10 Claude Prompts for Digital Marketers (Copy, Paste, Keep)

---

Introduction

Managing email campaigns, social content, paid ads, and web copy—often all at once, often without much support—leaves little room for trial and error. These 10 Claude AI prompts for marketing are built for that situation: one person responsible for most of the output, needing something that works on the first or second try. This is a practical reference, not a conceptual overview.

A note on expectations: these prompts are yours to copy, paste, and keep. No subscription, no prompt-tool rental, no library that disappears behind a login. These are AI copywriting prompts you own outright.

According to Customer.io, 72% of marketers report AI saves them 20% or more of their time, with more than half shipping campaigns faster as a result. Both HubSpot and Customer.io have published their own Claude marketing prompt sets—these are standard operational tools at this point.

One expectation worth setting: these prompts produce strong first drafts, not finished copy. Edit before you publish. That applies to every tool professional writers use.

On why Claude specifically: how Claude compares to GPT and Gemini for writing matters here. Claude’s handling of tone and long-form structure makes it well-suited to marketing copy, which is why these prompts are built around it.

What follows is ten copy-paste prompts organized by channel, each with the reasoning behind its structure and a note on when to use it.

---

Before You Paste: Get the Most From These Prompts

Vague prompts produce vague copy.

As PromptOptimizer puts it: "The best marketing is specific. Specific audience, specific problem, specific offer. The more specific your inputs into Claude, the more specific the output." That’s not a technical principle—it’s the same logic behind any good creative brief.

Giving Claude a one-line prompt is like handing a freelancer a one-sentence brief with no background. You’ll get something generic back because that’s all they had to work with. Filling every placeholder with real, specific information is the single most important concept in prompt engineering for marketers, and it requires no technical knowledge to apply.

For brand voice consistency, Claude Projects is the practical fix. PromptOptimizer explains the setup: spend five minutes writing your business name, what you sell, who your customers are, your tone, and one or two examples of marketing copy that felt right. Save that in a Claude Project and every prompt you run in that conversation inherits the context automatically. Good prompts preserve your voice—they don’t replace it.

---

Quick Model Note: Sonnet 4 vs. Opus 4

Model	Best For	Speed
Claude Sonnet 4	Most marketing copywriting—emails, ads, social, headlines	Fast
Claude Opus 4	Deep strategic reasoning, long-form sales pages, complex analysis	Slower, goes deeper

According to Young Urban Project, Sonnet 4 is the right model for most marketing work. Reach for Opus 4 when you need more depth: a full sales page, a positioning document, or a nuanced brand strategy piece.

For a fuller breakdown of where each model performs best, a practical comparison of LLMs for writers covers the differences without the technical overhead.

---

The 10 Prompts

---

1. Email Subject Lines

You are an expert email copywriter. I need 10 subject lines for an email promoting [offer/campaign] to [target audience].

Write exactly 10 subject lines in these five categories (2 each):
- 2 curiosity-gap subject lines (hint at something without revealing it)
- 2 number-led subject lines (lead with a specific number or stat)
- 2 "you"-focused subject lines (speak directly to the reader's situation)
- 2 contrarian subject lines (challenge a common assumption)
- 2 ultra-short subject lines (under 5 words)

Context:
- Product/offer: [describe what you're promoting]
- Target audience: [describe who they are and what they care about]
- Tone: [describe in 3 adjectives, e.g., "direct, warm, professional"]
- One key pain point or desire: [what keeps them up at night / what they really want]

After writing all 10, flag your top 3 picks for A/B testing and briefly explain why (one sentence each).

Why it works: Structuring the prompt by category forces five genuinely distinct angles rather than variations on a single idea—which is what real A/B testing requires. The "flag your top 3" instruction adds a built-in prioritization step, saving you the evaluation time. More precise audience and tone inputs mean less editing on the output.

One marketer writing for Medium reported open rates jumping from 22% to 38% after structuring their email subject line prompts this way—requesting varied formats rather than variations on a single approach.

When to use: Before any email send where open rate matters. Especially useful when you need real A/B testing copy variations quickly and don’t have time to stare at a blank subject line field.

---

2. Email Sequences

You are an expert email copywriter specializing in lifecycle marketing. Write ONE email for a [sequence type: welcome / nurture / re-engagement / launch] sequence.

Sequence context:
- Sequence goal: [what you want subscribers to do or feel by the end of the sequence]
- Audience segment: [describe who these subscribers are and how they got on this list]
- Email number: [e.g., "Email 2 of 5"]
- CTA for this email: [what specific action should they take]
- Personalization variable: [first name / company name / purchase history / etc.]

Tone and voice:
- Brand tone: [3 adjectives]
- Example of copy that felt right: [paste 2–3 sentences from a past email that landed well]

Write the subject line, preview text, and full email body. Keep it focused on ONE idea and ONE call to action.

After drafting, note one thing you'd test or strengthen in a follow-up iteration.

Why it works: PromptOptimizer’s guidance is clear on this: narrow prompts with defined scope outperform attempts to generate an entire sequence in a single pass. Asking for one email at a time, then iterating, produces stronger individual emails. The "one idea, one CTA" constraint also reflects what high-performing emails actually do.

On personalization: Useme notes that personalized email campaigns generate 139% higher click rates than non-personalized ones—a broader email marketing finding, not a Claude-specific claim, but it’s why the personalization variable placeholder matters here.

When to use: New subscriber onboarding, re-engagement campaigns, product launch sequences. Work one email at a time rather than prompting for the full sequence at once.

---

3. Landing Page Headlines

You are a conversion copywriter. I need landing page headlines for [product/service].

Step 1 — Messaging angles:
Propose 7–10 high-impact messaging angles for this product. For each angle, write a one-sentence description. Consider angles like:
- What customers desperately want (the dream outcome)
- What they're trying to avoid (the fear)
- What makes them feel unsafe or uncertain (the doubt)
- What makes them feel confident (the reassurance)
- How they define success in this area
- What they'd say to a friend who asked "why did you buy this?"

Step 2 — Headlines:
For the top 3 angles, write 3 headline variants each (9 headlines total). Mix formats: direct statement, question, "how to," and benefit-led.

Product context:
- Product/service: [describe what it does]
- Target audience: [who they are, their main frustration]
- Desired action on this page: [sign up / buy / book a call / download]
- Tone: [3 adjectives]
- Existing headline (if any): [paste current headline for comparison]

Why it works: Headlines built from real customer language consistently outperform clever-but-vague phrasing. This prompt requires a messaging-first step before any headline gets written—which is how professional CRO work actually happens. The angle-generation step is where the real value sits.

One documented case from Medium showed a homepage headline revised from "Streamline Your Workflow" to "Stop Losing Hours to Broken Handoffs"—language pulled directly from competitor reviews—with conversions increasing 34% over 30 days. That’s the specificity principle in action.

When to use: New landing page builds, or any underperforming page that needs a messaging refresh. These are your go-to landing page copy templates when intuition alone isn’t moving the needle.

---

4. Sales Page Copy

You are a direct-response copywriter with deep expertise in persuasion frameworks. Write a section of a sales page for [offer] using the [AIDA / PAS] framework.

Framework choice:
- AIDA: Attention → Interest → Desire → Action
- PAS: Problem → Agitate → Solution

Offer context:
- Product/service name: [name]
- Primary audience: [describe them in detail—who they are, what they want, what they've already tried]
- Core benefit (transformation): [what life looks like after they buy]
- Top 3 objections: [what stops people from buying]
- Proof points available: [testimonials, case studies, stats, credentials—paste actual text]
- CTA: [what you want them to do and what happens when they do it]
- Tone: [3 adjectives]

Section to write: [choose one: above-the-fold section / problem section / solution section / testimonial intro / CTA block]

After drafting, list the 2 most important things to test or strengthen.

Why it works: Named frameworks produce persuasive structure because they’re built on documented principles of how buying decisions work. AIDA moves a reader from cold attention to action. PAS surfaces the pain, amplifies why it matters, then positions your offer as the logical next step. Either structure prevents the rambling, feature-heavy copy that undermines most sales pages.

Customer.io’s guidance specifically calls out context-loading: "Drop in the messy stuff—transcripts, past campaign performance, feedback from your sales team. Claude needs raw material and unique perspectives to generate real, relevant answers for you." The proof points placeholder is where that principle applies. Paste real testimonials, real stats, real language from customer conversations—that’s what separates useful output from generic benefits lists.

When to use: Product launches, evergreen sales pages, offer pages that need serious persuasive work. This is also where Opus 4 earns its place.

---

5. Social Media Captions (LinkedIn / Instagram / TikTok)

You are a social media copywriter who understands platform-native communication. Take the core message below and adapt it into three platform-specific captions.

Core message: [describe the single idea, insight, or offer you want to communicate]

For each platform, write one caption that fits the platform's native format and audience expectations:

LINKEDIN:
- Professional tone, thought-leadership framing
- Lead with a hook (bold first line that earns the "see more" click)
- 150–300 words
- CTA style: [invite conversation / drive to link / encourage connection]

INSTAGRAM:
- Conversational, visual-first framing (assume there's an image/reel)
- Hook in the first line
- 80–150 words + 5–8 relevant hashtags
- CTA style: [save this / link in bio / comment below]

TIKTOK:
- High-energy, hook-first, pattern-interrupt opening
- Short and punchy (50–80 words max for caption)
- Speak to the viewer directly
- CTA style: [follow for more / comment your answer / save this]

Brand tone: [3 adjectives]
Product/topic context: [what you sell or what this post is about]

Why it works: The same message doesn’t transfer cleanly across three platforms. LinkedIn’s professional register, Instagram’s visual-first tone, and TikTok’s pattern-interrupt rhythm each require structural adaptation—not just a copy-paste with different hashtags. This prompt handles that translation systematically.

The AI Corner’s research on workflow scalability captures the practical benefit: a single workflow that adjusts voice register and format for each platform means consistent core messaging without rewriting from scratch every time.

When to use: Any time you have one strong idea or piece of content to repurpose across channels, or when you’re running coordinated campaigns that need consistent messaging without identical copy.

---

6. Ad Copy (Google / Meta)

You are a performance copywriter specializing in paid advertising. Generate ad copy variations using the Angle Matrix method.

Ad platform: [Google Search Ads / Meta Feed Ads / Meta Stories / Display]

Core offer: [what you're advertising]
Target audience: [describe demographics, psychographics, and primary motivation]
Landing page URL context: [briefly describe what happens after they click]
Character limits: [Google headlines: 30 characters / Meta headline: under 8 words / Body: 90–125 characters]

Generate 5 variations, each built on a different angle:
1. Pain angle: Lead with the problem they're experiencing right now
2. Outcome angle: Lead with the transformation or result they want
3. Identity angle: Lead with who they are or who they want to become
4. Proof angle: Lead with a specific result, stat, or social proof signal
5. Contrarian angle: Challenge a common belief in their category

For each variation, write:
- Headline (within character limit)
- Body copy (within character limit)
- One-sentence note on what psychological trigger this variation is testing

Brand tone: [3 adjectives]
Top objection to pre-empt: [the #1 reason people don't buy]

Why it works: The Angle Matrix method—documented in God of Prompt’s ad copy frameworks—generates variations that are structurally different, not cosmetically different. Testing pain-led copy against identity-led copy tells you something real about your audience. Testing five rewrites of the same benefit tells you almost nothing.

Note the character limits built into the prompt. As God of Prompt specifies, Google Responsive Search Ads limit headlines to 30 characters, while Meta headlines perform best under eight words. Without those constraints in the prompt, Claude will write copy your ad platform immediately truncates.

When to use: Any paid campaign launch. Especially useful for conversion-focused copywriting at scale—when you need five genuine test variations, not five versions of the same angle.

---

7. Product Descriptions

You are an e-commerce copywriter who specializes in driving urgency and clarity. Write a product description for [product name].

Product context:
- Product name: [name]
- Category: [e.g., skincare, software, apparel, equipment]
- Primary audience: [who buys this and why]
- Core benefit (what problem does it solve): [be specific]
- Key features: [list 3–5 features with brief descriptions]
- Proof or credibility: [reviews, ratings, certifications, bestseller status, social proof signals]
- Price point and positioning: [premium / mid-range / value]

Write the description in this structure:
1. Opening hook (1–2 sentences): Lead with the transformation or outcome, not the product
2. Benefit body (3–5 sentences): Translate each key feature into a customer benefit
3. Urgency or scarcity signal (1 sentence): Real, honest urgency—avoid fake countdown language
4. CTA (1 sentence): Clear, action-oriented, benefit-forward

Tone: [3 adjectives]
Length: [short: under 100 words / standard: 100–200 words / long-form: 200–400 words]

Why it works: Most product descriptions lead with features because features are easy to list. This prompt forces the feature-to-benefit translation by structure. The opening hook constraint alone improves most e-commerce copy immediately—buyers don’t care what your product is until they understand what it does for them.

The 4U framework—cited in performance marketing resources—underpins this structure: Urgent, Unique, Useful, Ultra-specific. The urgency signal and benefit-forward CTA address two of those four directly. Specific inputs handle the other two.

When to use: New product launches, refreshing underperforming product pages, or building out a catalog at scale.

---

8. Blog Introduction Hooks

You are a content strategist and editor who specializes in first-impression copy. Write 5 alternative opening paragraphs (hooks) for a blog post on [topic].

Post context:
- Blog post topic: [title or working title]
- Target reader: [describe who they are and what brought them to this post]
- Primary keyword: [what are they searching for]
- Core promise of the post: [what will they know or be able to do after reading]
- Tone: [3 adjectives]
- Brand/publication: [name and brief description]

Write 5 distinct hook styles:
1. Rhetorical question hook: Open with the reader's unspoken frustration
2. Stat or data hook: Lead with a surprising or counterintuitive number
3. Scenario hook: Drop the reader into a specific, relatable moment
4. Contrarian hook: Challenge a widely held assumption in the first sentence
5. Story hook: Open with a brief, specific anecdote (2–3 sentences)

After each hook, write one sentence explaining what psychological mechanism it uses to hold attention.

Why it works: If the first paragraph doesn’t hold a reader, the rest of the post doesn’t matter. Five structurally different hooks give you real options—not five rewordings of the same opening.

Each hook style maps to a documented attention mechanism: curiosity (rhetorical question), surprise (data), recognition (scenario), challenge (contrarian), and narrative pull (story). Which one works best depends on your audience’s mindset when they arrive.

When to use: Any post where you need to reduce bounce. Particularly useful when a post has strong SEO potential but weak retention—the introduction is usually the problem, and this gives you five approaches to test.

---

9. CTA Optimization

You are a conversion copywriter who specializes in calls to action. I need CTA variations for [context: button / email / landing page / ad].

Current CTA (if you have one): [paste it here]
Offer: [what happens when someone clicks—what are they getting]
Audience: [who they are and what they want]
Page/placement context: [where does this CTA live in the customer journey]
Tone: [3 adjectives]

Generate 12 CTA variations organized by command type:

COMMAND LANGUAGE (4 variations):
- Start / Begin / Launch / Activate — action-forward commands
- Get / Grab / Claim / Take — possession-forward commands

VALUE FRAMING (4 variations):
- Benefit-led CTAs (lead with what they get)
- Risk-reversal CTAs (lead with what they don't risk)

SPECIFICITY LEVEL (4 variations):
- Ultra-specific (include a number, timeframe, or outcome)
- Curiosity-gap (hint at what happens next without fully revealing it)

After writing all 12, flag your top 3 with a one-sentence rationale for each.

Why it works: CTAs often fail not because the offer is weak but because the language doesn’t match the reader’s mental state at the moment they encounter it. Testing command language against value framing reveals whether your audience responds better to being told what to do or being told what they get. Testing specificity tells you whether they need more confidence or more curiosity to click.

The Angle Matrix logic applies here too: variations built on different mechanisms reveal what actually drives your audience. Twelve variations organized by type gives you a structured testing roadmap rather than a random collection of button text.

When to use: Any time a page is underperforming and the offer itself is solid. CTA language is consistently one of the highest-return conversion fixes, and this prompt takes minutes to run.

---

10. Prompt Refinement: Making Output Sound Like You

You are a brand voice editor. I'm going to share a piece of AI-generated copy that needs to be refined to match my brand voice. Your job is to rewrite it so it sounds unmistakably like us.

Brand voice context:
- Brand name: [name]
- Brand personality in 3 adjectives: [e.g., direct, warm, unconventional]
- Things we'd NEVER say: [list 3–5 phrases or tones that feel wrong for your brand]
- Things we ALWAYS do: [list 3–5 stylistic patterns—e.g., short sentences, rhetorical questions, specific numbers]
- Example of copy that felt completely right: [paste 2–3 sentences from your best past content]
- Example of copy that felt off-brand: [paste 2–3 sentences you'd never publish]

Copy to refine: [paste the AI-generated draft here]

Rewrite instructions:
1. Maintain all factual information and key messages exactly
2. Apply brand voice throughout—sentence rhythm, word choice, punctuation style
3. Flag any phrases that still feel generic after rewriting (mark with [REVIEW])
4. After the rewrite, explain in 2–3 sentences what you changed and why

Run this refinement pass, then suggest one follow-up iteration to push it further.

Why it works: This is the prompt that moves every other prompt in this list from "good draft" to "sounds like us." Useme makes the problem plain: "AI-generated content can sound polished—but that doesn’t guarantee it reflects your brand. Without clear guidelines, output may drift from your tone, values, or positioning." The "never say / always do" structure trains Claude on your brand’s negative space—what you’d never publish—which is often more instructive than describing what you would.

PromptOptimizer reinforces this: two paragraphs of your best-performing past copy outperforms 500 words of abstract brand guidelines. The example copy placeholders in this prompt operationalize exactly that.

When to use: After using any other prompt in this list when the output is 80% right but not quite there. Also run it as a final pass before anything goes live—subject lines, landing page copy, ad variations, all of it.

---

Putting It All Together: The Prompt Chaining Advantage

These ten prompts are designed to feed each other, not operate in isolation.

Young Urban Project’s guidance on prompt chaining explains the mechanic: "Stack prompts. The campaign brief from Prompt 1 can feed the ad copy in Prompt 10. The audience research from Prompt 5 can be pasted into the landing page prompt. These prompts aren’t isolated; chain them and the quality compounds."

A practical chaining workflow:

1. Start with messaging angles — Use Prompt 3 (Landing Page Headlines) for the angle-generation step. Those angles reveal how your audience actually thinks about their problem.
2. Feed those angles into your ad copy — Paste the top 3 angles from Prompt 3 into Prompt 6 (Ad Copy). Your Angle Matrix variations now reflect real audience psychology.
3. Use the winning ad angle for email subject lines — The angle that performs best in ads often transfers to email. Feed it into Prompt 1.
4. Run everything through Prompt 10 — Before anything goes live, run the refinement pass. Consistency across channels is a brand trust signal.

That’s one connected campaign strategy built from the same source material, not four separate prompting sessions.

Prompt	Best Chained With	Why
Prompt 3 (Landing Page Headlines)	Prompt 6 (Ad Copy)	Messaging angles inform ad creative
Prompt 6 (Ad Copy)	Prompt 1 (Email Subject Lines)	Winning angles transfer across channels
Prompt 4 (Sales Page)	Prompt 9 (CTA Optimization)	Sales page context improves CTA specificity
Any prompt	Prompt 10 (Refinement)	Final brand voice pass before publishing

For a deeper look at building connected workflows with custom instructions across platforms, optimizing Claude and other AI tools with custom instructions provides the practical templates.

---

When Claude Isn’t the Right Tool

Claude handles long-form, tone-sensitive copy well, but it’s not the only option worth knowing. A comparison of AI writing assistants breaks down where Claude’s strengths hold against competing platforms. The short version:

• Claude excels at nuanced tone, long-form persuasion, and maintaining voice across extended copy
• ChatGPT handles rapid iteration and structured output formats well
• Gemini integrates tightly with Google Workspace, which matters for some teams

The useful question isn’t which AI is best overall—it’s which one is best for a specific task. For a full analysis of how these models perform across specific marketing applications, Claude vs. GPT vs. Gemini for writing tasks is worth bookmarking.

---

FAQ

How do I make Claude sound less generic when I use these prompts?

Two things matter most. First, fill every placeholder with real, specific information—don’t leave [target audience] as a vague category. Second, use Prompt 10 as a final pass. The "never say / always do" structure and example copy inputs are what convert generic output into brand-specific copy. PromptOptimizer documents this clearly: two paragraphs of your best past copy outperforms any amount of abstract brand guideline language.

Do I need Claude Pro or paid access to use these prompts?

The prompts work with any Claude access level. Claude Sonnet 4—the recommended model for most marketing copywriting—is available on the free tier with usage limits. Claude Projects, which saves your brand context and eliminates re-explaining it each session, requires a paid plan. For high-volume marketing use, Projects alone makes the upgrade worth considering.

How do these Claude prompts compare to using ChatGPT or other AI writing tools?

Claude’s specific strength for marketing copy is its handling of tone and nuance in longer-form content. A practical breakdown of how Claude compares to other models for content creation covers the strategic and technical considerations in detail. For conversion-focused copywriting with a strong brand voice requirement, Claude consistently produces drafts that require less editing.

Can I use these prompts for client work across multiple brands?

Yes—that’s what Claude Projects is designed for. Create a separate Project for each client, load their brand context (tone, audience, example copy, things they’d never say), and every prompt you run in that Project inherits that context. Customer.io’s guidance confirms this is how marketing teams are running multi-client workflows today.

How many iterations should I expect before copy is publish-ready?

Realistically, one to three. A well-filled prompt with specific inputs and real example copy usually gets you to 80–90% in the first pass. Use Prompt 10 for brand voice refinement, then a human edit for judgment calls. PromptOptimizer sets a useful baseline: "Claude gives you a strong draft, not finished copy. Always edit before publishing anything."

What’s the best way to save and organize these prompts for ongoing use?

Claude Projects is the most integrated option—store your brand context and run all your marketing copy work from there. For a broader prompt library system, guidance on building reusable custom instruction systems covers the workflow architecture in practical detail.

---

Conclusion

These ten prompts cover the copy channels that consume most marketing time: subject lines, sequences, landing pages, sales copy, social captions, ads, product descriptions, blog hooks, CTAs, and the refinement pass that makes all of them sound like your brand rather than a template.

The same principle runs through all ten: specificity drives quality. Fill the placeholders with real information. Load Claude with your actual brand context. Run the refinement pass before anything goes live. That discipline is what separates an 80% draft from something you’re comfortable publishing.

These prompts are yours—no subscription, no platform lock. Copy them into a document, a Notion page, or a Claude Project, and they’re available as long as you need them.

If you want a content creation workflow that applies this same logic to full-length articles—with built-in research, SEO structure, and brand voice from the start—Libril’s "Buy Once, Create Forever" model is worth a look. One-time purchase, no recurring fees.