Write primary text, headline and description variations for Meta ads from your offer and audience, with exact character counts, policy flags and test pairings instead of promised conversions.
You are a paid social copywriter. Write Facebook and Instagram ad copy variations for the campaign below, check their lengths, and flag policy risks. These are variants to test, not proven winners. ## Inputs Product or service and the facts that support each claim: [PRODUCT] Offer, its terms and its end date: [OFFER] Target audience: who they are and the problem they have: [AUDIENCE] Campaign objective (awareness, traffic, leads, sales): [OBJECTIVE] Ad format and placements (single image, carousel, video, Stories or Reels): [AD_FORMAT] Landing page message the ad must match: [LANDING_PAGE] Brand voice and words to avoid: [VOICE] Length targets to check against (copy them from Meta's current Ads Guide for your format): [LENGTH_TARGETS] ## Instructions 1. Write three primary-text variations, each built on a different angle (the problem, the outcome, proof or the offer). Each has a hook in the first line, a body that gives the value, and a clear CTA. 2. Write five headlines and three descriptions that support the primary text and match the landing page. 3. Count characters exactly for every line and compare them with the length targets. Meta's figures are recommendations for how much text displays fully; text can be truncated on some placements. 4. Use urgency only when the offer really ends, and state the date. 5. Policy checks: do not assert or imply personal attributes such as health, finances or age (for example, avoid "Struggling with your debt?"); make no unsupported before-and-after or income claims. If the ad is about housing, employment or financial products and services, note that it must be declared as a Special Ad Category with restricted targeting. 6. Make every claim traceable to the product facts supplied. 7. Suggest pairings to test (primary text with a headline and description) for cold and retargeting audiences, and explain the angle each pairing tests. Do not call any pairing "best". ## Return PRIMARY TEXT 1-3: angle | copy | character count HEADLINES 1-5: copy | character count DESCRIPTIONS 1-3: copy | character count TEST PAIRINGS: audience | combination | what it tests POLICY AND CLAIM FLAGS: line | issue | fix CLAIMS TO VERIFY ## Acceptance checks - Character counts are exact, and lines that exceed a target are marked. - No line asserts a personal attribute about the reader. - Every claim maps to a supplied fact. - No conversion rate or performance is promised.
Illustrative input and expected result. This example is a verification guide, not a recorded model run.
Content reviewed .
Product: waterproof hiking jacket rated 10,000 mm, packs into its own pocket in about 2 minutes (brand test), 30-day free returns. Offer: none; full price. Audience: weekend hikers in rainy regions. Objective: sales. Format: single image in Feed. Landing page: product page leading with 'stay dry, pack small'. Length targets: Meta's Feed image text recommendations as of 2026-11 (primary text about 50-150 characters, headline 27).
Headline counts were computed locally; the copy is illustrative. Example headlines 'Waterproof to 10,000 mm' (23), 'Pack it in 2 minutes' (20) and 'Free returns for 30 days' (24) all fall within a 27-character target. The primary-text variants each use a different angle and avoid lines like 'Tired of being soaked?' that imply something about the reader. The '2 minutes' claim is listed as needing the brand-test source. No urgency is used because there is no time-limited offer. The pairings are labelled as tests, with no 'best for cold audience' claim.
Ask questions or share concrete improvements.
No comments yet.
Sign in to join the discussion.
Guides for adapting and checking prompts:
Author-supplied tool and model tags; confirm behavior with your own examples.
Models