The FTC's Health Products Compliance Guidance states it directly: under FTC law, advertisers must have a reasonable basis for their product claims before disseminating an ad. Not after. Not during legal review. Before the ad runs. Civil penalties for violation can reach $53,088 per day per ad, and the FTC pursues not just the company but individual officers, agencies, and endorsers who participated in misleading marketing.
For brand and regulatory teams, that requirement creates a practical challenge: finding a positive study is not the same as having substantiation. The research has to be strong enough to support the specific claim being made, and the claim language has to accurately reflect what the research actually demonstrates. Both conditions must be satisfied before any claim reaches a label, an ad, or a product page.
This guide covers how to evaluate evidence, translate scientific findings into accurate consumer-facing language, and build the documentation that makes every claim defensible.
An evidence-based product claim is one where the language used to describe a product benefit is matched to the quality and scope of the research that supports it. The claim does not go beyond what the evidence shows, and the evidence is strong enough to support what the claim asserts.
The relationship between scientific evidence, consumer-facing language, and regulatory expectations is not a sliding scale where more creative language is just riskier. It is a matching requirement.
The FTC's substantiation standards training materials make this explicit: the advertiser must possess at least the level of substantiation expressly or impliedly claimed in the advertisement. If an ad says "clinically proven," the evidence must meet the standard that phrase implies.
The practical rule: claim language sets the evidence bar. Write the claim first, then ask whether the available evidence actually meets the bar that specific language implies.
The most common source of evidence-to-claim mismatch is starting with the marketing message and then searching for research to support it. This reverses the correct sequence. Science-backed product claims begin with an inventory of available evidence.
Sources of research that can support product claims:
For each source, the first question is not "does this support our claim?" but "what does this actually show?" Starting from the research allows the claim to describe what is actually demonstrated rather than what the brand would like to communicate.
Finding a relevant study is necessary but not sufficient. The study's design, scope, and outcomes determine what claim language it can support.
Study design is the primary determinant of evidence strength. The hierarchy from strongest to weakest for claim substantiation purposes:
The FTC's guidance on health product advertising specifies that for health claims, "competent and reliable scientific evidence" is the standard, and that this typically means human clinical data. Animal and in vitro research generally cannot support consumer-facing efficacy claims on their own.
A study's relevance to the specific product being marketed matters as much as its design quality. Key relevance questions:
Applying an ingredient study directly to a finished product without verification that the finished product delivers the same ingredient in the same form and concentration is one of the most common claim substantiation errors.
What the study actually measured and what it showed determines the claim. Review:
Research papers describe outcomes in technical terminology that serves scientific communication, not consumer decision-making. Translating findings into product claims requires a deliberate conversion process.
The four steps of translation:
An example: a study demonstrates a statistically significant improvement in a validated measure of sleep onset latency in adults with self-reported mild sleep difficulty. The scientific finding is specific and bounded.
A consumer-facing translation might be: "Shown in a human study to help adults fall asleep faster." Adding "dramatically" or "guaranteed" exceeds the evidence. Saying "helps you sleep better" broadly overstates what was actually measured.
How to write product claims from studies is one of the most specific and practically important skills in regulatory affairs and marketing. The framework below produces claims that are both consumer-friendly and evidence-accurate.
The evidence-to-language calibration is the most nuanced aspect of creating claims backed by research. Each tier of evidence supports a corresponding tier of claim language.
Strong evidence (multiple RCTs with consistent results, or a single well-designed RCT with adequate power):
Moderate evidence (one human study, or multiple studies with mixed results):
Limited or indirect evidence (ingredient studies, mechanism research, or observational data):
Using strong language for moderate or limited evidence creates regulatory exposure. The FTC's substantiation requirement is assessed against what the claim implies, not just what it literally states. A claim that "studies show" implies multiple studies. A claim that "clinically proven" implies the standard that phrase is understood to represent.
For a broader look at how labeling language becomes a legal liability, see our guide on how misleading labeling on products can lead to regulatory trouble.
A claim without documentation is a liability. Documentation transforms a claim from an assertion into a defensible record.
Every product claim should have a substantiation file that includes:
This record serves three purposes: it enables faster approval for similar future claims, supports efficient review when claims need updating, and provides the documentation that regulators can request in an inquiry.
A claim that is technically accurate may still create compliance problems through implied meaning, visual context, or the way it is understood by the consumer the FTC focuses on: the reasonable consumer.
Questions the review team should ask before approving any claim:
Label context matters as much as claim wording. Our guide on how to design a label for any product type covers how visual hierarchy and layout affect how claims are interpreted by both consumers and regulators.
Understanding where evidence-based product claims break down is as useful as understanding how to build them correctly.
For teams working in chemical, supplement, or food categories, understanding the mandatory elements that must appear alongside any marketing claims is equally important.
See our breakdown of the six elements every GHS label must include for the compliance architecture that surrounds claim language in regulated categories.
Building evidence-based product claims at scale, across product ranges, markets, and regulatory contexts, creates workflow challenges that manual review alone handles poorly. AI-assisted claim review addresses this by integrating compliance checking directly into the claim development process.
GetGenAI's compliance review platform integrates into label development workflows through API, MCP server, or AI assistant skills, allowing teams to run automated checks on claim language against current regulatory requirements at each revision.
For science-backed product claims specifically, this means catching language that exceeds the substantiation standard implied by the evidence, identifying wording that could be interpreted as disease claims, and flagging claims that require substantiation documentation before approval.
For teams in regulated health-product categories, GetGenAI's AI-powered regulatory compliance tools for supplements and pharma apply category-specific rulesets to claim review across the full product range.
The workflow integration options include API calls at each design milestone, MCP server connections for AI-native production pipelines, and drop-in skills for teams using Claude or Cursor in their review process. Each integration point runs compliance checks during development rather than at final submission, which compresses the feedback cycle from weeks to minutes.
Before any claim is approved for production:
The strongest product claims are not the boldest ones. They are the ones that accurately represent what the science shows and can be defended with documentation when challenged. The FTC's standard is prior substantiation, not post-hoc justification.
Teams that build the evidence-evaluation and documentation process into their claim development workflow produce claims that clear regulatory review faster, require fewer revision cycles, and carry less legal exposure than teams that write the marketing message first and find the evidence later. The discipline is front-loaded. The commercial benefit is the claim that survives.
Upload your product label and claim language to GetGenAI for automated compliance review against FTC substantiation standards and regulatory requirements before production.
What type of scientific evidence is considered strongest for supporting product claims?
Randomized controlled trials conducted in humans are the strongest form of evidence for consumer-facing product claims. The FTC's standard for health claims is "competent and reliable scientific evidence," which in practice means well-designed human clinical trials. Multiple RCTs with consistent results provide the strongest substantiation. A single well-designed RCT can support a claim when the study population and conditions are relevant to the marketed product.
Can brands make claims based only on ingredient studies, or do they need product-specific research?
Ingredient studies can support claims, but only when the finished product delivers the ingredient in the same form and at a comparable dose to what was used in the study. If the formulation, dose, or delivery mechanism differs from the studied ingredient, the ingredient study does not automatically transfer to the finished product. Brands relying on ingredient studies should document the connection between the studied ingredient and the marketed product as part of their substantiation file.
How do companies decide whether to use "supports" versus "clinically proven" in claims?
The language should match the strength of the evidence. "Clinically proven" implies a high evidentiary bar, typically multiple well-designed human clinical trials with consistent results, and the FTC will assess whether the evidence meets the standard that phrase implies to a reasonable consumer. "Supports" requires a reasonable scientific basis but can accommodate more limited evidence, such as a single human study or strong mechanistic research. The decision should be made by evaluating what the evidence actually shows and selecting the language tier that accurately reflects that evidence level.
Get intels on brand, platform, and regulatory changes, distilled by our compliance team
