Why is insurance advertising regulated state by state?

Insurance advertising has no single rulebook. Under the McCarran-Ferguson Act, states regulate the business of insurance, so every state runs its own mix of statutes, regulations, and bulletins. The National Association of Insurance Commissioners (NAIC) publishes model laws to line those rules up. A model is a template, though. States adopt it in full, in part, or with their own changes.

For a national carrier, that creates a specific problem. One campaign can run in all 50 states. A disclosure that works in most of them can still fail in a few, and a failure in one state is still a violation. Add agents, affiliates, social channels, and a steady flow of product updates, and the volume gets ahead of any manual process.

This guide covers how the rules fit together, what nearly every state requires, where states diverge, why review breaks down at scale, and how to build a process that holds up in a market conduct exam.

How are state insurance advertising rules structured?

Think of four layers, each with a different legal weight.

• State statutes. Unfair trade practices acts, many built on NAIC Model #880, prohibit unfair and deceptive practices in the business of insurance, including misleading advertising.

• State advertising regulations. These carry the detail. Most track NAIC models by product line: the Advertisements of Accident and Sickness Insurance Model Regulation (#40) and the Advertisements of Life Insurance and Annuities Model Regulation (#570). Medicare supplement ads sit under separate rules. Model #40 expressly excludes them.

• Bulletins and guidance. Departments of insurance use bulletins to state expectations, then test them in market conduct exams. The NAIC's model bulletin on insurers' use of AI is the newest example.

• Federal rules on top. The FTC Act and the FTC Endorsement Guides reach endorsements and testimonials. CMS rules govern Medicare Advantage and Part D marketing. State insurance law does not replace any of this. It stacks.

The practical takeaway: the models tell you what regulators think good looks like. The state text tells you what you can be cited for. You need both.

What do nearly all states require of insurance advertising?

The details differ, but the core obligations repeat across Model #40, Model #570, and the state rules built on them. Section numbers below refer to the NAIC models. State versions are numbered differently, so check the local text.

• Truthful in fact and by implication. Ads are judged by overall impression, not by the fine print alone (Model #40 §5, Model #570 §4.A). Each piece of a multi-part ad has to comply on its own.

• Conspicuous, nearby disclosures. Required disclosures go next to the statement they qualify or under accurate captions. Model #40 §4 bars captions such as "Extent of Coverage." Model #570 §5.A bars minimizing or obscuring required information.

• Limits described plainly. Exceptions, limitations, and preexisting condition language cannot be minimized (Model #40 §6B–C). Limited-benefit policies carry prominent labels.

• Loaded words under watch. Model #40 §6 restricts terms such as "all," "full," "unlimited," "low cost," and "tax-free." Model #570 §4.B restricts "investment," "savings," and "retirement plan" where they could mislead.

• Clear product and insurer identity. The ad has to say it is insurance. Model #570 §6 requires the insurer's name, bars trade names that blur identity, and bars any suggestion of a government connection.

• Fair comparisons. No incomplete comparative advertising claims and no disparaging of other insurers or producers (Model #40 §8, Model #570 §5.V).

• Sourced, current statistics. Numbers must be relevant, recent, and attributed (Model #40 §10, Model #570 §5.R).

• Genuine testimonials. Covered in the table below and in the next section.

• Insurer control of all advertising. Model #40 §2B–C makes the insurer responsible for ads, including those created by agents.

• An advertising file and an annual certificate. Model #40 §18 calls for a file kept four years or until the next regular exam report, whichever is longer, and an officer-signed annual certificate. Model #570 §9 sets five years after use ends, plus an annual certificate.

Common claim types map to those rules as follows.

Where do state insurance advertising rules differ?

Three areas cause most of the trouble: filing triggers, testimonial mechanics, and newer topics the models do not settle.

Filing and pre-review

The models mostly avoid mandatory pre-approval. Model #40 §20 offers optional prior filing at least 30 days before use. Model #570 §9.B lets the commissioner require submission of ads that could mislead, and its drafting note urges states to reconsider pre-approval and consider exempting generic or branding ads. States then go their own ways.

The lesson: "file or don't file" is a per-state, per-product question. A tracker that records the answer for every product and state you advertise in earns its keep the first time an exam asks for proof.

Testimonials and endorsements

The models set the baseline, individual states add mechanics. California, in 10 CCR §2536.4, requires confirmation from the author if a testimonial is used more than a year after it was given, and requires a "Paid Endorsement" disclosure sized to match the endorser's name. Florida's rule adds licensing: a spokesperson who solicits insurance, advises on policies, or extends an invitation to contract is treated as soliciting and must be a licensed agent. Florida also carves out qualifying association endorsements that do not describe specific benefits, coverages, or premiums.

Both states say the insurer adopts a testimonial's statements as its own. If the creator says something wrong, the carrier owns it.

Lead generators and third-party marketing

In December 2022 the NAIC's Market Regulation and Consumer Affairs (D) Committee approved a request to reopen Model #880 to address improper marketing of health insurance. The proposed amendments would define lead generators, bind them to the unfair and deceptive practices prohibitions, require at least two years of marketing and complaint records, and add fines. A model amendment is a proposal until a state enacts it, so treat this as a direction of travel and confirm current status with the NAIC.

Artificial intelligence

The NAIC's model bulletin on insurers' use of AI asks carriers for a written AI program covering governance, risk management, and internal audit, and to vet third-party data and AI vendors. As of March 2025, 24 states had adopted it with little or no material change. That count has likely moved since. The bulletin targets AI that supports decisions about regulated insurance practices, so whether AI-generated marketing copy falls inside a given state's scope is a question for counsel.

Where does insurance advertising review break down at scale?

The gap is often process, not knowledge. The rule was known, but the check did not happen, or the evidence was missing when an examiner asked. These are the patterns to look for.

• Agent and producer content. The model rules make the insurer responsible for advertising, including material agents create (Model #40 §2B–C). A producer's social post or a customized flyer is still your ad if it promotes your product.

• Stale testimonials. A video that cleared review in 2024 may not clear in 2026. California and Florida both require reconfirmation after a year. Few teams track testimonial age.

• One asset, many states. Approving the strictest version and running it everywhere works until the strictest state differs by product line. Then you are guessing, or over-disclosing.

• Channel sprawl. Model #40 defines "advertisement" to reach print, broadcast, web, and electronic material, plus sales aids and prepared agent presentations. Short video, creator posts, landing pages, and email variants all count.

• Edits after approval. Each piece of a multi-part ad must comply independently, so a trimmed clip or a cropped graphic that drops a disclosure is a new problem.

• Lead generators and affiliates. Traffic partners write their own copy. If it markets your product, expect questions about what you knew and what you checked.

• Evidence gaps. Testimonials that cite benefits received need claim data on file. Ads need a file that shows distribution and form numbers. Reconstructing this during an exam is slow and visible.

• Live content nobody rechecks. The website, the social archive, and the PDF library hold years of copy written under older product terms and older rules.

What is the enforcement risk?

Risk comes from several directions at once. Departments can cite violations in market conduct exams. Some states can order corrective advertising at the insurer's expense, as New York's Part 215 does for ads that substantially mislead the public. An authorized officer signs the annual compliance certificate under the models, which makes the review process an officer-level concern. And the proposed lead generator amendments to Model #880 show where regulators want more accountability next. Fines in that proposal run from $1,000 per violation (capped at $100,000 in aggregate) up to $25,000 per violation for conscious disregard (capped at $250,000 in aggregate).

How do you build a defensible insurance advertising review process?

A defensible process shows what was checked, against which standard, by whom, and when.

1. Build one rule inventory by product line and state. Record the applicable statutes, regulations, and bulletins, the filing trigger, and the retention period for each. Date every entry and assign an owner for updates.

2. Classify every asset. Model #40 distinguishes institutional ads, invitations to inquire, and invitations to contract, and Texas asks for that classification on filing. The type decides which disclosures and filings apply.

3. Write a master standard with state overlays. Hold the baseline in one place: truthful, conspicuous disclosures, fair comparisons, genuine testimonials. Layer state-specific rules on top, such as filing triggers and testimonial reconfirmation windows.

4. Review at the draft stage and keep the evidence. Catch issues before design and legal rounds, not after. Keep an audit trail with each asset: the reviewer, the rule applied, the version, and the decision.

5. Route filings and track them. Tag assets that trigger a filing, record the date filed and the department's response, and block publication where a waiting period applies.

6. Monitor what is live. Rescan websites, social channels, and document libraries on a schedule. Track testimonial age, retire expired claims, and re-review anything when a rule or product term changes.

What does this look like for a single asset?

Take a hypothetical 30-second customer testimonial video for a life insurance product. At draft, the reviewer confirms the testimonial is genuine and tied to the advertised policy, that claim data is on file if it mentions benefits received, and that any paid arrangement is disclosed in the form each state requires. The asset is tagged with its type and its states. The approved version and its evidence are filed with its form number and distribution plan. A cut-down version for social media goes back through review, because dropping a disclosure creates a new violation. Twelve months later, the system flags the testimonial for reconfirmation before the ad runs again.

How does Haast help insurance marketing compliance teams?

Haast puts your compliance standards to work inside the content workflow. It checks content from drafting through approval to publication, and it continuously scans websites, social channels, and internal repositories to flag risks in content that is already live. Your team still owns the standards and the decisions. Haast does the repetitive checking, so reviewers spend time on judgment calls.

For insurers, that means the review steps described above can run on every draft and every live page, not only the assets that reach Legal. Rules by product line and state, testimonial and disclosure checks, and an evidence trail for each review sit in one place.

Zurich, a global insurer, uses Haast across its Australia and New Zealand business, where reviews of product disclosure statements, target market determinations, social content, and customer communications were manual and line by line. With Haast, review time for drafting target market determinations and product disclosure statements was cut in half. Drafting teams get instant compliance feedback, which reduces rounds with Legal, and ongoing monitoring flags non-compliant content before it becomes an issue.

Read the Zurich case study or book a product tour to see how Haast handles your state advertising rules.

FAQ

Do the NAIC model regulations apply directly to my ads?

No. They are templates. They apply only where a state has adopted them, and states often change them. Check each state's own text.

Do I have to file insurance ads with state regulators?

It depends on the state and the product. Texas requires Medicare supplement ads to reach the department no later than 60 days before first use, while Utah asks for Medicare supplement ads on the commissioner's request. Model #40 treats prior filing as optional. Keep a per-state tracker.

Is the insurer responsible for ads that agents and producers create?

Under Model #40 §2B–C, the insurer must keep control over all advertising, including material agents create. State versions vary, so confirm the wording in each state where your producers operate.

How long do we have to keep advertising records?

Model #40 says four years or until the next regular examination report, whichever is longer. Model #570 says five years after the ad's use ends. State periods differ, and some require claim data behind testimonials to be kept as well.

Do testimonial rules apply to social posts and creator content?

Generally yes, where the content promotes your product. Model #40 defines advertisement to include web and electronic material, and state testimonial rules ask for genuine, current, accurately reproduced statements and clear disclosure of payment. The FTC Endorsement Guides apply on top.

Does AI-generated marketing copy trigger extra obligations?

The NAIC model bulletin asks insurers to run a written program for governing AI that supports decisions about regulated insurance practices. Whether marketing content is in scope depends on the state. Ask counsel, and keep records of how AI-assisted content is reviewed either way.