Claim Review Methodology

The review layer for public health and wellness claims.

AuditGPT turns a public page into a structured claim record: what is being said, what support is visible, what an AI answer may repeat, and what should be rewritten or escalated.

Built around public substantiation discipline.

AuditGPT is not a law firm, regulator, clinical reviewer, or compliance certification. Our methodology is designed to help teams operationalize claim-support discipline around public-facing pages. It is informed by public business guidance such as the FTC's Health Products Compliance Guidance, which emphasizes truthful, non-misleading claims and appropriate substantiation for health-related advertising.

FTC Health Products Compliance Guidance
How the review works
01

Identify express and implied claims

We separate literal statements from the impression a reasonable buyer, patient, prospect, or AI answer engine may take away from the page.

02

Map visible substantiation

Each claim is checked against adjacent proof: studies, device references, provider credentials, citations, before/after context, testimonials, disclosures, and linked evidence.

03

Classify support strength

Claims are labeled supported, weakly supported, overstated, or unsupported. The label is a business review signal, not a legal conclusion.

04

Test AI Answer Reality

We examine how answer engines may summarize, qualify, ignore, or distort claims when public evidence is missing or buried.

05

Recommend safer framing

The report gives replacement language that preserves the marketing point while narrowing the claim to what the public support can actually carry.

06

Issue a dated receipt

Paid reports produce a Claim Intelligence Receipt and reviewed-badge summary showing what was reviewed, what was found, and what remains pending.

What we review in health, wellness, and med-spa pages.

Treatment outcome claims

Results, speed, permanence, safety, discomfort, downtime, recovery, and predictability.

Device and FDA language

Whether FDA-cleared, FDA-approved, clinical, medical-grade, or patented language is used precisely.

Testimonials and before/after content

Whether consumer stories imply typical results without the context needed to avoid overstatement.

Credentials and provider authority

Whether titles, licenses, training, and supervision claims are visible and current enough to support trust.

AI answer distortion

Whether AI systems are likely to repeat, soften, contradict, or ignore the brand claim.

Cleanup path

What can be fixed with copy edits, what needs proof, and what should be escalated for professional review.

The output is a receipt.

A Claim Intelligence Receipt records the review date, claim categories, support status, safer rewrite options, cleanup status, and next review date. It says reviewed, not approved.

The boundary is explicit.

AuditGPT does not certify compliance, make medical judgments, provide legal advice, guarantee ad approvals, or determine whether a claim is lawful. It creates a structured review artifact so teams can fix and escalate intelligently.

Review one public page first.

Start with a snapshot, then upgrade to a paid Claim Intelligence Report when you need a dated receipt and reviewed-badge summary.

Run Free Snapshot