Stability Testing Requirements by Market

Stability testing is the one part of a product file that cannot be hurried by paying more. A brand can buy faster microbiology, a larger claim panel or a fuller metals screen, & all three arrive sooner if someone is willing to spend more. Stability runs at the speed of the conditions it is studying, so the only real lever is when the study starts.

That timing decision is where cross-border sellers lose the most. A product developed for one climate is shipped through another, stored in a third & sold online into a fourth, & the shelf life printed on the pack has to survive all of them. Brands that start stability work after the artwork is approved end up printing a period they cannot fully support, & then discover the problem when a distributor asks for the technical file or a marketplace asks why the expiry date is shorter than a competitor's.

This article covers what each type of stability work shows, why the pack is part of the test rather than an accessory to it, what the alcohol & fragrance question adds, & which changes force a study to start again. It is written for buyers & sellers rather than for laboratories, so the emphasis is on which decisions the data supports & what to ask before accepting a summary.

Shelf life is a claim a brand has to be able to support

The period printed on a pack is not a property of the formula that can be calculated from a table. It is a statement that, under the conditions the brand has defined & for as long as the product is likely to be kept & used, the product continues to meet its specification. That is why the same bulk can carry two different periods in two different packs, & why the answer changes with the market's climate, the channel's storage & the way the product is used.

Supporting the statement means being able to show three things: what specification was maintained, what conditions were studied, & what data exists at the time the claim is made. A brand that can produce the first two & not the third is in the most common position in the industry: it has a defensible approach, an unfinished study, & a pack that already says something.

The practical consequence is that the claimed period should be chosen at the brief, not at the printer. A shorter claim that is fully supported is worth more commercially than a longer one that invites a question, because the claim interacts with everything downstream: the production planning cycle, the retail delivery window, the marketplace returns policy & the pace at which a distributor will reorder.

What accelerated work can & cannot show

Accelerated conditions are used because they make a formula fail faster if it is going to fail. Elevated temperature, & often elevated humidity, drive the reactions that change a product: oxidation of oils, hydrolysis, separation of an emulsion, a shift in colour, a change in odour, a fall in viscosity, a drift in pH. A study that holds its specification under those conditions is a reasonable indication that the formula is robust.

What accelerated work cannot do is stand in for time. The relationship between behaviour at elevated temperature & behaviour at ambient temperature is not fixed, & it is different for different mechanisms. A formula can be excellent under accelerated conditions & still change slowly at ambient over a long period, or the reverse, because a different process dominates at each temperature. That is why accelerated data usually supports a statement about a shorter period, & why the pack claim is often set conservatively until real-time data accumulates.

Accelerated work is also the place where incompatibilities surface earliest. Separation, crystallisation, a colour shift in a fragrance, a shell that becomes brittle or a capsule that blooms: these tend to appear under stress before they appear on a shelf. Reading the accelerated results as a screening tool rather than as a shelf-life verdict is the position most technical buyers take, & it is the position that keeps the brand out of trouble.

Real-time work & the calendar problem

Real-time study means holding the product under the conditions it will actually meet & testing it at intervals across the period the brand wants to claim. It is the only route to a full claim, & it is also the slowest. Nothing about it can be compressed, which is why the decision to start it belongs to the first production batch rather than to the second year of selling.

The commercial problem is that real-time work produces no launch benefit. It costs money, occupies shelf space in a stability room & generates data that will be read long after the launch meeting is forgotten. Brands skip it, print a shorter claim, & then find that a distributor in a hot market wants a longer one, or that a marketplace's own freshness rules push the effective selling window below what the brand planned.

The workaround is to run real-time work as a maintenance activity rather than a project. Samples from the first commercial batches go into storage as a routine step, & checkpoints are read when they fall due. A brand that does this has a defensible position within a couple of production cycles, without ever having paid for a single large study.

Freeze-thaw & transport simulation

Products do not travel at ambient temperature. A container crossing a warm route can sit in high heat for a long stretch, an air consignment can spend a night in a cold hold, & a pallet can be moved between a heated warehouse & an unheated loading bay several times before it reaches a shelf. Freeze-thaw cycling is the standard way to look at what repeated temperature swings do to an emulsion, a gel, a suspended pigment or a filled capsule.

The failures this work finds are often physical rather than chemical: a cream that grains after cycling, a gel that becomes stringy, a serum that separates & does not recover when returned to room temperature, a capsule shell that becomes cloudy or sticky. For cross-border sellers these are the failures that matter most, because they happen away from any laboratory & are only visible to the customer.

Transport simulation is worth planning around the actual lanes the product will travel rather than a generic assumption. A brand shipping by sea to a humid market & by air to a temperate one is running two different journeys, & the study should reflect both. Where the destination is extreme, it is sometimes cheaper to change the pack or the shipping method than to strengthen the formula, & that is a decision the study data makes easier.

Light exposure, colour & odour drift

Light is the stability variable that brands underestimate, because most of the damage happens without any visible change to the container. Ultraviolet & visible light drive oxidation in oils, fade pigments, degrade some vitamins & change the profile of a fragrance. A product that is stable in a closed cabinet can drift on a shelf under store lighting or on a bathroom windowsill.

The pack is the main defence & also the main variable. An opaque bottle, an opaque tube, a UV-blocking treatment or a carton that keeps the primary pack out of light can change the picture completely. This is why light exposure work & pack choice belong in the same conversation: a formula that needs protection is a formula that constrains the pack, & a brand that chooses a clear glass bottle for shelf presence may be choosing a shorter claim as well.

Colour & odour drift are the two changes customers notice first, & both are difficult to defend after the fact. Recording the colour & odour of the reference sample at release, with enough description to be repeatable, is a small discipline that pays for itself the first time a customer complains. Without a reference, the argument becomes a matter of recollection on both sides.

Testing in the pack, not in glass

A formula tested in a laboratory container tells the brand about the formula. A formula tested in the production pack tells the brand about the product. The difference matters because the pack participates in the chemistry: plasticisers can migrate into an oil, an inner liner can absorb fragrance or a preservative, a pump can allow slow moisture loss, a laminate can delaminate at higher temperature, & a metal component can catalyse oxidation.

Testing in the final pack also captures the fill level, & fill level is a stability variable that brands rarely consider. Headspace in a bottle or a jar holds air that reacts with the contents, so the same formula filled higher or lower can behave differently. Where a brand plans several pack sizes from one bulk, the smallest & the largest are usually the ones worth studying, because they represent the extremes of headspace relative to volume.

Where the production pack is not yet available, a study can start on a container of the same material & construction from a stock source, on the understanding that it will be repeated on the production pack before the first commercial shipment. That is a reasonable compromise & should be written down as one, so that nobody later presents the earlier result as if it described the finished pack.

Alcohol, fragrance & the flash point question

Products built on alcohol behave differently in almost every respect. The alcohol content makes the product more volatile, so loss through an imperfect seal is a real mechanism rather than a theoretical one, & the loss changes both the scent & the texture. Alcohol also interacts with some fragrance components & with some pack materials, & it changes what the product does when it meets air at the surface of a pump or an atomiser.

The flash point matters commercially as well as technically. A finished product with a low flash point is handled differently for transport: it affects how the goods can be shipped, which documentation travels with them, which packing group applies & whether an air lane will accept them at all. A brand that only learns this at the point of booking freight is often choosing between a delay & a redesign.

The useful discipline is to establish the flash point of the bulk & of the packed product early, & to record it with the stability file, because the number is stable & cheap to determine while the consequences of not knowing it are neither. Fragrance-heavy formats add a second thread: the compound version affects both the sensory result & the label, so a fragrance substitution during a stability programme can invalidate a pack as well as a study. Where a fragrance is present, the twenty-six declarable allergens commonly tracked for labelling should be confirmed against the compound version actually in use, because that list can move when the compound does.

Compatibility: the pack is part of the formula

Compatibility work overlaps with stability work & is sometimes sold as a separate item, which is why brands regard it as optional. It answers a different question: not whether the product changes over time, but whether the product & the pack change each other. Migration from a plastic into the fill, absorption of an active into a liner, a gasket that swells until the pump stiffens, a tube shoulder that cracks after limited flexing: all of these are compatibility findings.

The consequences are rarely subtle. A pump that loses dose accuracy after a few weeks changes the amount of product a customer uses per application, which affects how long a pack lasts, which affects repeat purchase. A liner that absorbs a fragrance can flatten the scent even though the bulk analysis is unchanged. None of these would appear in a bulk study, & all of them would be blamed on the formula if nobody had tested the pack.

The cheapest arrangement is to enter the final pack, or a stock equivalent, into the same programme as the formula, so that physical & chemical observations are read at the same checkpoints. That gives the brand one file with one narrative rather than two reports that disagree, which is exactly what a distributor or a technical buyer will ask for.

Re-testing triggers after any change

Stability data describes a version. A change to the formula, the fragrance or its level, the preservative system, the pH target, the process, the supplier of a major raw material, the pack material, the fill volume or the shelf-life claim makes that description incomplete, & the question of how much needs repeating depends on what changed.

Some changes need a fresh full programme: a new fragrance compound, a different preservative system, a move from a tube to a jar. Some need a shorter check: a fill volume change within the same pack, a minor colour adjustment, a supplier change for an ingredient with the same specification & the same grade. Some need only a documented rationale: a supplier change for packaging components made to the same drawing & the same resin.

The mistake is to make the scale of the re-test a judgement taken after the change. It should be a rule written before it, with a named person confirming which category a change falls into & recording the answer. A factory & a brand that share that rule spend far less time arguing about whether an old report still applies, & far less money repeating work that did not need repeating.

What each piece of stability work shows

The table below compares the main types of stability & pack work against the question each one answers, what it typically shows about a product, & the decision it supports. It describes common practice rather than a fixed requirement, because the applicable expectation depends on the destination & on the product type.

The rows describe relative patterns rather than a fixed checklist for every product. A water-free capsule fill & a high-alcohol mist will draw on quite different rows, & the same brand selling both should expect two different programmes. What the table is useful for is sequencing: the earlier rows decide whether the product can be made at all, & the later rows decide how it is packed, shipped & sold.

Reading a stability summary before you accept it

A summary should name the formula version, the pack & fill volume, the batch, the conditions studied, the checkpoints, the specification against which each result was compared & the direction of any drift. Without the version, the document cannot be tied to anything; without the specification, a result has no meaning.

Three details are worth checking closely. The first is whether the reported result is a single reading or a trend, because a viscosity that has fallen slowly across the programme is more informative than one number inside a range. The second is whether the observations are described as well as the measurements, since odour & appearance changes often carry the customer complaint. The third is whether the conclusion states the period the data supports, rather than a longer period the brand would prefer.

It is also worth asking what happens to the file when the data is incomplete. A summary that says a study is continuing, with the checkpoints completed & the ones outstanding named, is more useful than a document that presents partial data as final. Buyers read the second kind once & the first kind with confidence.

What to put in the brief

Stability scope is decided by four things: the destinations, the shipping lanes, the packs & the period the brand wants to print. Give all four at the start, & say which of them is still open, because a study designed around an unsettled pack will be repeated. If the brand has already printed a period on artwork for a related product, say what it was, because consistency across a range is easier to defend than a set of different claims.

• The destination markets & the climates & storage conditions they imply.
• The shipping lanes, including any air freight or long sea route.
• The pack formats, sizes & materials under consideration.
• The period the brand intends to print, & whether real-time data is required for it.
• Whether the formula contains alcohol, a heavy fragrance load or a light-sensitive active.
• The process steps that differ from the pilot batch, since scale changes behaviour.
• The change types that should trigger a repeat, & who confirms them.
• Who holds the file & how a distributor or marketplace request is answered.

Then decide how the claim will be handled while the study is still running. The workable approach is to print a period that the completed data supports, with the remaining checkpoints named in the file & a route to extend the claim later if the results allow it. Extending a claim when the data arrives is a quiet piece of good news. Withdrawing one because the data never came is not.

Where our capability fits

We start stability work on the frozen formula & the production pack, & where tooling is not ready we begin on a stock container of the same material & repeat on the production pack, with that sequence recorded rather than glossed over. Freeze-thaw, light & humidity work is planned around the lanes the brand actually ships on, & observations on colour, odour & texture are written into the record alongside the measured values, so that a complaint can be compared with a reference instead of a memory.

We work with brands on private label (your brand on a proven base), OEM/ODM contract manufacturing & full custom formula development, & the stability programme is scoped to the format in each case. Our lines include hyaluronic acid serum softgel capsules & a retinol essence capsule, formats where shell integrity, fill stability & pack compatibility are read together rather than as separate questions.

Tell us the destinations, the lanes, the packs & the period you want to print, & we will come back with the work that supports it, the changes that would send us back to the start & the point at which the claim can be extended. Start at our inquiry page, email capsulebiotech@163.com, or message the right team on WhatsApp: Capsule Series (capsules only) +86 158-7655-5319, Cosmetics Series (serum, hair care, body care, fragrance) +86 158-7655-5319. The wider stability picture is in our article on accelerated studies, & the pack side in our guide to packaging compatibility testing.

General manufacturing information, not regulatory or legal advice. Stability expectations, shelf-life declarations, transport classification & documentation duties vary by market & by product, & no study result can be guaranteed in advance, so the applicable requirements should be confirmed for each destination before a product is placed on sale.