All prompts are curated and reviewed by John Allick. Content Policy →
The blank-page problem doesn't disappear when you have an AI model — it moves earlier in the process, from "what do I write?" to "what do I tell the AI to write?" This guide solves that second problem. Every prompt here starts with a specific editorial goal: a blog post targeting a competitive keyword, a newsletter built to reduce churn, a landing page designed to convert. They're built for writers who know what good output looks like and need AI to get them there faster — not for people who want the AI to make every creative decision. Use them as starting points. Adjust the bracketed variables to your topic, audience, and brand voice. The CRISP structure does the rest.
What Are AI Writing Prompts?
AI writing prompts are structured instructions that writers, marketers, and editors give to AI models — such as ChatGPT, Claude, or Gemini — to generate blog posts, SEO articles, landing pages, newsletters, product descriptions, and other written content. Unlike a casual question, a well-engineered writing prompt defines the role of the writer, the target audience, the format, the tone, the word count, and any SEO or brand requirements — all in a single instruction.
Editorial teams noticed something counterintuitive in 2024: switching AI models rarely improved output quality as much as rewriting the prompt that drove them. The model accounts for roughly 20% of the final result. The remaining 80% is the quality of the brief — the persona, the structural schema, the SEO requirements, and the purpose that tells the AI exactly what "a great article" means for this publication, this reader, and this search intent.
A well-crafted AI writing prompt can produce a 2,500-word SEO article, a complete landing page, a 30-day editorial calendar, or a polished newsletter issue in a single generation. The difference between a generic prompt ("write a blog post about productivity") and an engineered one (the templates on this page) is the difference between AI slop and a draft that needs only a light editorial pass.
The CRISP Framework for Writing Prompts
The topic, the publication, the reader's existing knowledge
An expert persona ('senior editor at a top publication')
Structure, format, word count, and SEO requirements
Brand voice, tone, reading level, banned words
The specific outcome — rank, convert, inform, persuade
Why Prompt Engineering Matters in Content Writing
A 3,000-word blog post that ranks has nothing in common structurally with a 3,000-word blog post that does not — except the word count. The difference is specificity: H2 schema, entity coverage, reader persona, search intent, and competitive benchmarking baked into the prompt before the AI writes a single word. Content teams encoding that rigor into their prompts publish high-ranking articles at the volume of a media operation with a fraction of the headcount.
Prompt engineering is now a core editorial competency. Content teams at publishers, SaaS companies, and agencies have built entire production workflows around structured AI prompts — using them to scale blog production, draft landing pages, brief writers, and automate the first draft of every deliverable. The writers who master this skill in 2026 are not being replaced by AI. They are quietly outproducing everyone who is not.
Best ChatGPT GPT-5.5 Prompts for Writers
ChatGPT GPT-5.5 is the strongest model for structured, format-constrained writing. It excels at SEO articles that follow strict H2 patterns, listicles with consistent item structure, FAQ blocks formatted for People Also Ask, and editorial briefs that need to follow a defined template. GPT-5.5's particular strength is following complex multi-step instructions precisely — which is why it is the top choice for SEO content teams who need to ship 20+ articles per month.
Best for: SEO articles, listicles, how-to guides, FAQ sections, editorial outlines, blog idea generation, content briefs
Claude Opus 4.8 Writing Workflows
Claude Opus 4.8 is the editorial favorite for long-form writing that requires nuance, careful argument structure, and voice consistency. Claude's reasoning depth and large context window make it uniquely suited for white papers, brand voice rewrites, op-eds, narrative case studies, and any writing where generic output would be obvious. Pair Claude with a strong style reference and it produces drafts that need only a light human edit before publishing.
Best for: Long-form articles, brand voice rewrites, landing pages, narrative case studies, technical documentation, op-eds, voice-sensitive copy
Gemini 3.5 Flash Content Writing Examples
Gemini 3.5 Flash stands apart because of its real-time web access and deep research synthesis. For writers, this makes it the first choice for research-driven content, competitive landscape pieces, trend reports, and articles that need to reference current statistics or recent industry events. Where ChatGPT and Claude work best with context you provide, Gemini can go find that context itself — making it the model of choice for content that needs to be both fresh and authoritative.
Best for: Topic cluster design, competitive content audits, research-driven articles, trend reports, fact-checked content, comparison posts
| Writing Task | ChatGPT GPT-5.5 | Claude Opus 4.8 | Gemini 3.5 Flash |
|---|---|---|---|
| SEO articles & listicles | ⭐⭐⭐ | ⭐⭐ | ⭐⭐ |
| Long-form essays & op-eds | ⭐⭐ | ⭐⭐⭐ | ⭐ |
| Brand voice rewrites | ⭐ | ⭐⭐⭐ | ⭐ |
| Research-backed content | ⭐ | ⭐⭐ | ⭐⭐⭐ |
| Landing page copy | ⭐⭐ | ⭐⭐⭐ | ⭐ |
| Social & short-form | ⭐⭐⭐ | ⭐⭐ | ⭐⭐ |
| Technical documentation | ⭐⭐ | ⭐⭐⭐ | ⭐⭐ |
| Topic cluster design | ⭐ | ⭐⭐ | ⭐⭐⭐ |
AI Prompts for Blog Writing & Long-Form Content
Blog writing is the highest-leverage application of AI prompts for most content teams. With the right prompt, AI can produce a fully structured outline, a polished 1,500-word draft, or a complete 2,500-word pillar article — including H2 structure, internal link suggestions, meta description, and FAQ block — in a single generation. The human editor's job shifts from writing to refining: adding original data, brand perspective, and the on-the-ground expertise that AI cannot replicate.
The most effective blog teams in 2026 use AI not only for individual drafts but for orchestrating their entire content calendar. A structured outline prompt turns a 30-minute brainstorm into a publishable outline; a long-form draft prompt cuts first-draft time from a day to under an hour. Pair these with rigorous human editing and the result is more content, not lower-quality content.
AI Prompts for SEO Content Writing
SEO writing is one of the most powerful applications of AI prompt engineering. A structured SEO prompt encodes everything your draft needs to compete in 2026: primary keyword placement, secondary keyword distribution, H2 hierarchy, internal link anchors, FAQ schema, Flesch-Kincaid reading level, and the People Also Ask format Google rewards.
The most advanced SEO teams use AI not just for individual articles but for building entire topic clusters — defining pillar pages, supporting cluster pieces, internal linking architecture, and publishing roadmaps. A well-structured topic cluster prompt can compress weeks of content planning into a single AI session. See the Gemini 3.5 Flash topic cluster prompt above for an example.
Key Inputs for an SEO Prompt
- Primary keyword + 3-5 secondary keywords with search intent for each
- Target word count tied to the top 3 ranking pages
- H1, H2, and H3 hierarchy rules with one keyword variant per H2
- Internal link anchor suggestions to 3-5 existing pages
- FAQ block formatted for People Also Ask snippets
- Meta title (under 60 chars) and meta description (under 155 chars)
- Reading level target (Grade 8-10 Flesch-Kincaid)
AI Prompts for Copywriting
Copywriting is where the difference between a generic prompt and an engineered one is most visible. Generic AI copy reads like marketing spam. Engineered AI copy reads like the work of a senior copywriter — because the prompt gave the model the same brief a senior copywriter would receive: a target customer, a single benefit, a specific psychological angle, a banned-word list, and a hard word count. Use these prompts for ad copy, sales pages, email sequences, and any place where word choice determines conversion.
Related: AI Marketing Prompts → · AI Email Writing Prompts →
AI Prompts for Landing Pages
Landing pages are where every word is measured in dollars. A well-structured landing page prompt asks the AI for three variations of the hero block in different psychological angles — pain-led, benefit-led, and status-led — so a marketer can A/B test the angles, not just the words. The same prompt pattern works for SaaS free trials, ecommerce product pages, consulting offers, and lead-magnet downloads. Used consistently, these prompts shorten the time from a positioning brief to a test-ready landing page from days to under an hour.
Pair an AI-drafted landing page with one round of editing by a senior conversion copywriter and the result is usually within 5 percent of what a full agency engagement would have produced. The Claude landing page prompt earlier on this page is built for exactly this workflow.
AI Prompts for Email Marketing & Newsletters
Email and newsletters still deliver the highest ROI of any digital channel — and AI prompts are the fastest way to capitalize on it. Whether you are publishing a weekly newsletter, launching a product, or running a re-engagement campaign, the structured prompts below give your AI the context to produce emails that are actually opened and replied to. The pattern is the same: specify recipient role, sender context, email goal, word limit, tone, and CTA type.
Also see: AI Prompts for Email Writing →
AI Prompts for Product Descriptions
Ecommerce product descriptions are the highest-ROI single application of AI writing prompts. A structured product description prompt produces three versions of the same product in different formats — full product page copy, meta description, and ad caption — in a single generation. Multiply that by hundreds of SKUs and a small ecommerce team can refresh an entire catalog in a week instead of a quarter.
AI Prompts for YouTube & Video Scripts
Video script writing benefits enormously from structured prompts. A good YouTube script prompt produces not only the script but also thumbnail title options, an SEO-optimized description with timestamps, B-roll suggestions, and on-screen text cues — all in a single generation. The result is a creator who spends more time recording and editing and less time staring at a blinking cursor.
AI Prompts for Brand Voice & Tone Consistency
Brand voice is the single biggest reason teams are nervous about AI writing — and the easiest concern to address with the right prompt. By including a reference sample of on-brand copy, a banned-words list, and explicit voice attributes, you can teach an AI model your brand voice in a single prompt. Claude Opus 4.8 is the strongest model for this kind of voice-matched rewrite, but every modern model can handle it when the prompt is structured properly.
AI Prompts for Content Repurposing
The highest-leverage AI workflow for content teams is repurposing — turning one pillar asset into 10+ derivative outputs across channels. A structured repurposing prompt takes a long-form article, podcast transcript, or webinar recording and outputs a LinkedIn post, a Twitter/X thread, an Instagram carousel, a newsletter teaser, a short-form video script, a Quora answer, and a set of pull-quotes in one generation. The result is one piece of original thinking distributed across every surface your audience uses.
AI Prompts for Content Strategy
Beyond individual drafts, the most senior application of AI prompts in content writing is strategy work: building editorial calendars, mapping topic clusters, planning launch sequences, and designing content distribution playbooks. AI is uniquely good at this because strategy work is structurally repetitive — the same questions ("what should we publish next?", "how do we tie these pieces together?") repeat across teams, brands, and niches. A well-engineered strategy prompt automates the structural part so the editor can focus on judgment.
Best Prompt Engineering Techniques for Writers
These five techniques consistently separate high-performing writing prompts from average ones:
Role Priming with Credentials
Start every prompt with a specific expert role that includes credibility signals. 'Act as a writer' produces generic prose. 'Act as a senior editor at a top publication who has shipped 500+ long-form features' produces editorial-grade output. The credentials prime the model to write at that level of craft.
✓ Good
"Act as a senior SEO content strategist with 10+ years of experience publishing for top-ranking sites."
✗ Weak
"Act as a writer."
Audience Precision Over Demographics
Vague audience descriptions produce vague writing. Specify role, expertise level, what they already know, and what they need to feel after reading. The more your prompt reads like a real reader brief, the more targeted the prose.
✓ Good
"Audience: mid-career SEO specialists who already understand keyword research and are looking for advanced topical authority strategies."
✗ Weak
"Audience: marketers."
Hard Output Constraints
AI models produce stronger writing when they have constraints. Word limits, reading-level targets, banned words, sentence-length caps, and format rules all improve quality because they force the model to be more deliberate with every word.
✓ Good
"Under 1,500 words. Grade 8-10 reading level. No buzzwords: 'innovative', 'cutting-edge', 'game-changing'. No sentence over 20 words."
✗ Weak
"Keep it short and engaging."
Voice References and Style Samples
Teaching an AI your brand voice by description is harder than teaching it by example. Always paste a 100-200 word sample of your best-performing on-brand copy into the prompt. The model will pattern-match more reliably to a sample than to adjectives.
✓ Good
"Match the voice in this sample: '[200 words of on-brand copy]'. Maintain the same rhythm, vocabulary level, and use of contractions."
✗ Weak
"Use a friendly but professional voice."
Chain-of-Thought for Editorial Work
For complex editorial tasks (long-form features, white papers, topical authority planning), use chain-of-thought prompting: instruct the AI to reason step-by-step before producing the final draft. This dramatically improves argument structure and reduces the need for heavy re-editing.
✓ Good
"First, outline the 5 H2 sections. Then draft each section. Finally, write the introduction last so it accurately previews the body."
✗ Weak
"Write a long-form article on [topic]."
Common AI Writing Prompt Mistakes
Even experienced writers make these prompt engineering mistakes — and they are the primary reasons AI produces disappointing first drafts:
No defined audience
Writing 'create a blog post' with no reader definition produces prose that appeals to no one. Always specify role, expertise level, and the question they came to your article to answer.
No output constraints
Without word limits, reading-level targets, or sentence-length caps, AI defaults to verbose, generic prose. Hard constraints sharpen every output.
Single-shot iteration
Generating one draft and iterating is much slower than asking for three angles upfront. Request multiple openings or variations in the first prompt.
No banned-words list
Without explicit exclusions, AI defaults to corporate clichés ('leverage', 'synergy', 'innovative') that kill credibility on the page. Always add a 'never use' list.
Missing brand voice sample
AI cannot pattern-match a voice you describe in adjectives. Paste 100-200 words of on-brand copy directly into the prompt as a reference.
Weak role assignment
'Act as a copywriter' is too vague. 'Act as a senior conversion copywriter who has shipped 8-figure landing pages' primes the model for expert-level craft.
No structural blueprint
Asking the AI to 'write a long-form article' without specifying H2 hierarchy, FAQ placement, or internal link slots produces output that needs heavy restructuring.
Ignoring SEO inputs
Skipping primary keyword, secondary keywords, and meta requirements means you will rewrite half the draft anyway. Bake SEO into the prompt.
The Future of AI in Content Creation
AI content writing is moving from individual prompt use to orchestrated workflows — where multiple AI models work in sequence, each handling the task it does best. A typical 2026 editorial workflow looks like this: Gemini 3.5 Flash conducts real-time research and competitive analysis, Claude Opus 4.8 drafts the long-form article in brand voice, ChatGPT GPT-5.5 generates the SEO meta block and FAQ, and a human editor adds original data, examples, and final polish before publishing.
The next frontier is agentic editorial systems — where AI agents autonomously plan topic clusters, draft pillar pages, refresh aging articles, and propose internal linking changes based on real ranking data. Models like GPT-5.5, Claude Opus 4.8, and Gemini 3.5 Flash now have the reasoning capability to close this loop in production environments.
For writers, the highest-leverage skill in this environment is not prompt writing itself — it is prompt system design: building the libraries of reusable, parameterized prompt templates that power your editorial workflows at scale. This page is the foundation of that library for content teams.
Generate a Custom AI Writing Prompt
Describe your topic, audience, and goal — get a production-ready writing prompt optimized for your AI model in seconds.
Frequently Asked Questions
The most common questions about AI writing prompts — optimized for Google's People Also Ask, Featured Snippets, and AI Overview features.
Final Thoughts: AI Writing Prompts in 2026
Prompt engineering is no longer optional for serious content teams. The gap between writers who use structured prompts and those who treat AI like a chatbot grows wider with every model upgrade. GPT-5.5, Claude Opus 4.8, and Gemini 3.5 Flash are extraordinarily capable — but only when given the structured context they need to perform at their best.
The prompt templates on this page are a starting point. The most effective editorial teams in 2026 build internal prompt libraries — catalogued by format, channel, audience, and AI model — and treat them as a core content asset. Every prompt that produces a ranked article, a converting landing page, or a high-engagement social post gets saved, refined, and reused.
Start by copying the prompts that match your most immediate need. Customize the placeholders. Test the output. Edit ruthlessly. Then build your own library from there. PromptPrepare's free generator can create custom writing prompts for any scenario you describe — use it whenever the templates on this page do not quite fit your specific context.
AI Prompts for Social Media Content
Social content is where AI-driven workflows deliver the fastest ROI. With structured prompts a creator can produce a complete week of content — LinkedIn posts, Twitter/X threads, Instagram captions, TikTok scripts — in a single two-hour session. The key is encoding your brand voice, audience profile, and platform-specific format requirements into the prompt so the AI returns platform-ready content rather than rough drafts.
Platform coverage: LinkedIn · Twitter/X · Instagram · TikTok · YouTube · Facebook · Threads · Pinterest
LinkedIn Thought Leadership Post (Under 300 Words)
Twitter/X Thread Generator (8-12 Tweets)