Video Ad Approval: The Meta and TikTok Policy Traps That Get You Rejected

The advice you usually hear about video ad approval is to "follow the policies," as if the rejection problem were a reading-comprehension failure. It isn't. The advertisers who get rejected most have usually read the policies; what they haven't done is model how the review systems actually score a video. Approval is not a content question, it is a classification question, and the difference matters because both platforms grade four signals at once and reject on any single one.

Look at the data and a clear shape emerges. Roughly four out of five rejections cluster into the same handful of triggers, and they are predictable enough to engineer around before you ever hit submit. To get a video ad approved on the first pass, you stop thinking about whether the ad is "good" and start thinking about which classifier each line, frame, and destination URL is going to trip. This piece walks the triggers in order of how often they fire.

Why is my video ad rejected on Meta or TikTok?

The generic "ad does not comply with our policies" notice maps to a short list almost every time: personal-attributes phrasing in the hook, an outcome claim such as "cure" or "guaranteed," a before/after or body-part visual, a landing page that contradicts the ad, or an undeclared restricted category. DeepClick's analysis finds that about 80% of rejections fall into the five primary buckets, with landing-page mismatch the most common and the most overlooked. The rest of this article is those buckets, ranked. If you want the fast version, skip to the pre-flight checklist near the end and run it against your next upload.

The biggest video ad approval blocker isn't your video

Advertisers spend their attention on the creative and ignore the destination. The review systems weight it the other way. Both Meta and TikTok run multimodal review that reads the footage, the on-screen text via OCR, the spoken audio, and the landing page as one bundle, then scores them for consistency.

DeepClick's 2026 TikTok breakdown puts landing-page mismatch first: about 40% of what advertisers report as "creative" rejections are actually landing page issues. TikTok's own policy names clickbaiting, defined as ad messaging that differs from the landing page, as a primary rejection trigger.

Meta scores the same dimension. As of Q1 2026, a headline claim has to appear on the landing page, pricing and discounts have to match the ad's promise, and the page cannot bury users in pop-ups or autoplay sound. Meta's classifier checks landing page content alignment alongside the ad's OCR and copy. Before you submit, put the ad and its destination side by side and verify five things:

The exact claim or offer spoken or shown in the video appears in text on the landing page.
The price and any discount in the ad match the price on the page. No "50% off" in the ad and full price on the page.
The product on screen is the product the page sells: same SKU, same outcome.
The page is mobile-optimized and does not auto-play sound or trap the user in pop-ups.
No benefit lives only in the video; every on-camera claim is substantiated in writing on the page.

This is the cheapest fix in the whole process and it removes the single largest failure category. When you test many variants it is why ad-to-landing-page match deserves its own pass rather than a glance.

Personal attributes: the trap that needs no claim

This rejection confuses people because the ad makes no health or product claim at all. Meta's Personal Attributes Policy prohibits copy that directly or indirectly asserts or implies a user's personal attributes: health, body image, age, financial situation, and more.

The mechanism is the second person. The moment your copy implies it knows something about the viewer, it trips. "Are you 45 and overweight?" is the textbook violation. In 2026 Meta widened the rule to catch indirect and conditional phrasing too: "For people dealing with financial challenges," "If you've been diagnosed with anxiety," even a sympathetic "We understand your pain" now gets flagged.

The fix is to stop addressing the viewer's condition and describe what the product offers instead. Gripas Marketing's compliant rewrites are a clean pattern to copy:

Rejected: "Are you suffering from anxiety?" to Compliant: "Explore resources designed to support calm, focus, and mental wellness."
Rejected: "Embarrassed by your weight?" to Compliant: "Discover fitness guidance designed to support a healthier, more active lifestyle."
Rejected: "Do you have bad skin?" to Compliant: "Discover skincare solutions designed to support a smoother, healthier-looking appearance."
Rejected: "Are you overweight and tired?" to Compliant: "Build healthier habits with guided fitness and nutrition support designed for long-term progress."

The reliable safe openers, per the same source, are "Explore resources for," "Learn practical ways to," and "Discover solutions designed for." Accelerated Digital Media adds that command phrasing like "Conquer your depression" is restricted too, so the rule reaches imperative voice, not only questions. For video this matters more than most people expect: the policy reads your spoken hook and your on-screen captions, not just the ad headline. A voiceover that says "tired of being out of shape?" fails exactly the way the headline would, so script the hook as if a transcript is being scored. If you work from a library of hook openers, audit the whole set against this rule once and you stop tripping it.

Before/after visuals: the most-pulled creative pattern

Side-by-side transformation imagery is the most common visual rejection in health, fitness, and beauty. Meta explicitly bans before-and-after weight-loss transformations, and the same ban covers cosmetic procedures: wrinkle treatments like Botox, dermal fillers, and any anti-aging treatment.

The stated rationale is negative self-perception, so the ban reaches past the split-screen to close-ups of body parts that reinforce insecurities, the classic "pinching fat" shot, and any framing that exploits insecurity to sell a beauty standard. Procedural footage such as needles in skin or invasive surgery is out as well. TikTok is stricter on body image: its weight-management policy prohibits suggesting an "ideal body type" exists, shaming users about their bodies, and any claim that ties appearance to "confidence, self-esteem, emotional wellbeing, desirability, popularity, or social standing." Its healthcare policy bans before-and-after comparisons for medicines and supplements outright.

You do not have to abandon proof. You change its form. The compliant substitutes drawn from the sources are:

Show usage, not the transformation. Film the product in use, reference progress "in a neutral way" and mention the time taken to see results rather than staging a reveal.
Lifestyle over body parts. Confident-person lifestyle imagery, testimonials in quote or video form, and educational content are permitted with disclaimers.
Narrate the mechanism. Explain what the product does and what is in it instead of cutting to a result.

Two exceptions are worth knowing. Meta carves out fitness services like Pilates, where transformations are allowed, and the restrictions apply to paid ads only. Organic timeline posts are exempt, which is why a brand's feed can show before/afters that would never clear ad review. To build credibility without the banned visuals, the patterns in social proof in video ads and testimonial-style ads without real customers yet map directly onto this constraint.

Claim language: trade outcomes for process

The fastest way to get a health, supplement, or beauty ad rejected is to promise an outcome. Both platforms block the verbs of medical intervention and guaranteed results. On Meta, the banned vocabulary includes "cure," "treat," "heal," "fix"; "guaranteed results," "100% effective"; and "instant relief," "overnight transformation," "results in 24 hours." Also flagged are "diagnose," "symptoms," and "diseases" from non-medical advertisers; superlatives like "most effective" or "strongest formula"; and pseudo-mechanism claims like "boosting metabolism," "cleansing toxins," "balancing hormones." TikTok's healthcare policy bans the same outcome verbs across all markets ("treat, cure, heal, or prevent any medical condition"), plus "miracle" or "secret" remedies and any claim that serious conditions such as cancer, HIV, or COVID-19 are treatable.

The substitution that passes is process-and-ingredient language. Compliant rewrites drawn from AdAmigo and The Graygency:

"Guaranteed to reduce joint pain in 30 days" to "many users noticed improved comfort"
"Treats insomnia" to "perfect for those enhancing their nighttime routine"
"Boosts immunity" to naming the specific ingredient and what it is
"Treats ADHD" to "improves focus"
"Cures rosacea" to "reduces redness"
"Beat obesity" to "boosts overall wellbeing"

What is actively permitted is references to clinical studies framed as "may help manage," "contains clinically evaluated ingredients," "historically used to support," and genuine FDA-approval statements. One 2026 caveat: supplement and wellness brands using "clinically proven," "doctor recommended," or "pharmaceutical grade" now need LegitScript certification or must drop to general wellness language. The discipline of substantiating every claim is the same one that anchors any working ad-script writing framework.

TikTok weight management: a stricter, separate rulebook

Anything weight-adjacent hits a dedicated TikTok policy that goes beyond Meta's. It completely prohibits:

References to unrealistic muscle gain or weight loss.
Claims that the product alone, without diet or exercise, leads to weight loss.
Suggestions that losing weight or gaining muscle is easy or guaranteed.
Tying body change to "improved life circumstances or self-image."

Entire product types are banned, not just claims: ingestible fat-burning products, fasting products, waist trainers, face and body sculpting products, and weight-loss injections or surgery. Where weight-loss supplements or meal replacements are allowed at all, it is only in select regions with TikTok sales-rep approval, and any permitted weight or muscle claim must be restricted to 18+ audiences and framed through "the promotion of a healthy lifestyle." The operating rule that keeps you clear of this policy: the product is never the sole cause of the result, never easy, and never tied to who the viewer becomes. Confirm your product type is not banned outright before you bother scripting it.

Restricted categories: the ones that need paperwork

Some verticals are not rejected for wording. They are rejected for a missing certification, and no amount of clever creative substitutes for the credential. On Meta the gated and special categories are:

Online pharmacies must be LegitScript certified.
Health verticals cannot optimize for lower-funnel events. Meta restricts or blocks "Purchase" and "Add to Cart" tracking for health brands and disables custom audiences with health-implied names such as "arthritis_interest_list" and conversions carrying metadata like "appointment booked."
Special Ad Categories (Housing, Employment, Credit, Politics) carry mandatory targeting limits. Per inBeat: no ZIP code targeting, all genders required, age fixed at 18 to 65+, location by country, region, or city only with a required 15-mile (25 km) radius in the US and 10-mile (17 km) in Europe, and no detailed targeting or lookalike expansion. AuditSocials adds that HEC lookalikes are capped at a minimum 1% similarity threshold. Declaring the category up front is mandatory; running these as standard ads is one of Meta's seven recurring disapproval patterns.

TikTok requires formal certification for crypto and digital assets, fintech, alt-pharma and regulated supplements, sweepstakes, alcohol, and dating. Without it, gray-niche verticals face a 40 to 60% first-pass rejection rate, and certification runs 7 to 21 business days. That is a lead-time problem you plan around, not a same-day fix. The specific gates: prescription medicines are banned almost everywhere; OTC medicines, medical devices (which need CE or UKCA marks), and pharmacies need regulatory proof; invasive cosmetic procedures are prohibited in most markets, hair transplants excepted, and Botox is banned in many European regions. Alcohol brands need explicit TikTok permission and a Registered Business Account, and CBD/hemp needs rep approval and excludes ingestibles.

Account quality score: the cost rejections leave behind

A single rejection is cheap. The hidden expense is what it does to your account quality score, sometimes called account health, which both platforms keep at the account level and which governs how every future ad is treated. DeepClick reports the Meta thresholds clearly:

Below 70: stricter scrutiny on everything you submit.
Below 50: manual-review delays that stretch approval times.
Below 30: spending limits and reduced reach.

AuditSocials notes that repeated rejections can escalate to delivery restrictions, account disablement, or forced identity verification. This is why a sloppy batch costs more than the creatives in it: a low score taxes the approval speed and reach of the clean ads you upload next. Two behaviors accelerate the decline, and both are avoidable. Do not make rapid consecutive edits to a rejected ad, because Meta treats that as a review-circumvention signal that hurts standing. And do not file a vague appeal; DeepClick finds vague appeals get auto-denied while specific ones that cite the exact policy resolve in 48 to 72 hours. Fix the root cause, then resubmit a clean variant or appeal with the precise citation.

A pre-flight checklist you can run before every upload

Bake this in so it runs once per creative, fast. It is faster than one appeal cycle and it protects the account score that governs everything else.

Read the hook as a transcript. Does any line, spoken or captioned, imply you know the viewer's health, body, age, or finances? If so, swap to "Explore / Discover / Learn practical ways to."
Scan for outcome verbs. Cure, treat, heal, fix, guaranteed, instant, 100%, overnight, transforms. Replace with "may help," "supports," "contains clinically evaluated ingredients."
Check every visual. No side-by-side before/after unless it is a Meta fitness service, no body-part close-ups, no needles or procedures, no "ideal body" framing.
Match the landing page. Every claim, price, and discount in the ad must appear on the destination. This step alone removes your biggest rejection bucket.
Confirm category and certification. Is this a Special Ad Category, or gated nutra, crypto, fintech, or alcohol? Declare it or hold the cert, and budget the 7 to 21 day lead time on TikTok.
Age-gate health and weight content to 18+ (25+ for prescriptions in New Zealand, per TikTok), and confirm you are not targeting minors.
Label AI voice cloning. If your voiceover clones an identifiable person, TikTok requires the AIGC label or a visible caption.

How rejections tax everything you ship next

Review time is the variable most advertisers underweight. Meta's standard review runs 24 to 48 hours, stretching to 5 business days in high-volume periods; TikTok runs from 2 to 6 hours for established clean accounts to 48 to 72 for gray niches. Every rejection resets that clock and adds an appeal cycle on top of it.

For a solo founder running several products, or an agency carrying a dozen accounts, the implication is concrete. If a compliant variant takes minutes to produce and gets pre-cleared against the checklist above, the approval pipeline shifts from a multi-day gate to background latency. You can ship a fresh creative per ad set, run more hooks per test, and refill the queue the moment something fatigues, instead of rationing creative because each one risks a stall. The case for that throughput is in why iteration speed compounds and how many ads you should actually run, and it is the same economics that lets agencies scale client volume without adding headcount. It only holds if compliance is cheap and built into production rather than discovered on a Friday.

FAQ

Why did my video ad get rejected with no specific reason?

The generic "doesn't comply" message almost always maps to a few triggers: personal-attributes phrasing in the hook, an outcome claim ("cure," "guaranteed," "instant"), a before/after or body-part visual, a landing page that doesn't match the ad, or an undeclared restricted category. Run the seven-step checklist; in practice about 80% of rejections fall into the five primary buckets, with landing-page mismatch the most common and most overlooked.

Can I use before-and-after images in video ads at all?

On paid Meta ads, no for weight loss and anti-aging or cosmetic treatments; side-by-side transformations are banned, with a narrow exception for fitness services like Pilates. TikTok bans before/after for medicines and supplements. Show product usage, lifestyle imagery, or testimonials instead. The ban applies only to paid Meta ads; organic posts are exempt.

How long does video ad approval take on Meta and TikTok?

Meta's most automated reviews finish in 1 to 4 hours; human-flagged ads take 24 to 48 hours and restricted verticals up to 72. TikTok new advertisers see 12 to 24 hours for e-commerce and 48 to 72 for gray niches; established clean accounts get 2 to 6 hours. Certification for gated categories is a separate 7 to 21 business day process.

What words automatically get health and supplement ads rejected?

Outcome and intervention verbs are the worst offenders: cure, treat, heal, fix, guaranteed, 100% effective, instant relief, overnight transformation. Also flagged are pseudo-mechanism claims (boosting metabolism, cleansing toxins, balancing hormones) and unqualified superlatives. Replace them with process language such as "may help manage," "supports," "contains clinically evaluated ingredients."

Do I need a certificate to run crypto, supplement, or alcohol ads on TikTok?

Yes. TikTok gates crypto and digital assets, fintech, regulated supplements, sweepstakes, alcohol, and dating behind formal certification. Without it, those verticals face 40 to 60% first-pass rejection, and certification takes 7 to 21 business days, so plan that lead time before launch rather than after the first batch fails.

Sources

Passing review is a checklist, and the expensive part is running it across dozens of variants by hand. Aitachyon makes the production side cheap enough that reframing a flagged hook, swapping a banned visual, and resubmitting a clean variant takes minutes instead of a re-edit, whether you are a founder running several products or an agency scaling client volume. When compliance is the only constraint left, you can fix it before an editor would have replied to your email.

Video Ad Approval: The Meta and TikTok Policy Traps That Get You Rejected