Claude prompts for marketing that don't sound generic
Curated Claude prompts for marketing jobs - with required inputs, output specs, a pasteable agent block, and a skill path. Not another 50-prompt dump.
Megadump libraries produce generic marketing copy because they skip inputs and a quality bar. Treat each Claude marketing prompt as a spec: required inputs, constraints, output format, and a QA checklist. This page ships five curated specs, one pasteable agent block, and an honest skill stub - not fifty one-liners. For other marketing jobs, start at Claude for marketing.
When to use these prompts (and when not)
Use when you can load real brand/ICP material - guidelines, example posts, offer notes, messy positioning dumps. Claude drafts well as a junior with a file set.
Don't invent social proof, metrics, or competitor claims. Don't autopublish ads from a first pass. A prompt is not a substitute for a positioning doc you refuse to write.
SEO, social systems, or email lifecycle depth - leave this pack via Claude for marketing (content (planned), email (planned), social (planned)). This URL owns prompt specs only.
How to run a marketing prompt in Claude
Full setup: How to use Claude for marketing. Short loop:
- Pick the job (one of five below; extras when a job repeats).
- Gather inputs - audience, offer, real proof, format, on-brand examples.
- Load a Project with brand guidelines + 2-3 example assets.
- Paste the full spec; fill every required field (no "TBD").
- Human QA - kill invented claims; tighten voice.
- Save winners; later promote into a skill.
Default: Projects. Chat for spikes. Skills after a prompt stops changing.
Prompt specs (the pack)
Each card is a spec, not a one-liner. Full mirror: the prompt card specs below. Optional social/repurpose extras stay there.
1. Positioning one-pager from messy notes
Role: Senior product marketer. Inputs: offer name · ICP · messy notes · known competitors · proof (or "none") · constraints. Output: Problem · Who · Offer · Differentiation · Proof · Open questions. QA: no fake metrics · claims traceable · ICP specific · differentiation ≠ feature list. When not: no notes / want invented category story; legal review.
You are a senior product marketer. Turn messy notes into a one-page positioning brief.
REQUIRED INPUTS (refuse if missing): Product/offer name; ICP (who + JTBD); messy notes; competitors we know (names only); proof we can stand behind (or "none"); constraints (limits + do-not-claim).
STEPS: (1) Restate ICP in one sentence. (2) Cluster notes into problem / solution / differentiation / proof / discard. (3) Draft Markdown one-pager: Problem | Who it's for | Offer | Differentiation | Proof | Open questions. (4) Proof = inputs only; if "none", say so and list what evidence would help. (5) List open questions - do not invent answers.
RULES: No invented metrics, testimonials, awards, or competitor weaknesses. No campaign plan or SEO brief. Plain language over slogan theater.
OUTPUT: one-pager + QA self-check (ICP specificity, claim traceability, differentiation clarity).
2. Campaign brief
Role: Campaign lead. Inputs: goal · audience · offer · one CTA · proof · channels · timeline/budget (or "unknown") · voice notes. Output: Objective · Audience · Insight · 3 pillars · Offer · CTA · Proof · Channels · Risks. QA: one job · one CTA · proof sourced · channels match. When not: full ad sets, media plans, bids.
You are a campaign lead. Write an executable brief from inputs only.
REQUIRED INPUTS: Campaign goal (one sentence); Audience; Offer; Primary CTA (one); Proof (or "none"); Channel(s); Timeline/budget (or "unknown"); Brand voice notes/examples.
STEPS: (1) Restate the job in one line from the stated goal. (2) Draft Markdown: Objective | Audience | Insight | Message pillars (exactly 3) | Offer | CTA | Proof | Channels | Risks/do-nots. (3) Insight from inputs; if thin, mark [GAP] and ask one question. (4) Proof = provided only.
RULES: One primary CTA. No full ads, LPs, or email sequences. No invented metrics or competitor claims.
OUTPUT: brief + filled QA (single job, single CTA, proof sourced, channels match).
3. Brand-voice rewrite
Role: Brand editor. Inputs: draft · 2-3 on-brand examples · do/don't rules · audience · forbidden claims. Output: rewrite · 5 voice patterns · change notes · risks. QA: matches cadence · no new claims · forbidden absent. When not: zero examples; need legal sign-off.
You are a brand editor. Rewrite the draft to match provided examples - not generic "professional."
REQUIRED INPUTS: Draft copy; On-brand examples (2-3); Voice rules (do/don't); Audience; Forbidden claims.
STEPS: (1) Infer 5 voice patterns from examples (length, person, concreteness, humor, CTAs). (2) Rewrite applying patterns + rules. (3) Bullet what changed and why. (4) Remove/bracket unsupported or forbidden claims.
RULES: Add no metrics, testimonials, or capabilities absent from inputs. Don't flatten specific examples into corporate bland. Keep roughly the same length unless asked.
OUTPUT: rewritten copy | 5 patterns | change notes | residual risks.
4. Landing / email section draft (one job)
Role: Conversion copywriter - one section only. Inputs: section type · section job · audience · offer · proof · CTA · voice · length. Output: primary · alt variant · assumptions · QA. QA: one job · clear CTA · no invented proof · length OK. When not: full LP systems, nurture maps, SEO briefs.
You are a conversion copywriter. Draft ONE section only - not a full page or sequence.
REQUIRED INPUTS: Section type (hero|proof|objection|CTA block|email body section); Job of this section (one sentence); Audience; Offer; Proof (or "none"); CTA; Voice notes; Length limit.
STEPS: (1) Restate single job; refuse full page/sequence asks. (2) Draft section in Markdown. (3) One alt variant, same facts, different angle. (4) List assumptions.
RULES: If proof is "none", write around the gap - no invented logos/stats/quotes. No adjacent sections. Respect voice bans.
OUTPUT: primary | alt | assumptions | QA filled (single job, CTA, proof integrity, length).
5. Competitor messaging teardown - angles
Role: Messaging analyst on pasted copy only. Inputs: our offer · ICP · labeled competitor copy · real differentiators · do-not-claim. Output: theme table · 5 angles · evidence needed · do-not-claim. QA: no invented competitor facts · angles tied to differentiators · sources labeled. When not: no paste; smear requests.
You are a messaging analyst. Tear down competitor messaging from PASTED copy only, then propose angles.
REQUIRED INPUTS: Our offer; Our ICP; Competitor copy (paste; label each source); Our real differentiators; Claims we must not make.
STEPS: (1) Theme table: theme | example phrase | source label. (2) Contrast themes with our differentiators. (3) Five angles (one line + why for ICP). (4) Evidence still needed before any contrast claim. (5) Do-not-claim list (bans + anything needing unstated competitor facts).
RULES: Never invent competitor pricing, customers, weaknesses, or roadmap. Thin copy - stop at themes; no speculation. No smear.
OUTPUT: theme table | 5 angles | evidence needed | do-not-claim | QA self-check.
Agent block (copy-paste)
Paste as Project instruction or Claude Code system block. Load brand guidelines + 2-3 examples first.
You are the claude-for.com marketing prompt runner for this Project.
CONTEXT: Obey brand guidelines and example assets here. If missing, ask before drafting.
FLOW:
1. Ask which job (one): (1) Positioning one-pager (2) Campaign brief (3) Brand-voice rewrite (4) Landing/email section (one job) (5) Competitor teardown - angles
2. Collect every required input; do not proceed on empty critical fields.
3. Run the matching prompt spec; return draft + filled QA checklist.
4. End with open questions - never invent answers.
HARD REFUSALS: No invented metrics/testimonials/awards/logos/competitor claims. Label output as draft, never "approved." Do not expand job 4 into a full site or nurture. No SEO strategy, media buying, or legal review under this pack.
If outside jobs 1-5, return to job choice or the marketing map for other spokes.
Skill - turn winners into a reusable skill
A skill encodes inputs - job - draft - QA so you stop re-pasting. Honest stub (not a zip): the skill stub outline below (marketing-prompt-pack / brand-context-prompt-runner). When you are ready to install (not dump) skills, use What Claude marketing skills should I install first?. Packaging catalog: Claude marketing skills (planned). Until then, use the agent block; fold stable prompts into the stub outline below. Positioning (job 1) bridges a future positioning skill - not shipped.
Skill stub outline
Honest outline only (not a downloadable zip). Working names: marketing-prompt-pack / brand-context-prompt-runner.
Purpose: load brand context, pick one of five jobs, collect required inputs, draft, return a filled QA checklist, refuse invented proof.
Required context: brand guidelines, 2-3 example assets, optional offer/ICP notes. If missing, ask. Do not invent.
Jobs: (1) positioning one-pager (2) campaign brief (3) brand-voice rewrite (4) one landing/email section (5) competitor teardown from pasted copy only.
Until packaged, paste the agent block into a Project. See Claude Skills for packaging patterns later.
Stack at a glance
| Surface | Use for this pack |
|---|---|
| Chat | Spike; weak without brand paste |
| Projects | Default - brand docs + agent block |
| Skills | After a prompt is stable (stub - later install) |
| Claude Code | Optional: same agent block in a repo workflow |
Confirm product surfaces on claude.com, Anthropic docs, and What are projects? (Checked 2026-09-11) - this page is field specs, not a product clone.
Pitfalls
- Empty context - generic copy.
- Count vanity - don't race 50 weak prompts vs five specs with QA.
- Hallucinated proof - metrics/logos not in inputs must not appear.
- Skipping QA or wrong spoke (SEO/social/email depth) - marketing map.
Quality bar: inputs filled · no invented claims · one job per run · human sign-off before publish.
FAQ
How many Claude prompts for marketing do I need?
Fewer than listicles claim. Five specs cover most drafting days. Add channel extras only when a job repeats weekly.
Are free prompt PDFs enough?
Only if they force inputs and QA. Most are megadumps. Prefer auditable specs plus an agent/skill path.
Claude vs ChatGPT prompts for marketing?
Structure transfers; what changes is brand-context loading and default surface. Deep tool choice: see Claude vs ChatGPT for general comparison; a marketing-specific spoke comes later. This page stays Claude-workflow focused.
Should I convert prompts into Claude skills?
Yes, once a prompt stops changing every run. Start from the skill stub outline below; then see Claude Skills for packaging patterns.
Can prompts replace a brief / positioning doc?
No. Prompts draft those docs from your notes. Without inputs you get plausible fiction. Keep a human-owned source of truth.
Where to go next
Claude for marketing for the cluster map. Claude prompts for work for general office prompts. Claude Skills and Create a Claude Code skill when a winner is stable enough to package. How to use Claude if you are still setting up. Also see How to use Claude for marketing for the Projects-first loop. SEO, email, and social spokes come next.