I have worked with hundreds of plush brand owners over fifteen years, and the single biggest predictor of whether a product launch succeeds or fails is the discipline with which the brand handles the sample approval phase. Brands that move slowly through sampling, document every revision, and only sign off when the sample is genuinely right tend to launch products that match expectations. Brands that rush through sampling tend to receive 5,000 units that “almost match” the sample — and almost is what kills retail launches.
Sample approval is not glamorous. It is paperwork-heavy, requires patience, and often involves three or four rounds of back-and-forth across continents and time zones. But this phase is where the entire economic value of the product gets locked in. A poorly approved sample becomes thousands of off-spec units. A well-approved sample becomes a consistent product line that retail buyers re-order season after season.
This article documents the exact 6-stage sample approval workflow we use at our Dongguan factory for custom plush toys destined for USA retail. The framework applies to any plush manufacturer, though the names of the stages vary by factory.
What “Sample Approval” Actually Covers
Before walking through the stages, let me clarify what sample approval is actually doing. It is not just “do I like how it looks.” Sample approval is the process of establishing three independent things:
First, the design intent is correctly translated from your brief into a physical product. This is the obvious part — does the bear actually look like your reference image? Are the proportions right? Does the face match the character design?
Second, the production specification is locked in writing. This means every measurable parameter — fabric type, stuffing density, accessory dimensions, embroidery thread color and stitch count — is documented in a spec sheet that both you and the factory have signed. The spec sheet becomes the legal contract for mass production.
Third, the golden reference is physically established. Both you and the factory hold an identical, sealed sample unit that serves as the master reference for the entire production run. When mass production samples are pulled for QC, they are compared against this golden sample, not against a photograph or a memory.
Get all three right and your mass production order will closely match your sample. Get any one wrong and you are setting up the kind of dispute that ends factory relationships.
Stage 1: Design Brief and Initial Pattern Development
The sampling workflow begins before any physical sample exists. The first stage is design brief submission — where you give your factory enough information to start building a pattern.
A complete plush toy design brief includes: the character or product reference (sketches, photos, or 3D renders); the target dimensions (height, width, depth in centimeters); the desired pose or stance (sitting, standing, lying); the fabric type and color expectations; the stuffing density and material; any specific construction details (jointed limbs, embroidery placement, accessory attachments); the safety standards the final product must pass; the target retail price point; and the target production quantity.
The factory takes this brief and produces a pattern. The pattern is a flat set of fabric pieces, cut from paper or thin cardboard, that when sewn together produces the 3D plush shape. Pattern development typically takes 3-7 working days for a standard plush product, longer for complex designs with multiple posable joints or unusual proportions.
Most factories will share the pattern with you for review before cutting fabric for the first sample. This is your last chance to catch fundamental proportion issues — once fabric is cut, changes become expensive. Review the pattern carefully, particularly checking that limb-to-body ratios match your reference, head size relative to body is correct, and the seam placement matches your design intent.
Stage 2: First Prototype Sample
After pattern approval, the factory produces the first prototype sample. This is the “rough draft” of your product — it answers the question “does the basic construction approach work?”
Expect the first prototype to be imperfect. Common issues at this stage include: face features positioned slightly off-center, seam tension creating subtle asymmetries, stuffing distribution causing unintended shape variations, fabric pile direction creating visual oddities, and color slightly off from your target swatch. Most of these are normal and correctable.
The first prototype typically arrives 7-10 days after pattern approval. Production cost is absorbed by the factory if you are working with established quote terms, or charged as a “sample fee” of $50-200 per design if you are dealing with a new factory.
Your job at this stage is to provide detailed, specific feedback in writing. Vague feedback (“it doesn’t look right”) will produce another vague sample. Specific feedback (“the right eye is 4mm too high relative to the left eye; the snout protrudes 8mm but should protrude 12mm per the reference; the ear stitching should curve inward rather than going straight”) will produce a corrected sample.
I recommend taking photographs of the sample alongside your reference image, marking up the photographs with annotations, and sending those alongside written notes. The combination of visual and written feedback eliminates the language and cultural translation issues that plague sample feedback rounds.
Stage 3: Revised Prototype Samples
Most plush products go through 2-4 revision rounds before reaching pre-production sample status. Each round addresses specific feedback from the previous sample.
Round 2 typically focuses on structural corrections — fixing proportions, repositioning features, adjusting seam tension, correcting fabric direction issues. These are pattern-level changes that require recutting fabric. Round 2 sample arrival is typically 7-10 days after Round 1 feedback is submitted.
Round 3 typically focuses on detail refinements — embroidery thread shade matching, accessory dimensions, stuffing density adjustments, color consistency between fabric batches. These are smaller changes that do not require pattern recuts. Round 3 sample arrival is typically 5-7 days after Round 2 feedback.
Round 4, when needed, typically focuses on production-readiness verification — confirming that the construction approach can be replicated at scale, that materials are available in sufficient quantity, and that labor time per unit is within reasonable bounds. Round 4 is more of an internal factory verification than a buyer-facing approval.
Through these rounds, communication discipline matters more than speed. A brand that takes 2 days to respond with comprehensive, specific feedback gets a better result than a brand that responds in 30 minutes with vague impressions. Buyers who insist on rushing samples often complain later that “the factory didn’t understand what I wanted” — when in reality the factory was given insufficient information to know what was wanted.
Stage 4: Pre-Production Sample
Once the revised prototype meets all design criteria, the factory produces a pre-production sample — sometimes abbreviated as “PPS.” This is the sample that gets formally signed off as the standard for mass production.
The pre-production sample differs from prototype samples in three important ways. First, it is built using the actual production materials — the exact fabric batch, stuffing lot, accessories, and packaging that will be used for the mass order. Earlier prototype samples may have used substitute materials for cost reasons. Second, it is built using actual production labor methods — the same workers, sewing equipment, and assembly sequence that will be used at scale. Third, it includes full packaging — the retail box, polybag, hangtag, or whatever the final consumer-facing packaging will be.
The pre-production sample is your last opportunity to catch issues before committing to mass production. Inspect it thoroughly. Compare to your reference. Confirm dimensions with a measuring tape. Test the packaging closure. If safety eyes are used, verify they pass a pull test (250 newtons of force per ASTM standards). If pellet pouches are used, verify the seams hold under twist testing.
If everything passes, you formally sign off on the pre-production sample. This is where the specification sheet gets locked in writing — every measurable parameter of the approved sample is documented, signed by both parties, and becomes the legal reference for mass production.
Stage 5: Golden Sample Designation
After pre-production sample sign-off, two identical units are produced and sealed as golden samples. One goes to you; one stays with the factory. These golden samples are the official reference for the entire mass production run.
The golden sample protocol matters because it eliminates “he said, she said” disputes during mass production. When a QC inspector pulls a random sample from the production line and questions its quality, the comparison is against the golden sample — not against a photograph, not against a written description, not against memory.
Golden samples should be sealed (typically in a transparent plastic bag with tamper-evident seals) and labeled with: the product name and SKU; the approval date; signatures of the approving brand representative and factory production manager; and the order number(s) the golden sample applies to. Both copies of the golden sample should be photographed from multiple angles, and the photographs stored in your project documentation.
I have seen brands skip the golden sample step thinking it was overkill. In every single case, those brands later had QC disputes that took weeks to resolve because there was no agreed reference standard. Skipping golden sample protocol is one of the most expensive shortcuts you can take.
Stage 6: Pre-Shipment Sample (PPS Verification)
The final stage of the sample approval workflow happens after mass production begins. Once the first 10-15% of the production order is complete, a pre-shipment sample is pulled from the production line and sent to you for verification.
This sample answers a single question: is the actual production output matching the golden sample standard? It is not a redesign opportunity — that ship has sailed. The pre-shipment sample either confirms production quality matches the approved standard, or flags a problem that needs to be corrected before the order ships.
Common issues caught at pre-shipment sample stage include: fabric pile direction inconsistency between batches, embroidery thread color variation between rolls, stuffing density drift as factory workers fatigue, accessory attachment variation as new workers join the production line, and packaging finish defects from print run variations.
If issues are found, you have three options. Option one is acceptance with discount — you accept the shipment as-is in exchange for a per-unit price reduction. Option two is rework — the factory corrects the issues at their cost before shipping. Option three is rejection — you refuse the shipment and require new production at the factory’s cost. Option three is rare and indicates a serious quality failure.
For most issues, option two is the right move. Reworking is cheaper than rejecting, and the factory generally agrees if the issues are documented against the golden sample standard.
Documents You Must Have Signed Before Mass Production Starts
The sample approval workflow produces three documents that should be signed by both parties before any mass production fabric is cut:
The Pattern Approval Document confirms that the flat pattern (and therefore the basic shape and proportions) has been reviewed and accepted. This is signed at the end of Stage 1.
The Specification Sheet is the master document listing every measurable parameter of the approved product: fabric type and weight, color codes (Pantone references), stuffing material and density, accessory dimensions and materials, embroidery thread colors and stitch counts, packaging configuration, label placement, and dimensional tolerances. The specification sheet is signed at the end of Stage 4 (pre-production sample approval).
The Golden Sample Designation is the formal record that the sealed reference samples have been created and distributed. It includes photographs of the golden sample from multiple angles, the seal tag details, and the storage location of each copy. Signed at the end of Stage 5.
Together, these three documents form the contract for what mass production should produce. Without them, you have no enforceable standard if production output disappoints.
Common Sample Approval Mistakes Brand Owners Make
Eight mistakes come up over and over in sample approval, each preventable:
Treating sample sign-off as optional documentation. Some brand owners view paperwork as the factory’s responsibility and just want to “see the bear.” Without signed sample approval documents, you have no legal recourse if production output differs from the sample. Insist on signing.
Approving samples based on photos only. Photos do not show stuffing density, do not reveal fabric hand-feel, do not display proper color (camera and screen color shifts), and do not let you test physical safety features. Always require physical samples before final sign-off, even at the cost of an extra week of shipping time.
Approving under time pressure. “The factory needs sign-off today to start production on time” is the most common pressure tactic. Resist it. A two-week delay in mass production is much cheaper than 5,000 wrong units. If the timeline is genuinely tight, the right move is to add resources, not to lower the approval standard.
Skipping pattern review. Many brands let the factory skip past pattern review to first prototype. Pattern issues are 10x cheaper to fix at the pattern stage than at the prototype stage.
Not documenting feedback. Verbal feedback over WhatsApp does not survive translation, does not get archived, and cannot be referenced if disputes arise. Always submit feedback in writing with photographs and specific measurements.
Not retaining golden samples properly. A golden sample in a desk drawer collecting dust is still useful. A golden sample lost in a move, donated by accident, or destroyed by a leaky roof leaves you with no reference standard. Store golden samples in sealed boxes in climate-controlled storage.
Letting different team members approve different samples. Approval authority should sit with one person (or one tightly aligned small team). If your design lead approves Sample 2 and your production manager approves Sample 4 with different priorities, you end up with inconsistent decisions and a confused factory.
Confusing sample approval with QC. Sample approval establishes the standard. QC enforces the standard. They are different functions, done by different people in different stages. Brands that try to combine them end up with neither.
How Long Should Sample Approval Take?
For a straightforward plush product with no major design challenges, the full 6-stage workflow typically takes 6-10 weeks from initial brief to pre-shipment sample sign-off. Here is a rough breakdown:
Pattern development takes 3-7 days. First prototype production and shipping to you takes another 10-14 days. Your first review and feedback round takes 3-5 days. Revised prototype rounds 2-3 each take 10-14 days including shipping. Pre-production sample takes 10-14 days. Golden sample sealing and distribution takes 3-5 days. Mass production begins, and pre-shipment sample arrives 14-21 days after that.
Faster timelines are possible — 4 weeks total is achievable for simple plush keychains with experienced factory relationships. Slower timelines are also common — 12-16 weeks for complex licensed character plush with multiple revision rounds.
The biggest variable is not the factory speed; it is your team’s response time to feedback rounds. Brands that respond within 24 hours of each sample arriving move through the process roughly 30-40% faster than brands that take 4-5 business days to respond.
When to Walk Away from a Factory During Sampling
Some factories should be fired during sampling rather than continued into mass production. Watch for these warning signs:
The factory ignores written feedback and produces revised samples with the same issues. This is usually a communication or capability problem that will only worsen at scale.
The factory cannot produce sample documentation (no spec sheet, no signed approvals) and resists when asked. This factory will not be able to enforce quality standards during mass production either.
The factory shifts deadlines repeatedly during sampling. If they cannot manage 1-2 sample units on schedule, they cannot manage 5,000 units on schedule.
The factory blames you for sample issues rather than taking responsibility. (“You did not specify that, so we cannot fix it.”) A good factory takes ownership of execution within the spec; a bad factory uses spec gaps as excuses.
The factory pressures you to skip steps in the approval workflow. (“We can move to mass production now and fix any issues in production.”) This factory wants to lock you into payment commitments before quality is locked in.
If you see two or more of these warning signs during sampling, walk away. The sample cost is sunk; the production order at stake is not.
Frequently Asked Questions
How many sample rounds should a plush toy go through?
Most custom plush toy products go through 3-4 sample rounds before reaching pre-production sample status. Round 1 is the first prototype (rough). Round 2 fixes structural and proportion issues. Round 3 refines details (color, embroidery, accessories). Round 4 verifies production-readiness. Simple plush designs may finish in 2 rounds; complex character plush may need 5-6 rounds.
How much does plush toy sampling cost?
For an established factory relationship, sample costs are often absorbed into the production order pricing and not charged separately. For new factory relationships, sample fees range from $50 to $250 per prototype, depending on complexity. The fees usually cover materials and labor for one sample unit; revision rounds may incur additional fees if the design changes significantly.
What is the difference between a prototype sample and a pre-production sample?
A prototype sample is a development version — built to test the design and confirm visual intent. It may use substitute materials or shortcut construction methods. A pre-production sample is built using the actual production materials, methods, and packaging that will be used at scale. The pre-production sample is the version you formally sign off for mass production.
What is a golden sample?
A golden sample is a sealed reference unit that represents the agreed standard for mass production. Two identical golden samples are produced — one held by the brand, one held by the factory. All production output is compared against the golden sample for quality verification. Golden samples are sealed with tamper-evident packaging and signed approval tags.
Can I skip the sampling process to save time?
No, not safely. Skipping or rushing samples is the single most common cause of mass production disasters in plush manufacturing. The cost of an extra 2-3 weeks for proper sampling is trivial compared to the cost of 5,000 wrong units that cannot be sold at retail. Even simple plush products should go through at least one prototype and one pre-production sample.
What should I do if I do not like the first sample?
Provide specific, written, photographed feedback. Mark up photos with annotations showing exactly what needs to change. Reference your original design brief and explain how the sample diverges from it. Be patient — the first sample is supposed to be imperfect. The revision round system is designed to converge on the right product over 2-4 iterations.
Who pays for sample revisions?
For revisions caused by factory production errors (sloppy execution of the spec), the factory pays. For revisions caused by buyer design changes (you decided you want a different ear shape), the buyer pays. Most factories are flexible on this distinction in established relationships and absorb minor revision costs. New factory relationships often require buyers to pay for any revision beyond the first one.