How to Build a Claim Substantiation Packet

A useful claim substantiation packet connects the exact advertising message to evidence the advertiser possessed before publication. It is not a folder of studies with no explanation. For each express and implied claim, record the marketed product, target audience, evidence standard, supporting materials, material limitations, approved wording, and reviewer decision.

Begin with the final claim, not the available study

The FTC's Advertising FAQs says objective advertising claims require a reasonable basis before an ad runs. The amount and type of evidence depend on the claim; health and safety claims generally require competent and reliable scientific evidence.

Start by capturing what the final content communicates:

• Spoken script and creator improvisation
• On-screen text, captions, and headline
• Demonstrations and before-and-after imagery
• Product name, graphics, and juxtaposition
• Testimonials and generally expected results
• Comparative claims and named competitors
• Landing page or offer language completing the message

Rewrite each message as a testable proposition. “Works fast” is not yet specific enough; identify the result, time, population, and conditions viewers are likely to understand.

Create one evidence card per material claim

An evidence card lets another qualified reviewer reproduce the decision. Include:

1. Claim ID and exact approved wording
2. Express and reasonably implied meanings
3. Product, formulation, dose, version, and conditions
4. Target population and market
5. Required level of support
6. Evidence titles, dates, authors, and locations
7. Study relevance, quality, limitations, and conflicting evidence
8. Required qualification or expected-results disclosure
9. Subject-matter and legal reviewer decisions
10. Approval date, expiration trigger, and affected assets

Do not copy a conclusion from a prior product without checking whether formulation, use, or audience changed.

Evaluate fit, quality, and totality of evidence

For health-related ads, the FTC's Health Products Compliance Guidance discusses competent and reliable scientific evidence and emphasizes human clinical evidence, study quality, clinically meaningful outcomes, and the totality of relevant evidence.

Common mismatches include:

• Ingredient research offered for a different finished product
• Laboratory or animal evidence used as proof of human benefit
• Association described as causation
• A study population unlike the advertised audience
• Dose, duration, or route different from normal use
• A statistically significant but practically trivial effect
• Cherry-picked favorable analysis or omitted negative research
• “Clinically proven” based only on a test that did not prove the claimed benefit

Record contrary evidence and explain how the reviewer weighed it. A packet designed only to defend approval is less reliable than one documenting uncertainty.

Connect evidence to the creator's actual cut

The sponsored-video review checklist helps identify every claim in the assembled media. The creator event disclosure guide shows why live and unscripted content needs the same claim control.

Build a claim table:

| Timestamp or frame | Claim ID | Status | Required action | |:---|:---|:---|:---| | 00:04 | C-01 | Supported with qualifier | Add approved on-screen condition | | 00:17 | C-02 | Outside evidence | Remove or escalate | | Caption line 2 | C-03 | Stale price comparison | Refresh market/date |

Approval of a brief does not cover a stronger creator paraphrase. Approval of a long video does not automatically cover a cutdown that removes the qualifier.

Separate evidence from mandatory disclosure

Several disclosures can appear in one ad, and each solves a different problem:

• Sponsor or material-connection disclosure explains the commercial relationship
• Expected-results disclosure addresses an atypical testimonial
• Qualifying language narrows a claim to what evidence supports
• Offer terms explain price, renewal, eligibility, or timing
• Synthetic-media labeling can identify generated or manipulated content

None is a substitute for support. Fine print cannot contradict a false headline, and #ad does not make an efficacy statement true.

Handle packet edge cases

Confidential study: Give qualified reviewers access and record the reviewed version; “confidential” is not evidence.

Supplier certificate: Verify what it proves. Ingredient identity does not establish consumer benefit.

Old formula: Add a hard stop when formulation or manufacturing materially changes.

Foreign study: Evaluate population and product relevance rather than rejecting or accepting solely by country.

Expert testimonial: Confirm qualifications and that the expert performed the evaluation ordinarily needed for the endorsement.

User-generated claim: Brand reuse can turn a customer statement into advertising; add it to the claim map.

No adequate evidence: Reject or materially narrow the claim. A post-publication research plan does not satisfy prior substantiation.

Version, retain, and refresh the packet

Store original evidence, expert notes, claim cards, final assets, approvals, publication markets, and corrections with immutable version identifiers. Set triggers for new evidence, safety signals, product changes, competitor changes, regulatory updates, translation, and new distribution.

“Claim substantiation packet” is an internal workflow term, not a regulator-mandated file format. Adapt it to risk and have qualified counsel and subject-matter experts decide sufficiency.

This is general U.S.-focused information, not legal, medical, or scientific advice. Evidence requirements are claim-, product-, audience-, and jurisdiction-specific.