms 2020, the seller clears the goods for export and delivers them on board the vessel nominated by the buyer at the named port of shipment. Risk transfers when the goods are on board. That legal structure was drafted for sea or inland waterway shipment, but for modern containerized exports it is often awkward because the seller usually hands cargo to a terminal or carrier before actual vessel loading and may not control or even witness the on-board event.
For many container shipments, FCA is the cleaner term. Under FCA, the seller delivers the goods to the carrier or another party nominated by the buyer at the named place, which can be the factory, a forwarder warehouse, or a container yard. That better matches real container operations where the seller can document handover at CY gate-in or at the forwarder's receiving point. If your program is containerized sherpa throws moving through a buyer-nominated forwarder, FCA named place often reduces documentary ambiguity around the exact transfer point.
Many buyers still use FOB operationally because internal systems, landed-cost templates, and carrier contracts are built around FOB China port naming. That is common, but buyers should understand the mismatch: if you insist on FOB Shanghai or FOB Xiamen for container cargo, the seller may control trucking, customs declaration, VGM filing, and CY cut-off delivery, yet not control the actual loading on board. If documents, delay claims, or damage claims arise, that gap can create argument over whether the goods were truly delivered under the legal term or merely handed into the export chain.
A practical sourcing rule is: use FCA when your forwarder takes custody before vessel loading and you want the handover point documented cleanly; use FOB only if your organisation accepts the container-shipment convention and can manage the risk/document mismatch. If the PO remains FOB, add an operating note: seller responsible for export clearance, pre-carriage, and timely CY handover to meet booked cut-off; buyer responsible for carrier booking and main carriage; any dispute over late handover judged against forwarder receiving records and cut-off times, not a vague 'port delivery' phrase. For adjacent cost-bucket logic, see EXW vs FOB cost items for blanket tenders and custom blanket lead times and shipping.
Use loadability maths with stated assumptions, not round numbers
A buyer-usable model needs explicit assumptions. For a 40HQ, internal geometric cube is typically around 76m³, but sherpa throws should not be planned at full geometric fill. For this SKU, a reasonable planning basis is 68-72m³ usable volume for floor-loaded, non-palletized export cartons. That range reflects door-end loss, imperfect carton interlock, bowing at the container sidewall, carton compression limits, and the need to avoid over-stressing vacuum-packed sherpa so badly that recovery, seam appearance, or carton integrity fails on arrival. The range is a planning assumption based on pack tests and practical loading discipline, not a universal constant.
State carton orientation and carton build. Example assumption set: 10 pcs/carton; cartons floor-loaded with the 60x45cm face down and 30-40cm height vertical; no pallets; no slip sheets; no hanging void fill; manual or clamp-assisted warehouse loading; export carton B/C flute around 5-7mm; stacking not to exceed compression level that leaves cartons deformed after 24 hours. If you palletize, usable cube drops sharply and the same FOB comparison changes. Do not mix floor-load maths with palletized dispatch rules.
Now show the arithmetic. At 150x200cm and 320gsm composite, nominal textile mass is 0.96kg/pc. If 10 pcs are packed in a standard carton at 60x45x40cm, carton volume is 0.108m³. At 68-72m³ usable cube, that yields about 629-666 cartons, equal to 6,290-6,660 pcs/40HQ. If controlled vacuum packing reduces the carton to 60x45x32cm, volume becomes 0.0864m³, giving about 787-833 cartons or 7,870-8,330 pcs. At 60x45x30cm, volume is 0.081m³, giving about 839-888 cartons or 8,390-8,880 pcs. Those figures are plausible for planning, but they should be validated by an actual pack test and carton compression check on the approved article.
For this program the container is usually cube-limited rather than payload-limited. A 10-piece carton contains roughly 9.6kg blanket mass, and complete carton gross weight may land around 10.2-11.5kg depending on polybags, inserts, carton tare, tape, labels, and actual article weight. That stays below common manual-handling caps and far below normal 40HQ payload limits. Freight per piece therefore moves more from carton cube than from net fabric mass. If your DC sets a gross-weight ceiling, write it into the PO, for example 16kg, 18kg, or 20kg maximum per carton depending on retailer policy.
Convert inland differences to per-piece effect. If one port costs US$450 more per 40HQ inland than the other, the delta is roughly US$0.072/pc at 6,290 pcs, US$0.057/pc at 7,870 pcs, and US$0.051/pc at 8,880 pcs. If the inland delta widens to US$900 during a tight trucking period, double those numbers. Buyers should ask for this arithmetic in the quote file, not a verbal 'small difference'. For adjacent compression logic, vacuum-compressed blanket CBM reduction and CIF costing is a useful comparison, even though mink and sherpa recover differently.
Compare two real operating patterns: Fujian via Xiamen versus Delta consolidation via Shanghai
Pattern 1: Fujian pack-out, FOB Xiamen. This is usually cleaner when cutting, sewing, inspection, vacuuming, carton packing, and export documentation all sit in Fujian. Typical buyer benefit is shorter drayage from factory or pack-out warehouse to port, fewer cross-province handoffs, and less carton abrasion or vacuum rebound before gate-in. If the cargo is truly local to Xiamen operations, the inland saving can be material and the chain is simpler to monitor. This pattern is strongest when your nominated forwarder already has stable allocations ex-Xiamen on the target lane and your booking does not rely on switching between multiple weekly mainline departures.
Pattern 2: Zhejiang or Jiangsu supply chain, FOB Shanghai. This usually wins when knitting, dyeing, cutting, sewing, final QC, or consolidation happen in the Yangtze River Delta. Shanghai typically offers denser sailing frequency, more carrier options, more forwarder-managed receiving points, and easier consolidation of mixed SKUs from Zhejiang/Jiangsu suppliers. For buyers with replenishment calendars, retail launch dates, or frequent LCL-to-FCL programme building, that wider schedule matrix often matters more than a modest inland saving elsewhere. A missed weekly sailing can cost more than a few cents per piece if the fallback becomes air top-up, markdown exposure, or launch slippage.
The port comparison should be made against origin cluster and lane type, not abstract port reputation. Xiamen often fits Fujian-origin cut-and-sew, especially when the throw program is self-contained and shipped as one full-container lot. Shanghai often fits Delta-origin programmes, mixed-SKU consolidation, and lanes where buyers want more backup sailings. In practical terms, if knitting is in Zhejiang, sewing in Jiangsu, and final pack-out near Shanghai, trucking finished sherpa throws to Xiamen can add cost, handling, and schedule exposure without compensating benefit. If the whole chain is in Fujian, the reverse is often true.
On container exports, cut-off reliability and rollover risk matter more than brochure frequency. Ask your forwarder for the last 8-12 weeks' operating view on your lane: CY cut-off tightness, booking release timing, rollover incidence, and whether backup sailings are realistically bookable or only theoretically listed. Shanghai often has broader fallback options on major US/EU lanes; Xiamen can still perform well if your carrier mix is aligned and you book early enough. During peak season, however, the port with more mainline choices and stronger forwarder allocation usually gives the buyer better recovery options after a production slip.
Customs and inspection friction should also be judged at factory-region level. Goods moving from local Fujian production through Xiamen may see fewer process touches than goods trucked long distance for a different named FOB port. Conversely, Delta factories and forwarders often have deeply routinised export files through Shanghai. The correct question is not 'which port is easier' in general; it is 'which port is easier for this exact factory, this exact document set, this exact month, and this exact forwarder'.
Use a simple decision rule: if inland savings are small and schedule reliability differs clearly, choose the more reliable booking port. If inland savings are large and sailing options are broadly comparable, choose the closer port. If the seller proposes a named FOB port that does not match the actual manufacturing chain, ask why and request a full inland and schedule justification. For background on mixed-material outdoor packing choices, choosing picnic, beach, and camping mat constructions is a useful contrast in how origin cluster drives export logic even when the article type differs.
Side-by-side TCO framework buyers can actually score
Use a total-cost-to-load view rather than only the quoted FOB price. A usable matrix for 320gsm sherpa throws is: 30% weight on pack-out location fit; 20% on booking control and fallback sailings; 15% on inland cost delta per 40HQ; 15% on cut-off reliability and rollover risk; 10% on customs/document friction; and 10% on QC hold risk before gate-in. Score each item 1-5 for Xiamen and Shanghai using your forwarder, your factory region, and your shipping month. A port with a slightly higher inland cost can still win if it protects the launch date and reduces rehandling.
A buyer-ready comparison table can be written directly into the sourcing pack: FOB Xiamen—best when pack-out is in Fujian; inland usually lower for Fujian-origin cargo; fewer unnecessary handoffs; risk if lane has thinner backup sailings or weaker buyer allocation. FOB Shanghai—best when supply chain is in Zhejiang/Jiangsu or mixed-SKU consolidation is needed; inland often higher for Fujian-origin cargo; stronger mainline frequency on many lanes; usually better fallback if first booking slips. Add your own lane-specific freight note rather than assuming one port is always cheaper on ocean rate.
For internal approval, build one row each for: inland trucking and local charges; expected buffer days from ex-factory to CY cut-off; document flow complexity; forwarder control points; estimated pieces per container based on approved carton; and risk note. Example shorthand: Xiamen may show lower inland cost and one fewer handoff for Fujian pack-out, but weaker recovery if rolled. Shanghai may show higher inland cost from Fujian yet stronger booking flexibility and easier mixed-factory consolidation. The right answer often changes by month.
Red flags that should push the port choice back for review are: seller asking to quote FOB from a non-local port without inland breakdown; no approved compression pack test before booking; carton cube still estimated rather than measured; named forwarder unable to confirm allocations on the intended lane; and supplier resistance to a written cost-sharing rule if the named FOB port is changed after PO issue.
Sherpa-specific QC: what to check before you let cartons move
If the article is vacuum-packed or compression-packed, the QC plan must be sherpa-specific. Generic fleece checks are not enough. Inspect pile crushing and recovery by opening packed samples after a defined dwell time, for example 24-72 hours after compression, then allowing recovery for 2 hours and 24 hours before visual and handfeel grading. Record whether the sherpa side remains matted, whether the flannel face shows panel marks, and whether edge curl increases after rebound. If recovery is poor, the theoretical cube gain is not commercially usable.
Check shade and texture balance between the flannel face and sherpa reverse. Even on solid colours, the two faces can read differently under D65 and warm store lighting. Approve a standard under at least one controlled light source and check lot-to-lot consistency against it. On piece-dyed polyester constructions, ask for rubbing and wash-fastness standards separately where dark shades are used; related methods include ISO 105-C06 wash-fastness checks and ISO 105-X12 rubbing-fastness checks for adjacent fleece articles.
Inspect seam grin, hem security, and needle damage. Sherpa bulk can hide skipped stitches or low seam margin until the article is stretched. Check edge seam allowance, overlock bite, thread tension balance, and any grin at corners after light extension. Look for shiny tracks, cut loops, or localized pile loss from blunt or overheated needles. If the article uses turned hems instead of overlock, verify bulk at corners and flatness after packing.
Control fiber shedding, loose lint, and odor. New sherpa can release loose pile after brushing and cutting; excessive shedding in the polybag is a retail complaint trigger. A practical inspection is shake test plus black-board or dark-garment visual check on a defined number of samples. Odor should be checked both immediately after opening and after short airing, since vacuum conditions can trap finishing or bag odor. Flag strong chemical smell, sour moisture smell, or solvent-like notes for hold.
Run a carton integrity and recovery check on the actual export pack. Measure carton dimensions, gross weight, strap or tape integrity, and rebound after a basic drop sequence suitable for your programme, such as one corner, three edges, and six faces from an agreed height if retailer specs require it. Then re-open and grade article recovery again. A carton that ships 5-8% more pieces but causes permanent matting, split bags, or deformed corners is false economy.
Use a lot release structure that links article quality to shipment timing: in-line checks during sewing; pre-compression approval; post-compression recovery check; final AQL inspection; and loading release only after carton dimensions and counts match the booking plan. For general inspection framework, blanket quality-control inspection, AQL 2.5 inspection checklists, and anti-pilling test requirements are useful companion references.
Acceptance methods, tolerances, and test language buyers should stop leaving vague
Replace vague wording such as 'composite GSM checked on cut sample or agreed area method' with a named procedure. A practical clause is: from each lot, cut five test specimens from approved in-process fabric using a 100cm² GSM cutter, avoiding selvage and visibly distorted areas; condition for at least 24 hours at approximately 20±2°C and 65±4% RH; weigh to 0.01g; convert to g/m²; and compare average result against the approved component target. For a finished two-layer article, do not try to derive composite GSM by cutting through bulky hems and seams. Instead, verify component areal density before assembly and verify complete article piece weight and size after assembly.
For piece-weight acceptance, write the article-level method separately: sample 10 finished pieces per production lot; condition for at least 24 hours in standard atmosphere; weigh each complete article including sewn labels and standard unit pack if shipped that way, excluding outer carton; average must fall within the agreed tolerance and no individual piece may exceed an agreed outer band, for example nominal ±6% individual and nominal ±4% average. The exact band should follow the risk you can tolerate on warmth, freight, and retail consistency.
For dimensions, write: measure 10 finished pieces after 24-hour relaxation on a flat table without stretching; length and width each within ±3cm unless otherwise approved; no more than one piece outside tolerance in the sample set unless reinspection is triggered. If laundering performance is part of the claim, specify the wash protocol and maximum dimensional change after laundering, for example ISO 6330 domestic laundering with dimensional change assessed after one or multiple cycles depending on claim language.
If appearance retention matters, add performance thresholds rather than relying only on incoming visual judgment. Common buyer controls for polyester fleece and sherpa include pilling to an agreed grade after a stated method, seam strength where applicable, colourfastness to washing, and odour acceptability by retained standard. If your market requires care label precision, align the sewn-in instructions to ISO 3758 care labeling. The point is to stop treating blanket acceptance as one soft visual call.
Paste-ready RFQ and PO clauses
Port and Incoterm clause: 'Trade term: FOB [named port] Incoterms 2020 only if Seller delivers against Buyer-nominated booking and export clearance for containerized shipment. Buyer may require FCA [named CY/forwarder warehouse/factory] Incoterms 2020 where carrier handover occurs before vessel loading. Seller shall not switch the named FOB/FCA place without Buyer's written approval.'
Cost-change clause: 'If Seller requests a change from the named FOB port after PO issuance, Seller shall disclose the reason in writing and all incremental costs, including but not limited to inland trucking, cross-dock or warehouse handling, storage, booking amendment, customs re-filing, document amendment, detention caused by delay, and any higher local charges. Unless Buyer approves otherwise in writing, Seller bears all incremental costs and schedule impact caused by Seller-requested port change. If Buyer requests the port change, Buyer bears approved incremental costs.'
SKU specification clause: 'Article: sherpa throw blanket, finished size 150x200cm ±3cm, finished composite weight 320gsm ±4% average lot tolerance, 100% polyester, flannel face 190±10gsm, sherpa reverse 130±10gsm, approved handfeel and shade per sealed standard, edge finish per approved pre-production sample, unit pack one piece per polybag, export carton pack 10 pieces unless otherwise approved. Finished article nominal textile mass reference approximately 960g/pc at specified size and gsm.'
Packing and loadability clause: 'Supplier shall submit pre-shipment pack test with measured folded size, carton dimensions, carton gross weight, and compression dwell/recovery results on approved bulk article. Booking quantity per 40HQ shall be based on measured export carton dimensions and agreed usable loading cube assumption for floor-loaded, non-palletized cargo. No shipment may use a tighter compression method than the approved pack test without Buyer's written approval.'
QC and inspection clause: 'Inspection level: AQL 2.5 major / 4.0 minor unless otherwise agreed, based on final packed goods. QC checkpoints shall include pile recovery after compression, shade consistency between flannel face and sherpa reverse, seam grin, loose fiber shedding, odor, needle damage, carton dimension verification, and carton integrity after agreed handling test. Buyer or Buyer's representative may place shipment on hold for recovery failure, odor failure, excessive shedding, or pack-out dimensions differing from approved test.'
Booking and cut-off clause: 'Seller shall provide cargo readiness date, carton count, measured CBM, and VGM-supporting weight information no later than [X] days before CY cut-off. If Seller misses agreed cargo-ready timing and rollover or amendment charges result, responsibility shall be allocated based on documented cause and forwarder records.'
Frequently asked
Is FOB or FCA better for containerized sherpa throw exports from China? For containerized cargo, FCA is often legally cleaner because delivery can be defined at the factory, forwarder warehouse, or CY where the seller actually hands over cargo. FOB is still used widely in practice, especially in buyer systems built around FOB China ports, but it can create a mismatch because the seller may not control actual on-board loading. If you keep FOB, add detailed operating clauses for cut-off, handover records, and port changes.
How many 150x200cm 320gsm sherpa throws fit in a 40HQ? It depends mainly on folded pack size and whether cartons are floor-loaded without pallets. With 10pcs/carton, a 60x45x40cm carton yields roughly 6,290-6,660 pcs at a 68-72m3 usable loading assumption. A 60x45x32cm carton yields roughly 7,870-8,330 pcs. A 60x45x30cm carton can reach roughly 8,390-8,880 pcs, but only if compression does not damage pile recovery, seams, or carton integrity.
When does FOB Xiamen usually make more sense than FOB Shanghai? FOB Xiamen usually fits best when cutting, sewing, inspection, and final pack-out are in Fujian and the buyer's forwarder has reliable allocations ex-Xiamen. The shorter inland leg can reduce cost and handling. If the supply chain sits in Zhejiang or Jiangsu, or if mixed-SKU consolidation and frequent fallback sailings matter, FOB Shanghai is often operationally stronger even if inland cost is somewhat higher.
What should buyers check on vacuum-packed sherpa throws before approving shipment? Check pile crushing and recovery after a defined compression dwell period, shade consistency between the flannel and sherpa faces, seam grin at edges and corners, needle damage, loose fiber shedding inside the polybag, odor on opening, carton integrity, and post-drop recovery if your retailer requires handling tests. A pack method that improves piece count but causes permanent matting or odor complaints is not a valid saving.
How should 320gsm be written in a PO for a sherpa throw? Write it as a finished composite article spec, not just a fabric claim. State finished size, finished composite gsm tolerance, component split such as flannel face 190±10gsm and sherpa reverse 130±10gsm, fiber content, edge finish, unit pack, carton pack, and article-level piece-weight method. This prevents disagreement over whether 320gsm refers to one face fabric or the whole blanket.
Have a project in mind? Send us your spec — we'll reply within one business day with indicative pricing and a sample plan.