Influencer and Endorsement Disclosures Under FTC Advertising Rules

If you were paid, gifted, discounted, employed, related to the founder, or hold shares in the company, say so — in the post itself, where an ordinary reader will actually see it. That is the whole of the Federal Trade Commission's endorsement rule in one sentence. The FTC's Endorsement Guides, revised in 2023, treat any material connection between an endorser and a brand as something the audience is entitled to know. A hashtag wall, a link labelled "more", or a note in a video description does not get there. The test is noticeability, not paperwork.

The single test every disclosure has to pass

Ask one question about any post: would an ordinary member of the audience notice the disclosure without hunting for it? If the honest answer is no, it fails, however correctly worded it is.

The FTC calls this standard clear and conspicuous, and it has a practical shape: unavoidable rather than optional, in the same medium as the claim — audio claims get audio disclosures, visual claims get visual ones — and placed near the endorsement rather than in a block at the end.

Wording matters less than placement, but it still matters. Plain words such as "ad", "sponsored" or "[brand] sent me this free" are understood; vague thanks and invented shorthand are not. The FTC's advertising and marketing guidance for businesses is the primary reference, and its consumer-facing site shows the same rules from the audience's side.

Watch out: A platform's built-in "paid partnership" toggle does not by itself satisfy the requirement. Those labels can be small, greyed, cropped out of reposts, or absent off-platform. Treat them as a supplement to your own disclosure, never a replacement.

What counts as a material connection

People assume the rule is about cash. It is broader: a material connection is any unexpected relationship that might affect how much weight the audience gives the opinion.

• Payment of any kind: flat fees, per-post rates, bonuses tied to sales.
• Free product, loaned equipment, review samples, or anything the creator keeps.
• Discount codes, comped travel, event access, meals, or accommodation.
• Affiliate links and commission arrangements, including small percentages.
• Employment, contractor status, or a role at an agency working for the brand.
• Family, household, or close personal ties to someone at the company.
• Equity, options, or any other financial stake in the business being praised.
• Entry to a contest or giveaway conditioned on posting.

Two further duties sit alongside disclosure. An endorsement must reflect honest opinion and actual experience — you cannot endorse something you never used. And any objective claim inside it needs the substantiation the advertiser would need to make the claim directly: "cleared my skin in a week" imports a health claim, and the burden of proof travels with it.

The FTC also has a rule aimed at fake and deceptive reviews and testimonials. As of mid-2026 those two instruments together are the working federal framework, reaching fabricated reviews, suppressed negative feedback and testimonials from people who never used the product. Where a genuine bad review ends and an actionable false statement begins is a separate problem, covered in our piece on online reviews, opinion, and platform issues.

Placement, medium by medium

Because the standard is noticeability, the right answer changes with the format — and the failures repeat.

Reposts deserve attention too. When a brand lifts a creator's video onto its own channel, the disclosure has to survive the move — cropped aspect ratios and stripped captions turn a compliant post into a non-compliant one. The same discipline applies to email, where a separate federal statute adds its own labelling duties, explained in our guide to the CAN-SPAM rules senders overlook.

Who carries the exposure

The creator is not the only party on the hook. The framework reaches the advertiser whose product is promoted and the intermediary — the agency, network or management company that arranged and paid for the placement. An advertiser that scripts a post, approves it and pays for it does not become a bystander because someone else pressed publish.

In practice brands are expected to run a program, not a hope: disclosure instructions in the contract, training on what a compliant post looks like, spot-checks of what went live, and a documented process for fixing posts that fall short. The FTC's business guidance hub is where a compliance team would start.

Contracts do not shift that duty. A clause requiring the creator to disclose helps with monitoring; it does not answer a regulator asking why the campaign ran unlabelled. Whether the clause binds the creator at all is a formation question, discussed in our analysis of when an online agreement binds.

Synthetic and AI-generated endorsers

Generative tools raise the same disclosure question plus a second one. A synthesised voice or face endorsing a product carries an added layer of state rules on likeness and digital replicas, still developing and varying widely; we track that in our piece on emerging state rules on synthetic media and likeness. The federal analysis does not change: if the audience is led to believe a real person tried the product, that has to be true.

The federal layer and the state layer

Endorsement disclosure is federal law, administered by the FTC, and it applies across the country. That is the baseline every campaign has to meet.

States add a second layer. Every state has some form of unfair and deceptive acts and practices statute — UDAP for short — letting a state attorney general pursue deceptive advertising within the state, and in many states giving private consumers a route to sue that federal FTC enforcement does not. California, for example, has a well-developed body of consumer-protection law that state regulators use against advertising practices independently of anything the FTC does. Some states go further and regulate advertising in specific sectors — health services, legal services, financial products, alcohol, cannabis — so a campaign that clears the federal standard can still breach a state rule for its industry.

One more federal regulator is worth knowing about. The Federal Communications Commission administers separate sponsorship identification requirements for broadcast and cable programming — a different regime with different mechanics, and a media plan spanning social and broadcast has to satisfy both.

Note: Using a brand's name or logo is a trademark question, not a disclosure question, and is judged under different rules. The USPTO's trademark pages are the starting point for that separate analysis.

Common questions

Is #ad at the end of a caption enough?

Usually not. Placement decides it, and the end of a caption is often below the fold, mixed into hashtags, or hidden behind a "more" link. Put the label at the start of the caption, where it shows without expanding the post, and add it on screen in any video. The word "ad" is fine — the problem is where it sits, not the wording.

I only got a free sample, so is there really anything to disclose?

Yes. Free product is a material connection even when no money changes hands and even when the creator dislikes the item and says so. Audiences weigh an opinion differently once they know it arrived at no cost. The same applies to loaned equipment, comped travel, event tickets, and discount codes given to people who post about the brand.

Can the brand be blamed when the creator forgets to label a post?

Yes, and this surprises marketing teams. The framework reaches the advertiser and the agency that arranged the placement, not only the person who published it. A contract requiring disclosure is useful but does not transfer the duty. Regulators look for an actual program: written instructions, training, monitoring of what went live, and a record of corrections when a post ran unlabelled.

Does an employee posting about their own employer need to say anything?

An employment relationship is a material connection, so yes — praise for your employer's product on a personal account normally has to make the connection clear. Many companies handle this with a short social media policy naming an approved form of words. Trade groups and paid ambassadors are treated the same way. The question is whether an ordinary reader would guess the relationship, and usually not.

Putting a disclosure practice in place

1. List every relationship first. Map who receives money, product, codes, equity or access. Everything on that list needs disclosing somewhere.
2. Write the words once. Agree a short approved label per format so creators are not improvising at publish time.
3. Fix placement per format. Decide in advance where the label sits in each format, using the table above as a starting point.
4. Check the claims separately. Any performance, health or comparison claim needs substantiation held by the advertiser before the post goes out.
5. Review what actually published. Look at the live post on a phone, not the draft, and check the label survives cropping and reposting.
6. Keep the record. Save the brief, the instructions sent, screenshots of the live post, and any correction made afterwards.

Practical step: Run one existing campaign through this list before writing a policy. Auditing real posts exposes the two or three placement habits causing most of the problem far faster than drafting rules in the abstract.