A cosmetic stability and packaging compatibility study should answer defined questions about an identified formula in an identified pack. Before commissioning work, agree what the study will evaluate, which configuration it covers, how results will be judged and who will interpret them. A quote labeled “stability test” does not settle any of that.

This guide covers how to request a useful study. Designing the laboratory protocol is the provider’s job. It includes a copyable request and a filled fictional example. Use the cosmetic product development roadmap to line the testing decision up with formulation, production and launch.

Before you begin: name the decision the study will support

State why you need the evidence. Comparing development candidates, evaluating a proposed pack and supporting a launch decision are different tasks. Ask the provider what its proposed work can and cannot establish for yours.

Prepare the current formula reference, sample or batch identity, intended use, target markets, pack details, existing reports and known changes. Name the brand decision owner and the technical reviewer. If the final pack is undecided, say so. Do not describe an early container as the commercial configuration.

For EU products, Annex I of the Cosmetics Regulation lists stability under reasonably foreseeable storage conditions and the relevant characteristics of packaging material among the information in the safety report. Your stability evidence therefore has to feed the safety assessment. The Regulation does not set a universal study schedule.

Step 1: Separate stability, compatibility and dispensing objectives

Give each part of the work its own question. The investigations may overlap, but each conclusion should be identifiable.

Intertek’s Cosmetic Stability Testing page distinguishes formula stability, final-pack compatibility and packaging functionality. Its examples include appearance, pH, viscosity, leaks and dispensing performance. These are one provider’s examples, not mandatory measurements for every cosmetic.

1. Formula stability: which changes in the product will the study track over time, under which conditions?
2. Formula-pack compatibility: what will the study examine about the formula and container together?
3. Packaging functionality: how will closure and dispensing behavior be evaluated?

A single “pass” is meaningless until you know which question it answers. Ask the provider to map each objective to an observation or measurement and a reported conclusion.

Scope microbial evidence separately. Physical observations and pack checks do not prove preservation. The challenge testing versus microbial limits comparison covers those questions.

Step 2: Identify the exact formula and complete pack

“Our moisturizer in a pump” is not a sample description. Make it traceable to the intended commercial product:

1. Record the formula version, sample or batch reference, source and any preparation or filling information from the manufacturer.
2. Identify the container, material specification, size, closure, liner or seal and dispenser, using supplier component references.
3. Record the intended fill quantity and any component variants or decorations.
4. Ask the lab to confirm sample receipt, identification, configuration and any difference from the request.

Keep the tested configuration distinct from the one intended for sale. If an early study uses a development sample or substitute container, have the qualified reviewer explain the limits of applying it to the final product. It is not final-pack evidence.

You do not need to choose technical parameters. Ask the provider which further sample details matter and how much material and how many packs it needs.

Step 3: Agree conditions, criteria and the limits of the evidence

The provider proposes the technical design. The brand needs to understand its scope before authorizing it.

1. Request the proposed conditions, duration, observation schedule and sample requirements, with the reasons for each.
2. Identify the endpoints (the observations or measurements to be evaluated) and who sets their acceptance criteria.
3. Agree who reviews unexpected findings, when the brand is told and what happens if criteria are not met.
4. Ask how interim and final findings will be distinguished and what further evidence is planned.

Agree acceptance criteria before results come in. Do not loosen a criterion after an unfavorable result to protect the launch date. If a criterion genuinely needs scientific revision, document the rationale and review it with the technical parties.

Real-time and accelerated evidence

Real-time observations show what happened over the elapsed study period under the stated conditions. Accelerated work uses stress conditions to probe behavior. Ask the provider why its chosen conditions are informative and what they can support for your product.

There is no rule here that converts a given accelerated exposure into a given shelf life. Ask the qualified reviewer to explain the basis and limits of any proposed durability conclusion and the role of ongoing real-time observations.

Step 4: Send a complete testing request

Send one request that the provider can accept, amend or quote against. This checklist is a procurement tool, not a laboratory method or a legally prescribed form.

Copyable testing-request checklist

REQUEST ID / VERSION / DATE:
BRAND CONTACT / TECHNICAL REVIEWER:
DECISION THIS WORK SHOULD SUPPORT:

PRODUCT IDENTITY
- Product name and intended use:
- Formula version and sample/batch source:
- Available manufacturing/filling references:
- Target markets and relevant existing evidence:

EXACT PACK CONFIGURATION
- Container supplier / component reference / material / size:
- Closure, liner, seal and dispenser references, as applicable:
- Intended fill quantity and relevant variants:
- Differences between supplied samples and final configuration:

SCOPE TO AGREE WITH THE PROVIDER
- Formula-stability objectives:
- Formula-pack compatibility objectives:
- Closure/dispensing functionality objectives:
- Proposed conditions, duration and observation schedule:
- Endpoints, methods and acceptance criteria:
- Sample quantities and receipt/identification arrangements:
- Exclusions, assumptions and interpretation limits:

REPORTING AND DECISIONS
- Interim outputs and final report contents:
- Unexpected-finding notification and investigation process:
- Result reviewer and decision authority:
- Further evidence needed for the intended decision:
- Changes requiring review before applying this evidence:
- Provider scope acknowledgment and unresolved questions:

Get the provider to acknowledge the agreed scope, not just confirm that a parcel arrived. Close any missing field that affects the work before it starts, and file the accepted version with the quotation and sample records.

Filled fictional request

All identifiers below are invented. No study has been performed.

• Identity: request T01, version 1, dated 23 October 2026; leave-on moisturizer M01, formula F03, development batch D02.
• Pack: proposed container C01, component specification CS01; pump PU01 and seal SE01; fill quantity in pack specification PS01. The brand supplies those specifications before scope acceptance.
• Decision: evaluate the formula-pack combination before a non-cancellable component order. This study does not authorize launch.
• Objectives: track defined formula changes, formula-pack suitability and pump/closure performance. Provider to propose endpoints and criteria for each.
• Plan requested: provider to propose conditions, duration, observation schedule and sample quantities, and to separate interim information from the final report.
• Known limitation: D02 is a development batch. Technical reviewer to assess its relationship to production material and name any further evidence needed.
• Owners and holds: manufacturer supplies sample-source information; packaging supplier supplies component specifications; test provider proposes the plan; brand technical reviewer judges its suitability. No testing authorization until scope and inputs are agreed.
• Change trigger: if PU01 is replaced, review whether the plan or existing evidence still covers the pack before accepting the substitute.

The useful outcome of this request is a provider response that closes each open question.

Step 5: Review the report before making the commitment

Record what the report means for the decision you named. A short summary helps, but keep access to the full agreed report and its limitations.

1. Check formula, sample, batch and pack identities against the request.
2. Check which conditions and observations were completed, including deviations and missing work.
3. Compare findings with the agreed criteria. Flag interim, inconclusive or unfavorable outcomes.
4. Have the qualified reviewer record what the evidence supports, what is unresolved and the next authorized action.

No reported problem is not the same as a resolved question. If dispensing was excluded, record the gap instead of calling the whole pack approved.

The decision might allow another development round, require an investigation or keep a component-order hold. It does not have to be a product-wide yes or no. For EU projects, Article 10 requires a separate product safety assessment and report. This study is one input to it, not a replacement.

The decision record should name the report reviewed, the product configuration it applies to, the reviewer, open issues and the permitted next commitment.

Step 6: Review changes before reusing earlier evidence

A change is a trigger for review. It does not automatically mean repeating the same test package.

1. Log proposed formula, raw-material, component, manufacturing or filling changes with references and reasons.
2. Identify which tested configuration and conclusions they might affect.
3. Ask the technical reviewer whether existing evidence still applies and what extra work, if any, is justified.
4. Record that decision before relying on the earlier evidence for the changed product.

The same product name does not make two versions equivalent. In the fictional example, replacing PU01 with PU02 reopens the pack question. It neither inherits the original conclusion nor dictates a particular retest.

Use the cosmetic pilot batch and scale-up checklist to plan production-representative work and transfer records. A lab sample and future commercial batches are not interchangeable by default.

Next step: ask for a scoped proposal, not a generic test price

Send the testing request with formula and pack references, state the decision it should inform and ask the provider to explain its evidence plan. Close the biggest scope gap before authorizing work. Keep the development roadmap next to the report so a good study result does not stand in for the other launch gates.