One drawing goes to three factories. The quotes come back at $2.46, $3.13 and $4.34 for a 25 cm character. A buyer looking at three numbers and one drawing concludes somebody is cheating.
Almost nobody is. The three figures are three different products that share a drawing. The difference is a list of specification choices, each of which can be priced - and the arithmetic below is the same model published in our unit-cost guide, so you can check it against your own rates.
The gap has four lines in it
Take the lean specification and our standard one and put them side by side. Same character, same 24 m² of cut panels, same trims list, same 12% margin.
| Cost line | Lean | Standard | Difference |
|---|---|---|---|
| Shell fabric | $0.385 | $0.544 | +$0.159 |
| Filling | $0.272 | $0.352 | +$0.080 |
| Embroidery | $0.308 | $0.396 | +$0.088 |
| Sewing labour | $0.643 | $0.911 | +$0.269 |
| Trims, packaging, overhead | $0.590 | $0.590 | unchanged |
| FOB per unit | $2.46 | $3.13 | +$0.67 |
That is a 21% gap on an identical drawing, and every cent of it is a specification decision rather than a negotiation. Push the specification the other way - 300 g/m² fabric, 165 g of fill, 5,500 stitches, 15 SAM - and the same character reaches $4.34, a 76% spread from the lean end.
FOB per unit, one drawing, three specifications
Lean $2.46 ####################
Standard $3.13 ##########################
Generous $4.34 ####################################
+----+----+----+----+----+----+----+
2.40 2.75 3.10 3.45 3.80 4.15 4.50
The five places the difference hides
A quotation that differs by 21% is not hiding it in the margin. It is in five places, and all five are visible if you ask.
Fabric weight. 220 g/m² against 260 g/m² is a 16% lighter shell. It drapes differently, it pills sooner, and it feels thin in the hand. Nobody prints GSM on a sample tag, so ask for the measured weight of the roll, not the nominal figure on the order.
Fill weight. 85 g against 110 g of fibre is a character that photographs the same and sags within a week of retail handling. The only honest number is grams per unit.
How the eyes are fixed. A push-fit eye and a washer-backed eye look identical on the face. Only one of them survives the 10 lb tension test of 16 CFR 1500.51(f)(1)(ii). The cost difference is cents; the failure mode is a choking hazard and a recall.
Print or embroidery. A four-colour chest print is cheaper than 600 stitches of embroidery - and it also converts an exempt dyed fabric into a component that needs lead and phthalate testing, because 16 CFR 1500.91(d)(7) exempts textiles only when everything on them is a dye.
Who inspects, and when. Final-visual-only is the cheapest QC line there is. A run with in-line control, AQL 2.5 pre-shipment sampling and 100% needle detection costs more per unit - and it is the line that finds the problem before the container leaves.
What a cheap quote usually leaves out
Cheap is rarely a lower price for the same work. It is usually a narrower scope.
| Item | In a lean quote | In a standard export quote |
|---|---|---|
| Third-party test reports (EN 71, ASTM F963, GB 6675) | Often excluded, “can be added” | Commissioned per component, evidenced |
| Needle and metal detection | Sometimes not mentioned | 100% pass, documented |
| In-line inspection | Usually not included | Included |
| Pre-shipment AQL inspection | Sometimes | AQL 2.5 to an agreed plan |
| Sampling rounds | One, revision charged | Included to approval |
| Carton and polybag specification | Buyer’s problem | Specified against the destination market |
| Social compliance audit (BSCI) | Not covered | Current report available |
None of those lines are optional to a retailer. They are optional only to the buyer who has not been told they exist.
The failures are not hypothetical
If cheap specifications only produced slightly thinner toys, this would be a comfort argument. The alert data says otherwise.
The European Commission published its 2025 Safety Gate annual report on 5 March 2026. The headline figures:
| Measure | 2025 |
|---|---|
| Alerts recorded | 4,671 - the highest since the system began in 2003 |
| Change against 2024 | +13%, and more than double the 2022 count |
| Follow-up measures taken by authorities | 5,794 - a record |
| Most common risk type | Hazardous chemicals - more than half of all notifications |
| Other significant risks | Injuries, and suffocation |
Toys were the second most notified category at 16%, behind cosmetics at 36% and ahead of electrical devices at 11%.
Share of all EU product safety alerts by category, 2025
Cosmetics 36% #####################
Toys 16% #########
Electrical 11% ######
All other 37% ######################
+----+----+----+----+----+
0 10 20 30 40 50 %
Two things follow from that table. Toys are a heavily policed category, so the probability of an unnoticed failure reaching the market is lower than it used to be - and chemical risk, the most common alert type overall, is exactly the part of a specification that a cheap quote tends to omit.
One failure costs more than the whole saving
Here is the arithmetic nobody puts on a quotation.
A 10,000 unit order. The lean specification saves $0.67 per unit.
| Line | Amount |
|---|---|
| Saving on 10,000 units at $0.67 | +$6,669 |
| Re-manufacture 2,000 replacement units at $3.13 | -$6,257 |
| Air freight for 2,000 units, 0.24 kg each at $7.50/kg | -$3,600 |
| Net position after one re-shipment | -$3,188 |
One air-freight replacement of a fifth of the order costs 1.48 times the entire saving. That is before a retailer chargeback, before a re-inspection fee, before the cost of a late delivery to a launch date, and before a recall - which is not a line item at all but an event that ends the relationship.
Saving from the lean specification vs the cost of one failure
Saving on 10,000 units +$6,669 ####################
Re-ship 2,000 by air -$9,857 ##############################
+----+----+----+----+----+----+
0 2 4 6 8 10 (k USD)
The point is not that cheap always fails. It is that the saving is bounded and small, while the loss is unbounded - and it lands in a single week.
What our specification actually includes
This is where we are, stated plainly so you can compare it with the other quotes on your desk.
| Our published parameters | Value |
|---|---|
| Workshop | 12,000+ m², pattern room, embroidery, needle detection and QC under one roof - nothing brokered out |
| Production staff | 200+ |
| Monthly output | 100,000 pcs |
| Sampling lead time | 5 to 7 days, revisions included to approval |
| Minimum order | 100 pcs per design |
| Inspection regime | 100% needle detection, AQL 2.5 pre-shipment, in-line control |
| Quality management | ISO 9001; BSCI social audit |
| Export experience | 18+ years |
That specification is why our number is usually the higher one on the sheet, by roughly 20% to 25%. We are not able to be the cheapest quote for a 25 cm plush and still run 100% needle detection, AQL 2.5 and the component testing. The choice is structural, not a negotiating position.
What a buyer gets for that difference is not a promise. It is a specification you can check line by line against the lean column in the first table of this article.
Five questions that price a quote for you
You do not need to trust either side. Ask these five, in writing, and the numbers become comparable.
| Ask | What it settles |
|---|---|
| Measured GSM of the shell fabric | Turns a material name into a weight you can price |
| Fill weight in grams per unit | The only defence against an under-filled delivery |
| How the eyes and nose are fixed, and to what standard | Predicts whether the toy survives the tension test |
| Print or embroidery, and the exact ink or thread | Predicts both the look and the chemical test scope |
| Inspection plan: in-line, final, AQL level, needle detection | Turns “we have QC” into a number |
A supplier who answers those five in a day is telling you they know their line. A supplier who answers with adjectives is telling you something too.
What we will not do
- We will not match a price we cannot build to. If a quote is 40% below ours on the same drawing, it is a different toy, and we would rather lose the order than ship one.
- We will not quote without the five numbers. A number without a specification is not a price.
- We will not bury the test cost in a line called “miscellaneous”. Test reports are itemised, because a buyer should see what compliance costs.
Where this stops
This article is about specification differences, not about accusing anyone. There are badly run expensive factories and well-run cheap ones, and the only way to tell them apart is the five numbers above - not the price.
Two honest caveats. The cost figures here are a model with illustrative rates, not a quotation; the arithmetic holds when you substitute your own. And a higher price is only justified if the specification behind it is genuinely higher - which is why we publish ours.
References
-
European Commission, Safety Gate: the EU rapid alert system for dangerous non-food products - results for 2025, annual report published 5 March 2026. Summarised by the Czech Ministry of Industry and Trade, Systém rychlého varování EU pro nebezpečné nepotravinářské výrobky - výsledky za rok 2025, 6 March 2026. https://mpo.gov.cz/cz/ochrana-spotrebitele/bezpecnost-vyrobku/system-vcasneho-varovani/system-rychleho-varovani-eu-pro-nebezpecne-nepotravinarske-vyrobky-_-vysledky-za-rok-2025—291954/
-
16 CFR § 1500.51, Test methods for simulating use and abuse of toys and other articles intended for use by children 18 months of age or less - the 10 lb seam and projection tension requirement referred to above. https://www.ecfr.gov/current/title-16/section-1500.51
-
16 CFR § 1500.91, Determinations regarding lead content for certain materials or products under section 101 of the CPSIA - paragraph (d)(7), the textile exemption that a print or coating removes. https://www.ecfr.gov/current/title-16/section-1500.91
-
16 CFR part 1303 - Ban of lead-containing paint and certain consumer products bearing lead-containing paint. https://www.ecfr.gov/current/title-16/part-1303
-
16 CFR part 1307 - Prohibition of children’s toys and child care articles containing specified phthalates. https://www.ecfr.gov/current/title-16/part-1307
-
Flechazo, What a Plush Toy’s FOB Price Is Made Of: A Unit-Cost Model You Can Audit - the worked model and rates used throughout this article. https://flechazoplush.com/blog/plush-fob-unit-cost-model/
-
ASTM F963 - Standard Consumer Safety Specification for Toy Safety, and EN 71-1 - Safety of toys - Part 1: Mechanical and physical properties. The voluntary and harmonised standards a retailer will ask to see evidence against.
Related reading
- What a plush toy’s FOB price is made of: a unit-cost model you can audit
- Lead and phthalates in a plush toy: which components actually need testing
- Age grading a plush toy: the small parts rules behind a 0+ or 3+ label
- Black-faced lamb plush: quoted at 26 cm and 35 cm, which is where a size difference shows up as a real cost line rather than a percentage
- The stock catalogue: the input data a comparable quotation needs
Flechazo quotes FOB Shanghai/Ningbo, CIF and DDP, and states the fabric GSM, fill weight, eye fixing, stitch count and inspection plan on the quotation so two suppliers can be compared rather than guessed at. MOQ from 100 pieces per design.