Cosmetic Claims and Labeling: How Beauty Brands Build a Defensible Evidence Plan

A defensible cosmetic claim starts with a clear statement, an accurate understanding of what the audience will take from it, and evidence that fits the actual product and intended use. The brand should identify the claim before it approves the packaging or launches an advertisement, then ask qualified people to review its wording, supporting information and market-specific requirements.

Three decisions need separate attention. First, does the intended use or message affect how a product is classified in a particular market? Second, does relevant evidence support the express and implied advertising claim? Third, is the finished product safe and appropriately labeled under the rules that apply to it? A positive answer to one question does not automatically answer the other two.

This guide helps beauty brands and their development, marketing, regulatory and quality teams organize those decisions. It uses selected United States and European Union sources as examples. It does not approve a specific claim, provide a universal label checklist or replace professional review for the product and market concerned.

What is a cosmetic claim?

A claim can be a direct statement about a product, such as a description of its purpose or an assertion about an attribute. It can also be communicated through images, comparisons, demonstrations, product names, video, testimonials or the way information is presented. The message that a reasonable audience takes away may be broader than the individual words a marketer intended to emphasize.

For that reason, keep the complete piece of communication when evaluating a claim. A short phrase on a box may appear alongside a photograph, a product name and a visual comparison. An online advertisement may combine a headline, a moving image, a caption and a link to more detail. Review the whole message in context, including what is prominent and what is only visible after an additional click.

The US Federal Trade Commission's advertising-substantiation policy addresses objective express and implied claims. It explains that advertisers need a reasonable basis for such claims before they are disseminated. [1] FTC guidance on health-related product advertising also discusses the message conveyed by the advertisement as a whole and the relationship between that message and the evidence. [2] These sources do not prescribe one study for every cosmetic claim or establish that a particular proposed sentence is compliant.

A useful first step is to write down the message in ordinary language. What does the brand intend to say? What might a buyer reasonably understand it to mean? Is the claim about the product's appearance, its performance under certain conditions, the materials in a package or something else? The answer guides the evidence review and helps the team identify an unintended implication before it reaches the artwork.

Classification, claim support and safety are different questions

Imagine a fictional brand developing a moisturizer. The team plans a package, a product page and a short social-media video. Before deciding which words to print, it should identify the product's intended use, the markets and the message conveyed across all three channels. This fictional example is a planning scenario; it is not a determination that any specific wording is legally permitted or supported.

The first workstream is classification. In the United States, FDA explains that intended use affects whether a product is regulated as a cosmetic, a drug or both. Intended use can be indicated by claims and other circumstances, and some products fall into more than one category. [3, 4] A supplier's description of an item as a cosmetic does not settle the classification of every version marketed with every possible claim.

The second workstream is claim substantiation. The team identifies the message and reviews whether the available evidence supports it. A claim can be accurate in one context yet too broad in another. Evidence about an ingredient, for example, may help explain a formulation choice but does not automatically establish the same effect for the finished product as sold and used.

The third workstream is product safety and applicable labeling. FDA describes the use of relevant existing information and appropriate additional work in assessing cosmetic safety, and says it does not provide one test list for every cosmetic or ingredient. [9] The European Commission describes an EU responsible-person framework and a cosmetic product safety report. [8] These are distinct from a conclusion that a particular advertisement is substantiated.

Keep the teams connected, but do not combine the decisions into one broad approval box. A product may need additional classification review even if the marketing team has collected data. A safety assessment may identify a question that changes the proposed label. A claim may require different supporting information from the information used for a separate safety decision. Assign owners and record which question each decision answers.

Build a claim register before artwork is final

A claim register is a working record of what the brand intends to communicate and the evidence linked to each message. It can be maintained in a shared document or another controlled system. The format matters less than the ability to identify the current wording, the product version and the person responsible for the next review.

Start with the exact phrase and the complete context in which it will appear. Record variants used on the primary package, outer carton, website, marketplace listing, advertisement, sales presentation and other relevant channels. A short headline and a longer product-page explanation may convey different levels of certainty even when they concern the same attribute.

For each version, record the intended audience, country or region, product identity, formula and package version where relevant, intended use and the likely takeaway. Note whether the message is proposed, awaiting evidence, supported within stated limits, rejected or under review. Those labels should reflect a real review process, not imply that a claim is approved merely because a row has been completed.

Add an evidence owner and links to the relevant underlying records. A useful entry identifies the source, what it actually assessed, its date, the product or sample to which it applies and its limitations. If the file is held by a supplier, record how the responsible reviewer can access it. If a claim relies on more than one document, explain how those documents connect rather than listing titles without interpretation.

Finally, set a change trigger. If the formula, package, use instructions, target audience, market or claim wording changes, someone should decide whether the prior review still applies. A register is not a legal requirement in every situation, but it can make the decision and its evidence visible before changes are made across multiple channels.

Map evidence to the finished product and the actual claim

Evidence has a purpose and a scope. Before accepting a report, ask what was studied, who or what was involved, which product or material was used, under what conditions and what was measured. Then compare those details with the message the brand proposes to communicate. A document is not useful support simply because its title contains a familiar ingredient or product category.

The sample may differ from the marketed product in composition, package or use. A concentration or ingredient level can be relevant when judging whether data on one material or formula applies to another, but the number alone does not establish that the finished product will have a particular effect. The responsible reviewer should understand the differences and document whether the evidence supports the intended conclusion.

The intended population also matters. Ask whether the people, model or conditions considered in the evidence resemble those relevant to the claim. A finding observed under a controlled condition may not support an unrestricted promise for every person or every pattern of use. Avoid turning a qualified observation into a broader statement simply because the broader wording sounds stronger.

The EU common criteria for cosmetic product claims address evidential support and the relationship between a claimed property and the product. The regulation's criteria include considerations about the relevance and quality of evidence, and explain that an ingredient's property does not automatically become a claim about the finished product. [6] The exact application depends on the claim and product, and the current rules for the intended market must be checked.

Packaging can matter when it is part of the message or the use being assessed. If an advertisement discusses a dispensing feature, the relevant evidence should relate to that feature and its actual configuration. If an environmental claim concerns a component, identify the specific component and its conditions rather than using unrelated evidence about a material in isolation.

When a supplier provides a summary, ask whether the people responsible for reviewing the claim can see enough information to assess relevance and limitations. Confidentiality can be handled through an agreed process, but an unsupported conclusion should not become an approved message merely because the underlying data is unavailable to the reviewer.

Select evidence that answers the question

There is no single evidence format that automatically proves every cosmetic claim. Depending on the message, relevant information might include a product-specific assessment, instrument measurements, consumer perception work, technical documentation, published research or a combination of sources. The appropriate approach depends on what is claimed and what the audience would reasonably understand.

When reviewing a study, ask how the question was defined. Was the endpoint directly related to the proposed claim? Was there a comparison or control relevant to the question? Were the measurement method, conditions and participants or model described? What limitations did the authors or laboratory identify? The answers help the reviewer decide whether the study is applicable; they are not a template for a universal study protocol.

Keep measurement and perception distinct. An instrument can measure a defined property under a specified method. A consumer study may ask people about their experience or opinion. Either can provide useful information for the right question, but a statement about a measured result should not be presented as if it were a statement about all users' experiences, or vice versa.

Consider whether the evidence addresses the same product version, use pattern and audience as the claim. If several pieces of evidence are combined, explain what each contributes and whether their conclusions are compatible. A literature review may provide context; it does not automatically replace work needed to connect a claim to the finished product. Similarly, a certificate or a safety record should not be relabeled as efficacy evidence without a reasoned basis.

FTC guidance for health-related advertising discusses the relevance of the product, claims, study design and evidence as a whole. [2] Its examples include categories beyond ordinary cosmetics, so do not assume that every example or protocol in the guidance is mandatory for all cosmetic products. The practical lesson for a brand is to ask a qualified reviewer what evidence is appropriate for the actual message and applicable law.

Record the review outcome in terms that preserve its limits. “Supports this defined statement for this product under these conditions” is more informative than a file name marked “approved.” If evidence is incomplete, identify the gap and hold or narrow the proposed wording while the responsible team decides what can be supported.

Use numbers, comparisons and visuals honestly

Numerical claims can be attractive because they appear precise. Their meaning, however, depends on the underlying measurement. Before placing a number on packaging or in an advertisement, identify the endpoint, method, relevant product, conditions and period to which it relates. Ask whether the full presentation creates a broader impression than the data supports.

Do not invent a percentage, a participant count or a result to complete a design. If a report contains a figure, the reviewer should check how it was calculated, whether it is an average or another measure, and whether the proposed wording accurately describes the finding. A number drawn from an ingredient study should not be presented as a finished-product result without a substantiated link.

Comparisons also need a clear reference point. “Better” or “improved” invites a question: better than what, for which property and under what conditions? An internal prototype, a prior product and an unnamed competing product are different comparators. The brand should understand the comparison before approving the claim and should not imply that a result covers comparisons the evidence did not make.

Images can communicate claims without using a numerical statement. A before-and-after image, a diagram or an illustration of a process may imply a product effect, a test result or a real facility. Use visuals that match the evidence and identify conceptual illustrations where appropriate. Do not fabricate clinical images, test records or outcomes to make a page more persuasive.

Disclosures can explain the conditions or limitations of a supported statement, but they are not a replacement for evidence. Consider whether a reader can notice and understand a qualification in the context of the complete message. The FTC's substantiation policy and EU claim criteria both make the substance of the representation important; a small note cannot turn an unsupported broad promise into a supported one. [1, 6]

Review label details as well as promotional copy

The product label carries more than advertising. Depending on the product and market, it may need to communicate identity, net contents, the responsible firm, ingredients, directions or warnings. Requirements can differ by package configuration, product category and jurisdiction. Begin the label review early enough to reserve appropriate space and obtain information from the relevant parties.

FDA's US summary describes a range of cosmetic labeling requirements, including applicable identity, net quantity, business information and ingredient declarations. [5] It also explains placement and other details that can depend on the product. The summary is a starting point, not a complete approval of an unknown label. Products that are also drugs may have additional or different requirements.

Identify who prepares the ingredient information and who checks it against the final formula. Ask how a late raw-material substitution or a packaging change will be communicated to the artwork owner. A product name, online listing and physical label should be reviewed together where they communicate the same use or claim, even when different teams create them.

For the EU, identify the relevant responsible person and the professionals who can review the actual product information and label for the intended countries. [8] Language, presentation and other requirements may depend on the destination and facts. An EU notification, a safety report and claim substantiation are separate elements; completing one should not be described as proof that every label and advertisement is correct.

Before production, agree on the authoritative artwork version and the sign-off process. Record open questions, the parties who supplied required information and the version reviewed. A final proof is easier to assess when the underlying formula, package dimensions and market are also identified. If a change occurs after approval, bring the affected elements back to the relevant reviewers.

United States: classification and advertising require coordinated review

FDA explains that a product's intended use is central to its US classification. The agency notes that intended use may be established by labeling and advertising, consumer perception or other relevant information, including known therapeutic uses of ingredients. [4] A product can be a cosmetic, a drug or both under the applicable definitions. A broad category description in a supplier catalog should not replace a review of the actual product and communication.

Claims about treating disease or affecting the structure or function of the body can raise drug-related questions in the United States. [3, 4] The precise classification of an actual product depends on its facts. Avoid assuming that a phrase will be acceptable because another brand uses similar wording, or that a disclaimer elsewhere on the page will necessarily determine the product's intended use.

FDA says it does not provide a preapproved list of cosmetic claims for manufacturers to select from. [3] This makes an internal process especially useful: record the exact proposed message, identify the likely takeaway, assess the evidence and ask the appropriate legal or regulatory professional to review the classification and labeling implications.

FTC advertising principles add another question: whether the advertiser has a reasonable basis for objective claims before making them. [1] The classification of a product and the substantiation of a message are related but separate. Evidence that supports one representation does not automatically establish that every stronger or differently framed representation is supported.

The same review should cover digital channels. A brand's website, paid advertisement, product listing and other marketing material may each present the product differently. Keep the evidence record linked to the actual versions used, including material supplied to partners. If a message changes, determine whether the existing support still answers the revised claim.

European Union: common criteria and supporting evidence

Commission Regulation (EU) No 655/2013 sets common criteria for claims made in relation to cosmetic products. Its annex identifies six criteria: legal compliance, truthfulness, evidential support, honesty, fairness and informed decision-making. [6] The criteria concern the substance and presentation of claims; a brand should assess the message in its actual context rather than relying on a list of attractive phrases.

The evidential-support criterion addresses adequate and verifiable evidence for claims and the relevance of studies where they are used. [6] That does not create one universal study design, participant count or test for every cosmetic message. A qualified reviewer should consider the claim, product, available information and current applicable requirements.

The 2017 Technical Document on Cosmetic Claims, prepared by the Sub-Working Group on Claims and endorsed by the Working Group on Cosmetic Products, explains approaches to applying the common criteria and includes examples concerning certain types of claims. [7] It is an explanatory best-practice tool, not a European Commission document; it should be read alongside the regulation and current applicable rules rather than treated as a blanket authorization for any phrase. Its examples do not establish that an unspecified product is safe or that every national interpretation is identical.

The EU framework also includes a responsible person and a cosmetic product safety report. [8] These are part of the broader product framework, not a substitute for checking the evidence and presentation of each claim. Similarly, an ingredient name appearing in CosIng is not an approval of the ingredient for every use or proof that the finished product can be marketed as described. [11]

Identify the actual EU destination, product, responsible person and language requirements with qualified support. The United Kingdom is a separate jurisdiction and should not be treated as automatically covered by an EU-specific statement. When rules or the product change, revisit the evidence and the communication rather than relying indefinitely on an old review date.

Review sensitive wording in context

Terms such as “natural,” “free-from,” “hypoallergenic,” “recyclable” and “sustainable” can carry different meanings for different readers. The brand should define what it intends to communicate, identify the evidence and check the complete presentation in each target market. Avoid creating a broad list of universally safe or prohibited words without reference to the product and context.

A “free-from” statement should be reviewed for what it implies as well as for its literal accuracy. The EU technical document discusses examples and considerations for such claims. [7] A brand should check whether the claim is fair, relevant and supported for the actual product rather than assuming that the absence of an ingredient makes every comparative implication appropriate.

“Hypoallergenic” should not be treated as a promise that no person can experience a reaction. FDA discusses the variability of individual sensitivities and the US context of allergen-related information. [12] An EU guidance example and a US discussion do not automatically produce the same legal conclusion in every market. Ask the responsible specialists what the intended statement communicates and what evidence supports it.

Environmental claims require the same precision. The FTC's Green Guides caution against broad, unqualified environmental-benefit claims and discuss the need for clear qualifications. [10] A material property, a supplier statement or a conceptual image of packaging does not establish that the complete package is recyclable everywhere or that a product has a general environmental benefit.

Do not use “clean” or “natural” as a substitute for explaining the actual product. If the brand uses such language, define the relevant meaning internally, review the possible implied message and connect any objective assertions to support. A preferred tone or visual style cannot replace a factual basis for a claim.

Maintain the evidence when the product or message changes

Claim support is connected to a particular product and communication. A change to the formula, supplier, package, manufacturing process, use instruction, intended audience or market may change the questions that need review. It does not follow that every change invalidates all prior evidence; the responsible team should assess the change and record the rationale for its decision.

Use version identifiers that connect the current formula, packaging and artwork to the claim register. Keep the report version and the relevant review date visible. If a supplier replaces a raw material or changes the specification, identify who informs the brand and who assesses whether the evidence remains applicable. If the package artwork changes, check whether the new layout changes the message or the visibility of a qualification.

The same applies to marketing updates. A shorter headline, a new photograph or a translated phrase can alter the takeaway even when no formula has changed. Route the revised material through the appropriate review and retain the context in which the claim appears. Do not assume that approval of one version covers all later variations.

Define a path for evidence gaps. The team may be able to narrow a claim, gather relevant information, change the proposed creative or remove the message. Document the decision and the owner. If the underlying support is no longer available or no longer applies, continuing to use the old claim should not be the default simply because the artwork is already prepared.

A fictional claim-evidence matrix

The following example is fictional and contains no test result or approved claim. A brand is considering a general moisturizing message for a proposed finished product. The team has an ingredient information sheet and a sample, but the final formula and communication are still being reviewed.

Decision item	Question for the project	Evidence or action to identify	Status in this fictional example
Intended message	What would the complete label and advertisement lead the audience to expect?	Record exact wording, visuals, market and intended takeaway.	Pending review.
Product identity	Does the evidence relate to the final formula and intended use?	Confirm the current sample, formula and package versions.	Final version not yet confirmed.
Claim support	What information supports the specific proposed statement?	Ask qualified reviewers to assess relevant product and ingredient data.	Evidence not yet evaluated.
Classification and label	Does the presentation raise a classification or labeling question in the target market?	Obtain product-specific regulatory and legal review.	Pending.
Change control	Who will review a later formula or wording change?	Assign an owner and connect the decision to versioned records.	Owner to be agreed.

The matrix does not say that a study is always required, that an ingredient sheet is sufficient or that the fictional claim may be used. It shows how to keep the question, evidence and decision separate. The appropriate work depends on the exact product, claim, market and available information.

Questions to ask your development and marketing teams

Before a claim reaches production, ask the people who create, supply and review the message to work from the same product version. These questions can make the review more useful:

• What exact statement, image or comparison will the audience see, and what might it imply?
• Which product, formula, package and intended use does the statement concern?
• Which countries and channels are included in the review?
• Who is responsible for classification, safety, labeling and claim substantiation?
• What does each report actually measure, and how does it relate to the finished product?
• Are relevant limitations, comparators and conditions preserved in the proposed wording?
• Can the qualified reviewer access the underlying evidence, including confidential supplier information where needed?
• Which label details and artwork version have been checked for the actual market?
• What happens if the formula, packaging, audience or wording changes?
• Which claims remain pending or should be removed until a decision is made?

The answers should point to identifiable records and decision owners. A supplier's general assurance can be a starting point for a discussion, but it is not a substitute for understanding what a specific document supports. Keep unresolved questions visible until the appropriate person has addressed them.

Frequently asked questions

Can an ingredient claim be used as a finished-product claim?

Not automatically. Ingredient information may be relevant, but the reviewer should consider the finished formula, intended use, audience and exact message. EU claim criteria specifically distinguish an ingredient's properties from claims about the finished product. [6] Determine whether the available evidence supports the actual statement before using it.

Does a cosmetic safety report approve all marketing claims?

No single document should be assumed to answer every question. Safety evaluation, product classification, labeling and advertising substantiation have different purposes. Identify which issue a report addresses and ask qualified reviewers whether additional evidence or review is appropriate for the proposed communication. [1, 8, 9]

Is a consumer opinion the same as a measured product result?

They can answer different questions. A consumer study may describe perceptions or experiences, while an instrument may measure a defined property under a method. The proposed claim should match the information collected and its limits. Do not turn an opinion into an unqualified measured effect, or a controlled measurement into a promise about every user's experience.

Can a disclaimer fix an unsupported claim?

A clear qualification can explain the scope of a supported statement, but it cannot replace a reasonable evidentiary basis. Review the complete presentation, including prominence and the likely audience takeaway. If the main message is broader than the evidence, consider narrowing or removing it rather than relying on a small note. [1, 6]

Does an ingredient's presence in CosIng mean the product is approved?

No. The European Commission describes CosIng as an information resource and explains that inclusion of an ingredient name is not an approval for every use. [11] Check the current applicable legal provisions and the exact formula, use and product with qualified professionals.

Do the same words work in every market?

Do not assume they do. The United States and European Union have different frameworks, and other markets may apply different definitions, labeling rules and enforcement approaches. Review the actual wording, product and destination. A translation should preserve the intended meaning and should also be assessed in its new context.

When should a claim be reviewed again?

Reassess the claim when a relevant product, evidence, market or communication change occurs, and according to the brand's documented review process. A change does not always require a new study, but the responsible team should determine whether the existing support remains applicable and record the decision.

Start with the exact message and the product behind it

Bring the proposed label and marketing material into the development conversation early. Identify the product version, intended use and markets, then map each objective message to the evidence and people needed to review it. Keep classification, substantiation, safety and labeling decisions visible as separate parts of the same project.

A clear evidence plan helps a brand ask better questions of its manufacturer, laboratory and advisers. It also gives the team a practical way to handle changes before an unsupported statement becomes part of packaging or a public campaign.

References

Sources checked 5 November 2026. The dated FTC policy, EU regulation annex and EU technical document are cited for the provisions and guidance described; laws, standards and guidance may change. Confirm current requirements and the actual product and market with qualified professionals.

1. US Federal Trade Commission. Policy Statement Regarding Advertising Substantiation, 1984.
2. US Federal Trade Commission. Health Products Compliance Guidance, 2022.
3. US Food and Drug Administration. Cosmetics Labeling Claims.
4. US Food and Drug Administration. Is It a Cosmetic, a Drug, or Both? (Or Is It Soap?).
5. US Food and Drug Administration. Summary of Cosmetics Labeling Requirements.
6. European Commission. Commission Regulation (EU) No 655/2013, Annex: Common Criteria, 2013.
7. Sub-Working Group on Claims, endorsed by the Working Group on Cosmetic Products. Technical Document on Cosmetic Claims, 2017.
8. European Commission. Cosmetics legislation.
9. US Food and Drug Administration. Product Testing of Cosmetics.
10. US Federal Trade Commission. Environmental Claims: Summary of the Green Guides.
11. European Commission. Cosmetic ingredient database.
12. US Food and Drug Administration. Allergens in Cosmetics.

Back to all articles ↗