TL;DR: Switching a cosmetics brand’s secondary packaging from offset-printed labels to direct pad printing reduced per-unit decoration cost by 31% and cut SKU changeover time from 4.2 hours to 47 minutes across a 14-SKU product line.
TL;DR: In this project, ink adhesion on the curved PP compact bodies required a pre-treatment dyne level of 44 mN/m — below 40 mN/m, field peel tests failed within 6 weeks of shelf exposure.
Surface Energy, Substrate Geometry, and Why This Project Almost Failed in Week Two #
When a mid-size Australian skincare brand came to us with a 14-SKU compact and bottle range, the brief looked manageable on paper: direct decoration on PP and ABS substrates, two-colour logo plus batch code, 80,000 units per run. The structural forms included flat-face compacts, slightly domed cap tops, and a narrow cylindrical lip gloss tube — three distinct geometries, two materials, one project timeline.
Our standard pre-qualification flow (what we call the QP-14 substrate clearance check) flags any substrate arriving without a confirmed surface energy reading. PP, in particular, varies significantly by grade and any added release agents from the moulding process. The first incoming lot from the brand’s existing mould supplier tested at 34 mN/m, well below our 40 mN/m minimum threshold for reliable pad print adhesion. We sent those components back and specified re-flame treatment to reach 44 mN/m before re-inspection.
That single step delayed the sampling phase by nine working days — but avoided what would have been a complete adhesion failure across the first production run.
The pad print parameters we landed on after substrate qualification:
| Substrate / Geometry | Cliché Depth | Ink Type | Post-Print Cure | Adhesion (Cross-Cut, ISO 2409) |
|---|---|---|---|---|
| PP flat compact face | 28 µm | 2-part polyurethane | 60°C / 30 min forced air | 0 (no loss) |
| ABS domed cap | 22 µm | Single-part acrylic | UV, 200 mJ/cm² | 0 (no loss) |
| PP cylindrical tube | 25 µm | 2-part polyurethane | 60°C / 30 min forced air | 0 (no loss) |
The cliché depth differences matter here. The domed ABS cap had shallower relief because the pad compression arc was longer — too deep a well on a curved transfer and you get ink shear at the trailing edge, producing a ragged second colour boundary. We confirmed this during press trials: at 28 µm on the dome, edge definition dropped from our target ±0.15 mm register to roughly ±0.35 mm, visible under normal inspection lighting. Dropping to 22 µm and adjusting pad shore hardness to 12° Shore A brought that back within spec.
What Actually Goes Wrong on Multi-Geometry Projects — and the Sequence Matters #
The brand’s original brief assumed a single pad setup could handle all three geometries in rotation. That assumption drove three separate failure scenarios during our pre-production trials, and working through them in the wrong order would have cost an extra two to three weeks.
The first failure involved ink pick-up consistency on the cylindrical tube. A cylindrical surface means the pad contacts the substrate at a tangential angle rather than a flat compression. The pad we initially specified — a 75° round pad — left a consistent thin band of missing ink at the contact initiation point on roughly 1 in 8 impressions. We diagnosed this as pad geometry mismatch, not ink viscosity. Switching to an asymmetric flat-top pad with a 60° profile resolved the drop-out. The lesson: substrate geometry determines pad profile selection before ink specification even enters the conversation. Pad profile is not a consumable decision — it is a tooling decision.
The second failure was a colour shift between production shift changeovers on the two-part polyurethane ink. Two-part PU inks have a working pot life of 6–8 hours at 22–24°C. During our first extended run, the ambient temperature in the printing hall peaked at 29°C, dropping pot life to approximately 4 hours. The ink batch mixed at the start of a 10-hour shift was printing noticeably more saturated in the final two hours as viscosity climbed. We measured a delta-E of 4.2 between early and late impressions — against a brand tolerance of delta-E ≤ 2.0 per the brand’s colour standard (aligned to Pantone Matching System spot verification). The corrective action was a 2-hour ink refresh cycle on PU inks whenever ambient temperature exceeded 26°C, logged under our production form IP-22 ink interval control.
The third failure was more subtle: slight ghosting on the batch code print in the second colour pass. The root cause was insufficient cure between passes. Our UV cure unit was delivering 200 mJ/cm² per pass, but conveyor speed had been increased by a technician to improve throughput without re-measuring actual dose at speed. A UV radiometer check showed effective dose had dropped to 140 mJ/cm² — below the 180 mJ/cm² minimum for tack-free surface state between passes. We now lock conveyor speed via our press setup sheet and record radiometer readings at the start of every production run.
Does Direct Decoration Always Beat Labelling for This Category? #
For rigid moulded compact and closure formats above 30,000 units per SKU, direct pad printing typically becomes the more cost-efficient route — but the crossover point depends heavily on how many colours are involved and whether the substrate is label-receptive.
On this project, the 14-SKU range had an average run of 5,700 units per SKU, which is well below the classic break-even threshold. The economics still favoured direct printing because the brand was already paying a label applicator and managing label stock for 14 references — indirect costs that rarely appear in a per-unit label price quote. Once we factored in label inventory management, applicator line time, and the recurring label minimum order quantities, the total decoration cost per unit dropped from USD 0.089 under the label model to USD 0.062 under direct pad print — a 31% reduction that held across the full 80,000-unit programme.
This calculus changes for substrates with heavy texturing or for formats under 20mm diameter where pad registration becomes unpredictable. For those, self-adhesive labels with a matching matte or metalite finish remain the practical choice.
Specification Notes for Brand Partners #
When you brief us on a direct pad printing project for cosmetics secondary packaging, the most critical information we need upfront is substrate material and grade (not just “plastic”), confirmed surface energy readings or moulding process details, and physical samples of each geometry — ideally three to five pieces per SKU.
A gap we encounter frequently: brands provide render files or CAD drawings but not actual moulded parts. Pad profile selection and cliché geometry both depend on the physical substrate, and even a 0.5mm change in dome radius changes the pad compression arc enough to require a different tooling setup. A render tells us what the part looks like. A physical part tells us how it prints.
The common brief gap that costs the most time: no colour standard. If you need delta-E ≤ 2.0 colour consistency, we need a confirmed Pantone reference or a physical colour standard before sampling begins, not after first samples are reviewed.
Our standard sampling timeline for a multi-geometry pad print project is 18–22 working days from confirmed substrate receipt and approved artwork. Projects with pre-treatment requirements (flame or plasma, which we handle in-house for runs above 30,000 units) add 3–5 working days to that window.
Frequently Asked Questions #
What surface energy level does our PP substrate need to reach before pad printing?
Our minimum is 40 mN/m, and for two-part polyurethane inks on compact faces we specify 44 mN/m as the target — below that, adhesion failure becomes likely within weeks of normal handling and shelf exposure.
Can the same pad setup run multiple substrate geometries in one shift?
It depends on how different the geometries are. Flat and shallow-dome forms can often share a pad type with a shore hardness adjustment. Flat-face and cylindrical forms in the same session require different pad profiles entirely — attempting to run both on a single setup introduces consistent print defects at the contact initiation point, which is a tooling problem rather than an ink or speed problem.
How do you control colour consistency on long runs using two-part polyurethane inks?
We set a 2-hour ink refresh interval whenever ambient temperature exceeds 26°C, measured and logged on production form IP-22. Two-part PU inks have a working pot life of 6–8 hours at 22–24°C, and that window compresses significantly in warmer conditions. A delta-E reading at the start and end of each ink batch is the quickest way to confirm consistency before it becomes a sorting problem.
What is a realistic per-unit cost saving when switching from labels to direct pad printing?
On this project it was 31%, bringing cost per unit from USD 0.089 to USD 0.062 across an 80,000-unit programme. That figure included indirect label costs that are often excluded from initial comparisons. The saving narrows on very short runs or on heavily textured substrates where label application is actually more reliable.
Does ISO 2409 cross-cut adhesion testing tell you everything you need to know about ink durability?
Cross-cut to ISO 2409 is our standard incoming acceptance criterion, and a Grade 0 result is required before we release samples. For cosmetics packaging specifically, we also run a simulated shelf-exposure soak test at 40°C/75% RH for 96 hours, followed by a repeat cross-cut and a finger-rub abrasion cycle. The ISO test alone does not replicate the combined effect of humidity cycling and product contact that compact or closure surfaces experience in distribution and retail.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
