Filling, Capping and Pack-Out: Where Cosmetic Quality Is Won

Short answer: the formula decides how a product feels; the filling line decides whether the carton arrives looking like the sample you approved. Fill accuracy, capping torque and seal integrity are the three measurable outputs of pack-out, and they are also the three causes of most post-arrival complaints — leakage, underfill findings and pumps that will not prime after a cold crossing. All three are specified before production and verified in the batch record.

Buyers spend their development attention on the bulk — texture, scent, actives, absorption — and then judge the result on the unit. A batch that meets every bulk specification can still generate a claim if the jars weep in transit, if the declared net content is challenged at retail, or if the sprayer dribbles. Those are pack-out failures: cheap to prevent at the filling set-up, expensive to fix after eight pallets are loaded.

Why pack-out decides complaints

Two numbers sit behind every unit on a shelf: what the label says is inside, and what is actually inside. The gap between them is where money and regulatory risk both live.

• Overfill is a real cost that never appears on an invoice: every extra gram across a large run is paid for at raw-material price, plus container volume you did not sell.
• Underfill is not a cosmetic problem but a declared-quantity problem. Net-content systems work on the average of a sample plus a tolerable negative error rather than on every unit, so a line centred slightly low can be compliant at the mean while repeatedly failing at the bottom of the range — and a badly underfilled jar eventually reads as a defective pack.

The filling target is therefore a decision, not a setting: it has to sit deliberately above the declared quantity, by an amount that matches the machine's real variation rather than its brochure figure.

Four filling principles and what each suits

Different formats fail in different ways, and the filling principle is chosen around viscosity, accuracy requirement and pack geometry.

Two formats sit outside that table. Airless and vacuum filling fills from the bottom of the pack, because an air pocket inside an airless bottle defeats the system's premise. Hot filling keeps the bulk mobile until it sets in the pack, trading fill accuracy for melt-point discipline.

The principle also constrains packaging choice: some formats cannot be filled accurately on the machines a factory has, so the pack is best chosen with the line in mind — the trade-offs between pump, airless, tube and jar are in our packaging format comparison.

The controls that keep a run centred

Frequency of checks matters more than any single measurement. A control plan for a jar line that samples once at start-up certifies nothing about hour six.

• Fill checks at start-up, mid-run and end-of-run, with the sample size stated in the batch record
• Tare and density verification for the specific bulk, because a volumetric setting is only correct for one viscosity at one temperature
• Product temperature at fill, recorded rather than assumed
• Line clearance and reconciliation: containers issued, filled, rejected and returned have to sum to the issued quantity
• Reject handling: failed units quarantined with a reason, and the count in the record

That last point is where the record earns its keep: a batch record showing a rejected-unit count and a reason is evidence of a controlled process, while zero rejects across a long run usually means the check was not performed.

Capping, torque and closure integrity

Capping is the least glamorous step and the most common cause of leakage. The specification that matters is an application torque window — a minimum that seals the closure, and a maximum that keeps it removable and undistorted.

• Torque outside the window. Below it the liner does not compress and the pack weeps; above it the closure cracks, the liner deforms permanently, or the cap will not come off.
• Cross-threading, which can seal visually and fail under vibration, usually tracing back to neck-finish tolerance in the mould rather than the capping head.
• Liner compatibility with fragrance, essential oils or high alcohol content, all of which will find a weak liner.
• Pump and sprayer priming. An airless pump needs its actuator primed and the pack de-aerated; a trigger sprayer needs its dip tube cut to reach the base without curling.
• Vacuum fluctuation in transit, which can draw product past a marginal seal on a route through high ground or extreme cold.

All of it is verifiable before shipping: torque checks at a defined frequency, a leak test and a transport simulation — reasonable questions for any supplier, alongside the checks in our manufacturer verification checklist.

Test the pack, not only the batch

Release testing answers whether the bulk is correct. Pack-out testing answers a different question: whether the unit survives being sold, shipped and used.

• Leak and seal integrity — vacuum or pressure decay testing, dye penetration, and hot-fill seal checks where the format calls for them.
• Transport simulation — drop, vibration and compression tests that mirror the actual route, on a carton in a full pallet rather than a single unit.
• Cold and heat exposure — at the temperatures the shipment will meet, checking for separation, gelling, sweating and pump function.
• Component compatibility — product in contact with the pack for the intended shelf life, watching for colour shift, fragrance uptake, swelling, stress cracking and print transfer.
• Consumer-realistic dispensing — a pump tested for its full dose count, not for one actuation.

Compatibility is a study rather than a spot check: it runs for the length of the shelf life, which is why pack choice and shelf life are decided together, and why it belongs on the same calendar as the stability programme rather than in a separate project.

Failure modes buyers see at the port

The same handful of defects accounts for most post-arrival claims, and each one points back to a specification rather than to luck.

• Product in the carton, or a wet pallet base — application torque below the window, or a liner incompatible with the bulk.
• Jars that look underfilled against the approved photo — a fill target centred too close to the declared quantity for the machine's real variation.
• A pump that will not prime after a cold crossing — an air pocket left by filling; the pack was never de-aerated.
• Loose caps or cracked closures on arrival — torque above the window, or neck-finish tolerance in the mould.
• Colour shift or scent loss at month six — component–product interaction, or a liner extracting fragrance.
• Lifting labels or transferred ink — adhesive and finish not matched to the pack surface and the climate it travels through.

The pattern is consistent: pack-out defects surface as complaints, not as rejections, because they appear after the goods have travelled. That is why the batch record and the pack-out checks are worth more than a final visual inspection — and why the documents that travel with the order, described in CoA and batch documents, should reconcile with the carton codes instead of arriving as a standalone PDF.

What this changes in a quote and a schedule

Pack-out capability is one reason two quotes for the same formula differ. Cheap filling usually means a least-cost machine, a wide fill tolerance and a capping station without instrumented torque — all invisible in a unit price. Three questions separate the quotes:

1. What fill tolerance is quoted, in grams, for my bulk at my viscosity — and how is it verified in the record?
2. Is capping torque measured and recorded, and at what interval? If the answer is "by hand feel", the leakage risk is yours.
3. Which pack-out tests are included, and which are charged? Leak, transport simulation and compatibility studies sit inside or outside the unit cost, and the difference shows up in the per-unit cost structure rather than on the quotation line.

Pack-out sits at the end of the timeline — within the 10–15 day mass-production window, depending on the manufacturing system — but the decisions that make it go well are made when the pack is chosen and the fill target is set. Our production timeline breakdown shows where it sits relative to bulk manufacture and release.

FAQ

Q: What fill tolerance should I expect on a cosmetic run? A: It depends on the bulk viscosity, the container and the filling principle — a flow-meter filling a thin serum holds a tighter window than volumetric filling of a thick cream. What matters is that the tolerance is quoted in grams for your bulk and verified at defined intervals in the record, rather than assumed from a machine's rated accuracy.

Q: Why do products leak in transit when they passed inspection? A: Leakage is usually a closure problem rather than a bulk problem: application torque below the sealing window, a liner incompatible with the formula, or a mould tolerance that only fails under vibration. A leak test plus a transport simulation on the packed carton catches most of it.

Q: How is declared net content checked? A: Net-content systems assess the average of a sample together with a tolerable negative error, rather than requiring every unit to reach the declared quantity. That is why the fill target is centred above the declaration to match the line's real variation, and why a run can be compliant on average while producing underfill findings.

Q: Does the packaging choice affect how accurately the product can be filled? A: Yes. Airless and vacuum formats fill from the bottom to avoid trapped air, stick formats are hot-filled and rely on melt-point discipline, and some packs cannot be filled accurately on a given factory's machines — which is why the pack is chosen with the line in mind.

Q: What pack-out checks should appear in the batch record? A: Fill checks at start-up, mid-run and end-of-run with the sample size stated, tare and density verification, product temperature at fill, capping torque checks at a defined frequency, leak or seal results, and a reconciliation of containers issued against units filled and rejected.