Call(WhatsApp Us)
+8615360806100Email Us
cherry@laeyo.comSee the factory by live video walkthrough
Book an on-site factory visit in GuangzhouMost skincare brands audit the factory that fills the bottle — and never audit the two companies that actually make the bottle and print the label. That gap is where recalls, marketplace takedowns, and…
Most skincare brands audit the factory that fills the bottle — and never audit the two companies that actually make the bottle and print the label. That gap is where recalls, marketplace takedowns, and quiet quality failures start: a pump that doses 0.8 ml instead of 0.5 ml, a label that lifts off a PCR bottle after three weeks in a warehouse, or a printed net quantity statement that is technically present but too small to read. This guide gives you an audit structure you can run yourself, with the exact documents, tests, and acceptance criteria to request from your manufacturer and its subcontracted packaging and print suppliers. Informational only, not legal advice.
In a skincare supply chain, three different companies usually touch the same finished unit, and each one owns a different failure:
A standard factory audit reviews the third layer in depth and takes the first two on trust. That is a mistake, because the buyer — through its Responsible Person — carries the final responsibility for the label and the claims on it. The factory’s obligation is narrower but absolute: the formula, INCI, specifications, technical data, and safety documentation it supplies must be true and complete. Your audit should verify both sides of that line.
| Layer | What typically fails | What to request | Who owns the fix |
|---|---|---|---|
| Component | Neck finish out of tolerance, pump dose drift, tube orifice mismatch, surface treatment that repels adhesive | Drawing with tolerances, component COA, dimensional report, material declaration, drop/leak test | Component supplier, coordinated by manufacturer |
| Decoration / print | Color drift between lots, barcode grade failure, small-print illegibility, label lift, scuffing in transit | First-article sample, ΔE record, barcode verification report, adhesion test, artwork version log | Printer / decorator |
| Fill & assembly | Product–pack incompatibility, leakage, cap back-off, missing or unreadable lot code | Compatibility report, torque and dose records, pilot run report, line clearance and label reconciliation | Filling factory |
| Regulatory content | Wrong INCI order, missing net quantity, claims without substantiation, “FDA approved” style wording | Approved artwork with version number, claims file, technical inputs sheet, label review sign-off | Brand / Responsible Person (factory supplies inputs) |
Ask for the first-article sample and compare it line by line against the locked file — not against a PDF on a screen. Check the version number, the INCI order, the net quantity, the warning block, and the placement of variable data. Approve the sample on the filled pack: a label that looks correct on paper can lose contrast on a frosted or recycled surface.
Request the color standard and the measured color difference on the first-article print, taken under controlled lighting. Also ask for the barcode verification report; a scannable symbol is a functional requirement, not an aesthetic one. If the pack uses a soft-touch coating or a dark matte finish, request a scuff and rub test after the product has been in contact with the surface.
Product identity must be clear on the principal display panel, net quantity must be accurate and visible, and required warnings must be prominent. Do not accept “minimal design” as a reason to shrink legal text. Apply the same test to your evidence cues — the small line carrying “contains X% active” or “dermatologist tested.” View the pack artwork at roughly 400 px wide, or at real viewing distance: if the proof cue cannot be read, the substantiation you paid for delivers no value on shelf or in a thumbnail.
Confirm who prints the lot code, expiry, and period-after-opening symbol, and with which method. Inkjet on a curved, coated surface is a known legibility risk. Request a sample where the code survives a rub test and is machine-readable.
| Phase | Buyer action | Manufacturer action | Evidence output |
|---|---|---|---|
| 1. Specification lock | Confirm market, Responsible Person, and functional specs | Issue packaging spec sheet with drawings and tolerances | Signed specification, role matrix |
| 2. Component sourcing | Approve sub-tier supplier list | Order components, collect COAs and dimensional reports | Component COA, dimensional report |
| 3. Artwork and regulatory review | Review label content, claims, and legal text | Supply technical inputs, INCI, IFRA certificate | Approved artwork with version number |
| 4. Print first article | Approve sample on the filled pack | Print, measure color, verify barcode | First-article sample, ΔE record, barcode report |
| 5. Compatibility and stability | Review reports before pilot | Run compatibility, stability, and micro testing | Compatibility report, stability and micro results |
| 6. Pilot run | Attend or request full documentation | Fill, assemble, code, and reconcile labels | Pilot report, line clearance, reconciliation, reject photos |
| 7. Documentation handoff | Archive and sign off | Deliver final document pack | COA, SDS, batch record, retained samples |
Fastest safe path: compress artwork review and print first-article approval by running them in parallel with component sourcing, and use an existing proven pack where possible. Do not compress compatibility testing or the pilot run — those two phases are where expensive failures surface. Standard path: run every phase sequentially, add a second print first-article round for dark or metallic substrates, and hold two weeks of buffer before your retail or marketplace launch date.
| Symptom | Root cause | Prevention | Evidence to request |
|---|---|---|---|
| Labels lift at the edges after a few weeks | Adhesive not matched to a recycled or coated bottle surface | Test adhesion on the production surface before proof sign-off | Adhesion test report on filled packs after 72 hours |
| Visible leakage in transit | Pump or cap specified without testing against the real viscosity | Run dose, torque, and inverted leak tests on the actual formula | Leak and torque records, transit test report |
| Print color differs between lots | No measured color standard or uncontrolled ink batch | Approve a physical standard and require measurement per run | ΔE record under controlled lighting |
| Product listing blocked over label content | Missing net quantity, unclear product identity, or unsubstantiated claim | Route artwork through regulatory review from version one | Approved artwork, claims file, label review sign-off |
| Batch documents cannot be produced during an audit | Subcontractor records never consolidated into the finished-goods file | Make documentation handoff a contractual deliverable | Complete batch record with label reconciliation and sub-tier COAs |
| INCI on pack does not match the formula | Artwork updated separately from the formula change | Single change-control gate covering formula and artwork | Version log, formula-to-artwork cross-check |
| Lot code unreadable or missing | Variable data printed by an unnamed third party with no verification | Name the printer, define the method, verify legibility | Coding sample with rub test, reconciliation record |
| Barcode fails at retail scanning | Symbol printed below minimum grade on a curved surface | Verify grade on the printed pack, not the digital file | Barcode verification report |
SHARE the target market and Responsible Person details for each SKU.
PROVIDE the full sub-tier supplier list for bottles, closures, tubes, and labels.
CONFIRM which party prints lot code, expiry, and PAO, and by which method.
ATTACH the approved packaging spec sheet with drawing numbers and tolerances.
CONFIRM label substrate, adhesive type, caliper, and finish per SKU.
PROVIDE the color standard and measured color difference on the first-article print.
APPROVE the first-article label on the filled pack, not on paper.
TEST label adhesion on the production surface and report results after 72 hours.
TEST product-to-pack compatibility for four weeks at accelerated conditions.
TEST pump dose, cap torque, and leak performance on the filling line.
VERIFY barcode grade on the printed pack and share the verification report.
REVIEW the principal display panel for product identity, net quantity, and warning prominence.
REVIEW small-print evidence cues at 400 px width for legibility.
CONFIRM INCI order and fragrance declaration against the current formula.
CONFIRM the IFRA certificate matches the actual use category and current amendment.
LOCK the artwork version and record the version number in the batch record.
PROVIDE line clearance and label reconciliation records from the pilot run.
PROVIDE the pilot run report with defect categories and photographs of rejects.
ATTACH COA, SDS, stability, and micro reports for the pilot batch.
CONFIRM written change control covering any pack, print, or formula change.
CONFIRM document retention period and who retrieves records during an audit.
REVIEW the claims file so every on-pack claim maps to a test report.
You should require it contractually. Ask for a sub-tier supplier list per component and per print process, plus written notification before any substitution. Undisclosed supplier changes are one of the most common causes of a pack that suddenly behaves differently from the approved sample.
The brand, through its designated Responsible Person for the market, owns final label and claim compliance. The manufacturer owns the accuracy and completeness of the technical inputs it supplies: formula, INCI, specifications, COA and SDS, safety support, and facility information. Your audit should verify both sides.
No. Facility registration and product listing are mandatory compliance steps with renewal cycles; they are not an approval, certification, or quality endorsement, and they must not be used as a marketing claim. The Responsible Person still needs adequate safety substantiation records for each product.
Allow about two weeks for documentation review, two to three weeks for print first article and approval, and four weeks minimum for compatibility testing before the pilot run. Sequential planning with buffer is safer than compressing test phases.
Usually through your manufacturer rather than independently. Request the documentation package, a first-article sample, and where feasible a witnessed pilot run. If direct access is refused without reason, treat that as a risk signal rather than a negotiation point.
Send the checklist above to your current or prospective manufacturer and ask them to return the evidence pack against it. If you would like a quote built around your packaging direction, target market, SKU count, and timeline, Request a Quote — include your market, formats, positioning, and packaging direction for a faster response.