FLECHAZO
Prototyping & Sampling Peer-Reviewed by Factory QA Lab 8x (Fill volume when linear size doubles - while the seam allowance stays exactly the same)

Scaling a Plush Design: Why 20 cm and 40 cm Are Different Products

A buyer wanting the same character in three sizes assumes a scale factor. Surface area grows with the square of the size, fill volume with the cube, seam allowance and hardware do not scale at all, and pile length stays where the mill made it - so a doubled size needs a re-graded pattern, a different fabric and a different fill target. Here is what actually changes across a size ladder.

M

Marcus Vance

Director of Factory Operations & Sourcing Quality Control

7 min read

Executive Summary & Core Benchmarks

  • ✓ Doubling a design's linear size quadruples the fabric area and multiplies fill volume by eight, but the seam allowance stays a constant 6 to 8 mm, so the small version of a design is proportionally a different object.
  • ✓ Pile length does not scale. A 5 mm pile reads as shaggy on a 10 cm toy and as a short, flat nap on a 45 cm one, so a size ladder usually needs two fabric specifications rather than one.
  • ✓ Facial features have a floor and a ceiling: eyes must clear the small-parts rules at the bottom of the ladder and remain legible when embroidered at the top, which is why a 8 cm and a 45 cm version are separate pattern jobs.
  • ✓ Cost is not linear in size. Fabric rises with area, fill with volume and freight with packed volume, so the FOB of a 45 cm style is typically far more than twice the 20 cm version rather than a markup on it.
  • ✓ MOQ applies per size per colourway. Three sizes of one design is three production runs, not one, which is the part of a size ladder that changes the commercial arithmetic.

A size ladder is one of the first things a brand asks for: the same character at a keyring size, a shelf size and a display size. It sounds like a resizing job. It is closer to three separate products that happen to share an identity.

The reason is arithmetic. A plush toy is not a flat drawing that can be enlarged, and the numbers that govern it do not move together.

What scales, and what refuses to

Doubling a design’s linear size is an eight-fold change in volume and a four-fold change in surface, while several of the most important dimensions stay exactly where they were.

PropertyHow it moves when linear size doublesConsequence
Fabric area per unitx4 (square of the scale)Fabric is the largest BOM line, so this dominates cost
Fill volume per unitx8 (cube of the scale)Stuffing time, fill weight and carton weight all follow
Seam length per unitx2 (linear)Sewing time rises, but not in step with the fabric
Seam allowancex1 — unchanged6-8 mm is a constant, so it eats a much larger share of a small panel
Eye, nose, zip, ringx1 — cannot scaleHardware is bought in fixed sizes from a catalogue
Pile lengthx1 — set by the millA 5 mm pile looks shaggy at 10 cm and flat at 45 cm
One design, three sizes, not to scale

  8 cm          25 cm                 45 cm

  area    1x  ---------->  9.8x  ---------->  31.6x
  volume  1x  ----------> 30.5x  ----------> 178x
  seam    1x  ---------->  3.1x  ---------->   5.6x
  allowance   6 mm   (identical at every size)
  pile        5 mm   (identical at every size)

  Ratio to the 8 cm toy. The gap between the area and the volume lines is
  the reason a bigger toy feels like a different material proposition.

The seam allowance problem

Every sewn panel carries a seam allowance, commonly 6 to 8 mm, consumed inside the seam and never seen. On a 45 cm body panel it is a rounding error. On an 8 cm leg it is a substantial fraction of the piece, and the pattern maker is fighting for millimetres of curvature over a seam that cannot get narrower.

This is why small toys are not scaled-down big toys. Below roughly 12 cm the pattern has to be redrawn: panels get removed, curves get simplified, and tighter radii are replaced with shapes a needle can actually negotiate. A straightforward scale operation produces a small toy whose head is subtly the wrong shape and whose limbs are too thin to stuff.

The pile problem

Pile length is a property of the fabric roll, not of the design. A 5 mm pile on a 10 cm keychain covers roughly a twentieth of the toy’s height; the same 5 mm on a 45 cm display piece covers a fraction of a percent and reads as a flat, cheap shell.

The practical result is that a size ladder normally needs two fabric specifications, not one:

  • Small sizes (8-15 cm): flat velboa or a short-pile plush, so the pile does not swamp the face and embroidery stays legible.
  • Large sizes (35-45 cm): a longer pile or a mink, so the toy reads as soft at arm’s length rather than looking like printed cloth.

Using one fabric across the ladder is possible, and it is the single most common reason a display-size version of a design looks wrong next to its keychain.

Faces do not scale nicely either

Facial features sit between a legal floor and a practical ceiling.

At the bottom of the ladder, eyes are the constraint. Plastic safety eyes below a certain diameter are hard to fix reliably and, on a toy marketed under 3 years, become a small-parts question. Most small plush therefore uses embroidered eyes, which changes the face entirely compared with a plastic-eye 45 cm version.

At the top of the ladder, embroidery is the constraint. A stitch file authored for a 25 cm face cannot simply be enlarged: stitch density, underlay and the minimum stitch length all have a working range, and scaling up produces long, loose stitches that read as a different, coarser face. A large version needs its own stitch file, not a scaled one.

So a ladder is genuinely two pattern jobs and often three stitch files, not one design resized.

Cost is not linear, and it is not flat either

Because the cost drivers move at different rates, the price of a size ladder does not follow a straight line.

Cost lineBehaviour across the ladderWhy
FabricRises with area (x4 on doubling)Cutting and consumption are area-driven
FillRises with volume (x8 on doubling)Grams of fibre, and stuffing machine time
LabourRises faster than linear, slower than areaSewing tracks seam length; stuffing tracks volume
HardwareFlat per unitThe same eye, zip or ring fits a range of sizes
SamplingCharged per sizeEach size is a separate pattern and sample
TestingOften per size or per component setA 45 cm toy is not covered by the 20 cm test report

The consequence for a quotation is simple and frequently missed: the 45 cm version is not “the 20 cm price plus a bit”. It carries four times the fabric, eight times the fill and its own sample, and if it is a separate size it is a separate production run with its own minimum order.

MOQ applies per size, per colourway

This is where a size ladder becomes a commercial decision rather than a design one.

A three-size ladder in two colourways is not one order of 300 pieces. It is six runs of 100, each with its own setup, its own cutting ticket and its own inspection lot. The setup charges that look trivial against a 20,000 piece run are material at 100 pieces, which is exactly the arithmetic in a unit-cost model.

If the volume for a size is genuinely small, the honest options are usually to launch the hero size first and add sizes on the second order, rather than to spread one minimum order across six runs and get a weak first shipment of each.

Packing and freight move faster than the price

Packed volume scales roughly with the cube, and it is what the carton maths uses. A 45 cm plush occupies far more container space per unit than a 25 cm one, and the ratio is worse than the price ratio - so the landed cost gap between sizes is larger than the FOB gap.

Small sizes are the opposite: a keychain at 8 cm travels at a density no large plush can match, which is why keychains are the one category where ocean freight per unit is close to irrelevant. The mechanics are set out in plush freight and CBM.

What to put in the size brief

A brief that gets an accurate quotation on the first pass answers five things:

  1. Which size is the hero SKU - the one the programme is really built around. It decides which pattern gets the most attention.
  2. Whether the sizes share a pattern or are re-graded. Say it explicitly. “Scaled from the 25 cm” and “designed at each size” are different jobs at different prices.
  3. Finished-size tolerance. Plush is sewn from soft goods and is never dimensionally exact; a stated tolerance of ±5% is realistic, ±2% is not.
  4. Fill firmness intent per size. A display piece and a keychain usually want different densities even from the same shell.
  5. The colourway count per size, because that is what multiplies the MOQ.

The catalogue already carries three designs in two sizes each, which is enough to read the arithmetic against real records rather than a formula:

Each product page states the size, the fabric and the MOQ, and the custom plush service covers a new ladder from artwork. If the intent is a size that is not in the catalogue, the sampling process is the same 5 to 7 days per size - which is the other reason to launch one size first and prove it before committing to a ladder.

M

Marcus Vance

Director of Factory Operations & Sourcing Quality Control

Writes the test scope for each export order against EN 71, ASTM F963 and GB 6675, and coordinates third-party testing with accredited laboratories before shipment. Focus areas: mechanical stress testing, electromagnetic metal detection, and destination-market import compliance.