The bottle comes back and the colour is wrong. The factory says your Pantone card is old; you say their mixing is off. Both can be true, and usually the truth is split across three places: the card, the screen you compared it on, and the plastic itself. Knowing which one moved saves you from paying for a re-run that fixes the wrong thing.
A Pantone number is a condition, not a fixed paint
Pantone defines a colour under specific lighting and substrate. The same number on coated versus uncoated paper already looks different, and on plastic it shifts again. So ‘off from Pantone’ isn’t a single deviation — it’s a comparison between your reference and the moulded part under your light. Shijin Packaging, a cosmetic plastic bottle manufacturer in China since 2003, reads colour against a signed golden sample with a spectrophotometer, because the number alone doesn’t tell you what ‘right’ means on a bottle.
Your screen is not the card
The colour you approved on a monitor is rendered by your display’s profile, not by ink or resin. Two screens show the same hex differently, and a phone in a shop will show it differently again. If the only reference anyone saw was a PDF or a webpage, then ‘off’ may have started on the screen, long before the factory mixed anything. A physical chip beats every digital stand-in, because only it is the same object the spectrophotometer will later measure.
The card itself drifts between print runs
Pantone books are printed, and print fades and shifts with age, light and handling. A card that’s been on a bench for years can read a half-step from a fresh one. So when a factory says ‘your card is old,’ they may be technically right — not as an excuse, but as a real source of the gap. The fix is to agree on one current, controlled reference before production, not to argue about which old card wins.
The plastic changes the read
Resin grade, wall thickness and even fillers move the colour. The same masterbatch in PET versus a different polymer, or in a thick versus thin wall, lands at a different ΔE. This is why a colour matched on a plaque or a rival’s bottle won’t automatically transfer to yours. The substrate has to be part of the spec, or the factory is matching a target on the wrong material and the bottle you receive was never the real reference.
How to stop the blame game before it starts
Before mass production, lock one reference: a current physical chip or a moulded sample in your own hand, measured under a standard light source. Write the acceptable ΔE into the PO. When everyone is comparing the same object under the same light, ‘factory or card’ stops being a guessing game and becomes a measurement — and measurements don’t argue, they just report a number against the signed sample.
What to do when it’s actually the factory
Sometimes it is the mixing: the masterbatch ratio drifted, or the cap and body were moulded on different machines and read differently. That’s a process fault, and it’s on the supplier to correct within the agreed tolerance at their cost. “+FULL+” That kind of documented process is what lets a colour dispute resolve from the sample, not from opinion — if the production part diverges from the golden sample beyond the stated ΔE, the re-run is their responsibility, not a new charge to you.
The sign-off that ends the argument
The golden sample is the tie-breaker. Approve it under standard light, keep it, and measure every batch against it. If you never signed one, you’ve handed the factory the right to match ‘Pantone’ by their own eye — and then ‘off’ is everyone’s fault and no one’s. Sign the sample, and the question of factory-versus-card answers itself from the part in front of you, with a number both sides agreed to in advance.
See how these specs show up on real parts: 220 ml PET cosmetic bottles, 300 ml PET cosmetic bottles.
Frequently asked questions
Is a Pantone colour on plastic exactly the same as the swatch?
No, and it shouldn’t be expected to be. Pantone is defined under specific lighting and substrate; the same number on coated paper, uncoated paper and moulded plastic reads differently because the material and surface change the light. What you’re really buying is a consistent match to an agreed reference — a signed golden sample — within a stated ΔE, not a perfect copy of the paper swatch.
Can an old Pantone card cause a colour mismatch?
Yes. Pantone books are printed and they fade, yellow and shift with age, light and handling, so a card that’s been on a bench for years can read a half-step from a fresh one. That doesn’t automatically excuse a factory, but it is a real source of the gap. The cure is to agree on one current, controlled physical reference before production rather than debating which old card is right.
Why does the same colour look different on cap and body?
Cap and body are often moulded on different machines, sometimes in different resins or wall thicknesses, and each shifts the read slightly. If they’re matched from one masterbatch but moulded separately, a small ΔE between them is normal. The PO should state the tolerance for cap and body separately, and the golden sample should include both pieces signed as a set so the factory can’t tune one and ignore the other.
How do I prove the factory got the colour wrong?
With a signed golden sample and a spectrophotometer reading. If the production part diverges beyond the ΔE written into your PO, that’s a process fault and the re-run is the supplier’s cost. Without a signed sample, you have no fixed reference — the factory matched ‘Pantone’ by eye and ‘off’ becomes a dispute with no tie-breaker. Sign the sample first; proof follows from the part.
What should I write into the purchase order for colour?
Name the exact Pantone, state the reference is a current physical chip (not a screen), specify the acceptable ΔE against the approved golden sample, and require cap and body to be measured separately. Add that the golden sample is signed under standard light and kept as the production tie-breaker. A PO line like that turns ‘factory or card’ from an argument into a measurement anyone can check against the signed part.