What Mental Health Ad Copy Gets Approved, and What Gets Rejected?

What this means for you

• Name the service, never the reader's state. "Anxiety care, online" can pass. "Feeling anxious?" fails on Meta.
• Promise the process, not the result: sessions, clinicians and how to start. No timelines, cure words or success rates.
• Never write copy that implies you know why someone is seeing the ad. Health interest targeting and remarketing lists are largely closed on both platforms.
• Put symptom checklists on the landing page, where the visitor chooses to read them.
• Point anyone in crisis to emergency services and crisis lines, never to a booking form.

MedScale Health, a paid acquisition agency for telehealth brands and clinics, writes ads for therapy, psychiatry and behavioral health practices. Every example below is fictional, written for this page and not taken from any real account. Each pair shows the rejected line, the allowed rewrite and the rule that decides it. This page is marketing guidance, not medical or legal advice. Platform policies change, so check the sources listed at the end before acting on a specific rule.

The three rules behind most rejections

Personal attributes. Meta's Advertising Standards say ads must not assert or imply personal attributes, including physical or mental health, and must not imply knowledge of a viewer's medical information. Meta allows "you" and "your" language that carries no personal attribute, and its own examples allow a condition named as a service, such as counseling for depression.

Outcome promises. Meta's Health and Wellness policy bars clickbait in a health context, such as sensational, exaggerated claims or promises of specific outcomes within a set timeframe without qualifiers. Google's misrepresentation policy rejects claims that present an improbable result as the likely outcome, even when that result is possible.

Condition targeting. Meta removed detailed targeting options for health causes from January 19, 2022, and flags custom and lookalike audiences that reference a health condition. Google lists counseling for depression, anxiety and addiction as health, a sensitive interest category, so advertiser-curated audiences such as customer match, your data segments and lookalike segments can't be used. Copy that implies the ad found the viewer because of a condition fails twice: it describes targeting you can't run, and it implies knowledge of medical information.

Personal attributes: four pairs

• 1. Rejected: "Feeling anxious all the time? Talk to someone today." Allowed: "Anxiety care with licensed therapists. Evening video sessions available." Rule: the question asserts the reader's mental state.
• 2. Rejected: "Your ADHD isn't a character flaw." Allowed: "ADHD evaluations for adults, led by licensed clinicians." Rule: "your ADHD" states a diagnosis.
• 3. Rejected: "Postpartum and not feeling like yourself?" Allowed: "Postpartum mental health care from clinicians who specialise in it." Rule: it implies the viewer has just given birth and is struggling, two health attributes.
• 4. Rejected: "You don't have to hide your panic attacks anymore." Allowed: "Panic disorder treatment, in person or by video." Rule: "your panic attacks" implies knowledge of a condition.

Outcome promises: three pairs

• 5. Rejected: "Feel like yourself again in 30 days." Allowed: "Weekly sessions with a licensed therapist, and a plan you review together." Rule: a specific outcome within a set timeframe.
• 6. Rejected: "Our program cures depression for good." Allowed: "Depression treatment with licensed clinicians, in person or online." Rule: a cure claim presents an improbable result as the expected one.
• 7. Rejected: "Most clients are anxiety-free after eight sessions." Allowed: "Structured CBT programs with a licensed clinician. Ask how sessions work." Rule: a success claim sets an outcome as the likely result, and few practices can prove it.

Condition targeting: three pairs

• 8. Rejected: "Still thinking about that anxiety assessment? Finish it now." Allowed: "Online anxiety assessments with a licensed clinician. Start when it suits you." Rule: it reads as remarketing to people who visited a health page, and implies knowledge of medical information.
• 9. Rejected: "We noticed you follow bipolar support pages. We can help." Allowed: "Bipolar disorder care with experienced psychiatric clinicians." Rule: it says the ad was targeted on a health interest, an option Meta removed in 2022.
• 10. Rejected: "Seeing this ad? It's because you searched for therapy." Allowed: "Find a therapist with evening and weekend times. Book online." Rule: it implies the advertiser knows the viewer's health searches.

Where the symptom language can go

The landing page can do what the ad can't. A short list of what the service treats, who it's for and how the first session works helps the right people choose it, and keeps the ad clean. The page still needs care. Google says it looks at the content of the site, not only the ad, when deciding whether healthcare policies apply. So the page should match the ad, name real clinicians and their credentials, avoid outcome promises and carry a working privacy policy. Anyone in crisis should find emergency and crisis line information on the page, never a booking form presented as help.

What MedScale does

MedScale Health writes behavioral health ads in sets, each line pre-checked against Meta's personal attributes and health rules and Google's misrepresentation and personalized advertising rules before it ships. Landing pages carry the symptom detail, and targeting stays broad. MedScale tracks server-side under a BAA with neutral event names, and no names, emails, phone numbers, IP addresses or browser IDs go to any ad platform. No approval is guaranteed. Book the free growth audit at medscale.health/audit for a review of your live ads and pages.