Copy-paste ChatGPT prompts for blog posts, headlines, meta descriptions, product descriptions, email copy, social captions, case studies, and content repurposing. Each prompt includes the exact text, what it produces, and tips for better results. Tested on real content campaigns at ScaleGrowth.Digital.
Last updated: March 2026 · Reading time: 18 min
A ChatGPT content writing prompt is a structured instruction that tells the model your role, audience, format, and constraints so it generates a specific type of written content rather than generic filler.The pattern we’ve seen across 45+ campaigns: writers who use ChatGPT as an editor and brainstorming partner produce content 3x faster than those who ask it to “write a blog post about X.” According to Coursera’s 2026 prompt engineering guide, specifying role, context, format, and constraints in every prompt lifts output quality by 40-60% compared to open-ended requests.
“We stopped asking ChatGPT to write content for us and started asking it to challenge our content. The quality gap is enormous. A prompt that says ‘act as a senior editor and tear this apart’ produces better work than ‘write me a blog post’ every single time.”
Hardik Shah, Founder of ScaleGrowth.Digital
Create a detailed blog post outline for the topic "[topic]" targeting the keyword "[keyword]". The audience is [describe audience]. Include: a working title (under 60 characters), a meta description (under 155 characters), 6-8 H2 sections with H3 subsections, the key point for each section (1 sentence), word count targets per section, and 3 internal linking opportunities. The tone should be [professional/conversational/technical]. Format as a structured outline I can hand to a writer.What it produces: A ready-to-assign content brief that covers structure, SEO elements, and writer guidance in one pass. Pro tip: Add “Include one data point or statistic placeholder per section” to force the outline to account for evidence-based writing.
I'm writing a skyscraper post to outrank this existing article: [paste URL or summary of competing article]. My target keyword is "[keyword]". Analyze the competitor's structure and create an outline that: covers every topic they cover, adds 3-4 sections they missed, includes more specific examples and data points, and uses question-based H2 headings for featured snippet potential. Target word count: [number] words.What it produces: A gap-filled outline that systematically addresses every angle the top-ranking content covers, plus areas they skipped. Pro tip: Paste the actual competitor text, not just the URL. ChatGPT can’t browse the web, so it needs the content directly.
Create an outline for a listicle blog post: "[Number] [Topic] [for/that] [Benefit]". For each list item, provide: a descriptive H2 heading, a 2-sentence summary of what to cover, one example or proof point to include, and an estimated word count (100-200 words per item). Include an introduction (150 words), a comparison table after item 5, and a conclusion with a clear CTA. Total target: [number] words.What it produces: A structured listicle outline where each item has substance rather than padding. Works for “best tools” posts, tips roundups, and example collections. Pro tip: Specify “Order items by usefulness to the reader, not alphabetically” to get a more engaging sequence.
Create a step-by-step how-to guide outline for "[task]". The reader is a [experience level] who wants to [specific outcome]. Include: a prerequisites section (tools, accounts, knowledge needed), numbered steps (5-8) with sub-steps where needed, estimated time for each step, common mistakes at each step and how to avoid them, and a "what to do next" section. Each step should be independently useful if someone lands on it from search.What it produces: A process-oriented outline with built-in troubleshooting. Each section answers the implicit question “what could go wrong here?” Pro tip: Add “Write each step heading as an action verb + outcome, e.g., ‘Set Up Your Tracking Pixels to Capture Every Conversion'” for clearer step names.
Create an outline for a comparison post: "[Product A] vs [Product B]". Include: a TL;DR comparison table (6-8 dimensions), individual overviews (150 words each), head-to-head sections for pricing, features, ease of use, integrations, customer support, and best for. End with "Choose [A] when..." and "Choose [B] when..." verdict sections. The reader is a [role] evaluating both for [use case]. Take a position on which is better for this audience.What it produces: A complete comparison framework with the comparison matrix upfront (search engines love tables) and detailed analysis below. Pro tip: Tell ChatGPT “Don’t hedge. Pick a winner for this specific audience and explain why.” Fence-sitting comparison posts don’t rank well or earn trust.
I'm building a content cluster around the pillar topic "[topic]". Plan 8-10 supporting articles that link back to the pillar. For each supporting article: suggest a title (under 60 characters), the target keyword, the search intent (informational, commercial, transactional), the content format (how-to, listicle, comparison, template), and word count target. Also suggest 2-3 internal linking paths between the supporting articles themselves. The audience is [describe].What it produces: A topic cluster map with internal linking architecture. We used this prompt to plan a 12-article cluster for a B2B SaaS client that generated 340% more organic traffic than their previous unstructured blog approach. Pro tip: Run this prompt, then validate each suggested keyword in Semrush or Ahrefs before committing to the cluster.
Write an intro paragraph (3-4 sentences, under 80 words) for a blog post about "[topic]" targeting "[keyword]". Use the BLUF method: sentence 1 answers the core question directly, sentence 2 provides one supporting data point or context, sentence 3 tells the reader exactly what they'll get from this article. No throat-clearing, no "In today's world..." openings. Include the keyword naturally in the first sentence.What it produces: A tight opener that puts the answer first. This matches how Google extracts featured snippets and how AI models identify authoritative answers. Pro tip: Add “The reader just typed ‘[keyword]’ into Google. They want an answer, not a lecture.” This framing consistently produces sharper intros.
Write a 4-sentence intro for a blog post about "[topic]". Start with a specific, concrete scenario that the reader will recognize from their own experience (not a hypothetical, not "imagine..."). Follow with the tension or problem that scenario creates. Then bridge to the solution this article provides. End with what the reader will walk away with. Under 100 words. Include "[keyword]" naturally.What it produces: An experiential opener that creates recognition. Works best for problem-solving content where the audience has felt the pain firsthand. Pro tip: Specify the reader’s job title. “The reader is a B2B marketing manager” produces very different scenarios than “The reader is a solo founder.”
Write a 3-sentence intro for a blog post about "[topic]". Open with a specific, sourced statistic that creates urgency or surprise. Follow with what that stat means for [audience]. Close with a one-sentence preview of the article's content. Under 70 words. Do not use "In today's" or "It's no secret that" or similar filler openings.What it produces: A data-driven opener that establishes credibility from sentence one. Verify the stat ChatGPT generates against real sources before publishing. Pro tip: Feed ChatGPT an actual stat you’ve verified. “Use this stat in sentence one: [stat + source]” works better than asking it to find one.
Write a 4-sentence intro for a blog post about "[topic]" that challenges a common assumption. Sentence 1: state the conventional wisdom most people believe about [topic]. Sentence 2: challenge it with evidence or a counterpoint. Sentence 3: explain what's actually true. Sentence 4: preview what this article covers. Include "[keyword]" in sentence 1 or 2. Under 90 words.What it produces: A pattern-interrupt opener that earns attention by challenging expectations. Best for topics where the common advice is outdated or oversimplified. Pro tip: This format works well for LinkedIn and social sharing. The “wait, what?” reaction drives clicks from social feeds.
Write a 4-sentence intro using the Problem-Agitate framework for a blog post about "[topic]". Sentence 1: name the specific problem the reader faces (be precise, not vague). Sentence 2: agitate by describing the consequences of not solving it. Sentence 3: hint at the approach this article takes. Sentence 4: tell the reader the specific outcome they'll achieve. Under 85 words. Include "[keyword]" once, naturally.What it produces: An empathy-driven opener that validates the reader’s frustration before offering direction. The agitation step is key: it separates this from a bland “this article will show you…” intro. Pro tip: Specify the real consequence. “Agitate with lost revenue” hits harder than “agitate with general frustration.”
Generate 10 title tag options for a blog post about "[topic]" targeting the keyword "[keyword]". Requirements: each title must be under 60 characters (show character count), include the keyword in the first 30 characters where possible, use a proven format (how-to, number, question, or benefit-driven), and avoid clickbait. Rank them by your estimate of click-through rate potential and explain your top pick.What it produces: 10 title variations with character counts. The ranking forces ChatGPT to evaluate rather than just generate, which improves the top options. Pro tip: Add “Also show the Google SERP preview (title + meta description) for your top 3 picks” to see how they’ll look in search results.
I have this working headline: "[headline]". Rewrite it 8 ways, each using a different emotional trigger: curiosity, urgency, fear of missing out, aspiration, authority, surprise, specificity, and social proof. Keep each under 65 characters. Mark which emotion each variant targets. Then pick the 2 strongest for a B2B audience and the 2 strongest for B2C.What it produces: Eight headline angles from one concept. The emotion labeling helps you choose deliberately rather than going with whatever “sounds good.” Pro tip: B2B headlines that include numbers outperform those without by 15-25% on average. Specify “Include a number in at least 4 variants.”
Generate 5 pairs of A/B test headlines for a [content type] about "[topic]". Each pair should test one variable: pair 1 tests number vs. no number, pair 2 tests question vs. statement, pair 3 tests benefit-led vs. curiosity-led, pair 4 tests short (under 40 chars) vs. long (55-65 chars), pair 5 tests "you" language vs. third-person. Show character counts for each. Explain what insight each test would produce.What it produces: Structured headline tests with clear hypotheses. Most content teams test headlines randomly. This prompt creates experiments with learning value. Pro tip: Run the winning headline through A/B testing on your actual audience. ChatGPT predicts well but your audience has the final vote.
I'm publishing a blog post with this headline: "[headline]". Write 5 email subject lines (under 45 characters each) and 5 social media headlines (under 100 characters each) that promote this post. The email subject lines should create curiosity without being clickbait. The social headlines should work for LinkedIn. Show character counts. Don't repeat the blog headline verbatim in any option.What it produces: Distribution-ready copy for two channels. Most content marketers use the blog headline everywhere, which wastes the opportunity to tailor messaging by platform. Pro tip: Add “The email open rate benchmark for our industry is 22% (Mailchimp, 2025). Write subject lines that would beat that.”
Score these 5 headlines on a 1-10 scale across 6 dimensions: keyword presence, emotional pull, specificity, length appropriateness (for SEO title tags, under 60 chars), click-through potential, and uniqueness vs. competing titles in the same space. Show a table with scores. Explain the biggest weakness of each headline and suggest a specific fix. Headlines: [paste 5 headlines]What it produces: A structured evaluation that catches weaknesses your eye skips after staring at headlines too long. The table format makes comparison easy. Pro tip: Include competing titles from the SERP: “These are the current top 5 titles ranking for ‘[keyword]’: [paste]. My headline needs to stand out from these.”
Write 5 meta description options for a page about "[topic]" targeting "[keyword]". Each must: be between 150 and 160 characters (show count), include the keyword naturally, contain a clear benefit statement, end with a call to action or value hook, and avoid starting with the brand name. Rank them by estimated CTR.What it produces: Five meta descriptions with character counts, ranked by likely performance. The character count constraint prevents truncation in SERPs. Pro tip: Google bolds the search query in meta descriptions. If your keyword is “content writing prompts,” that entire phrase will be bold, drawing the eye. Front-load it where possible.
Write 4 meta descriptions for a [product/service] page about "[offering]". Include: the primary benefit the customer gets, a differentiator from competitors, and a specific CTA. Each must be 150-160 characters. Two should lead with a benefit, two should lead with a pain point. Show character counts.What it produces: Conversion-oriented meta descriptions split by approach (benefit vs. pain). Test both angles in Google Search Console to see which drives higher CTR for your audience. Pro tip: Include pricing or a specific number when possible (“From $49/mo” or “Join 10,000+ teams”). Specificity in meta descriptions signals credibility.
Here are 10 pages that need meta descriptions. For each page I'll give you the title, URL, target keyword, and a 1-sentence summary. Write a unique meta description for each (150-160 characters, include keyword, end with action hook). Format as a table: Page Title | Keyword | Meta Description | Character Count. Pages: [paste list]What it produces: A bulk meta description sheet ready for implementation. We used this prompt to write meta descriptions for 200+ pages in under 2 hours for a site migration project. Pro tip: Add “Each meta description must be clearly different from the others. No repeated sentence structures or phrases across the 10.” This prevents the pattern repetition ChatGPT falls into on batch tasks.
Here are 5 existing meta descriptions that aren't performing well (CTR below industry average). For each: diagnose why it might be underperforming (too long, no CTA, keyword missing, bland, etc.), rewrite it at 150-160 characters, and explain what you changed and why. Current meta descriptions: [paste with page URLs and target keywords]What it produces: A diagnostic-plus-fix for underperforming meta descriptions. The “why it’s underperforming” analysis teaches your team to spot problems proactively. Pro tip: Pull actual CTR data from Google Search Console and include it: “This page gets 2.1% CTR at position 4. Industry average at this position is 6.5%. Fix this meta description.”
Write a product description for [product name]. Here are the features: [list features]. For each feature, translate it into a customer benefit. Structure as: 1-sentence hook, 3-4 benefit paragraphs (40-60 words each), specifications section, and a CTA. Target keyword: "[keyword]". Tone: [professional/casual/luxe]. The buyer is a [persona] who cares about [primary purchase driver].What it produces: A benefit-led product description. The feature-to-benefit translation is where most product copy fails: it lists specs instead of explaining why the customer should care. Pro tip: Add “Never use the phrase ‘perfect for’ or ‘designed to.’ Show the benefit in action instead of claiming it.”
Write category page copy for a [type] e-commerce store's "[category name]" page. Include: an SEO-optimized intro paragraph (80-100 words) with the keyword "[keyword]", a buying guide section (150 words) explaining what to look for when shopping this category, 3-4 FAQ questions shoppers ask about [category], and a brief closing CTA. Don't describe specific products. Focus on the category's value and what differentiates good from bad options.What it produces: Category-level content that ranks for broader terms and answers the “what should I look for?” question shoppers have before they click into individual products. Pro tip: Include “[Category] averages 12,000 monthly searches. Competitors ranking for this term have 200-400 words on their category pages.” This context produces more strategic copy.
Write two versions of a product description for [product]. Version A: lead with the emotional benefit (how it makes the customer feel or what it enables). Version B: lead with the rational benefit (specifications, measurable outcomes, comparisons). Both versions: 120-150 words, include "[keyword]", end with a CTA. The customer is a [persona]. Explain which version you'd test first for this audience and why.What it produces: Two testable product descriptions with different psychological frames. The recommendation at the end provides a testable hypothesis. Pro tip: Luxury and lifestyle products usually perform better with emotional leads. Technical and B2B products perform better with rational leads. But test both anyway.
Write an Amazon-style product listing for [product]. Include: a keyword-rich title (under 200 characters), 5 bullet points (each starting with a benefit in ALL CAPS followed by an explanation, max 250 characters per bullet), and a product description paragraph (200 words). Target keywords: "[keyword 1]", "[keyword 2]", "[keyword 3]". The listing must read naturally while including all target keywords at least once. Competing products sell at [$X-$Y]. Our price is [$Z].What it produces: A marketplace-optimized listing. Amazon’s A9 algorithm weights title keywords heavily, so the title structure matters more than it does in standard product pages. Pro tip: Feed ChatGPT 2-3 competing listings and say “Match this format but differentiate on [your unique selling point].”
I need product descriptions for 5 products in the same category. For each, I'll give: product name, 3-5 features, target keyword, and price. Write each description at 100-130 words with: a unique opening hook (no two descriptions should start the same way), feature-benefit pairs, and a CTA. Maintain consistent brand voice across all 5 but make each description distinct. Products: [paste list]What it produces: Five descriptions with consistent tone but varied structure. The “no two should start the same way” constraint prevents the copy-paste feel that batch product descriptions often have. Pro tip: Process in batches of 5-10. Larger batches cause ChatGPT to fall into patterns. If you have 50 products, run 10 batches of 5.
Write a newsletter content block (80-120 words) about "[topic]". Structure: 1-sentence hook that creates curiosity, 2-3 sentences of the key insight or takeaway (give enough value that non-clickers still learn something), and a CTA link sentence. Don't use "click here." Use action verbs like "Read the full breakdown" or "Get the template." The newsletter audience is [describe]. The brand voice is [describe].What it produces: A newsletter section that delivers value and drives clicks. The “give value even without clicking” approach builds trust and keeps subscribers engaged long-term. Pro tip: Write 3 newsletter blocks at once: “Write 3 newsletter content blocks about 3 different topics: [A], [B], [C]. Each block should have a different opening structure.”
Write a 3-email sales sequence for [product/service] targeting [audience]. Email 1 (problem awareness): subject line + 150-word body highlighting the problem without pitching. Email 2 (solution introduction): subject line + 150-word body introducing the approach with a case study or data point. Email 3 (direct offer): subject line + 120-word body with clear pricing, social proof, and a deadline. All subject lines under 45 characters. All emails end with a single CTA (not multiple).What it produces: A three-part sequence that moves from problem to solution to offer. Each email has one job, which is the single biggest differentiator between effective and ineffective email sequences. Pro tip: For more email sequence structures, check our welcome email template and cold email template.
Generate 10 email subject lines for an email about "[topic]" sent to [audience]. Requirements: each under 45 characters, 3 using curiosity, 3 using specificity (numbers or data), 2 using urgency (without being spammy), and 2 using personalization placeholders ([First Name]). Show character counts. No exclamation marks. No ALL CAPS words. Rank by predicted open rate and explain your top 3.What it produces: Ten categorized subject lines. The constraint against exclamation marks and ALL CAPS prevents spam-trigger patterns that hurt deliverability. Pro tip: Subject lines between 20-40 characters perform 45% better than longer ones on mobile devices, where over 70% of email opens happen (Litmus, 2025).
Write a 2-email cart abandonment sequence for [store type]. Email 1 (sent 1 hour after abandonment): subject line (under 40 chars), 80-word body reminding them what they left, addressing the #1 reason people abandon carts in this category (usually [reason]), and a CTA to complete purchase. Email 2 (sent 24 hours later): subject line, 100-word body adding urgency or an incentive, a customer review or trust signal, and a final CTA. Both emails should feel helpful, not pushy.What it produces: A two-part recovery sequence. Cart abandonment emails recover 3-14% of abandoned carts depending on the industry (Baymard Institute, 2024). Check our abandoned cart email template for the full framework. Pro tip: Add “Include a dynamic product image placeholder and the product name in the subject line” for higher personalization.
Write a re-engagement email for subscribers who haven't opened in 90+ days. Subject line (under 40 chars, pattern-interrupt style). Body (100 words max): acknowledge the silence without guilt-tripping, remind them of 1 specific benefit of staying subscribed, offer a quick win (a free resource, a useful tip), and give an easy unsubscribe option. Tone: respectful and direct. The brand is [describe].What it produces: A win-back email that respects the reader’s time. The explicit unsubscribe mention actually improves deliverability by cleaning inactive subscribers naturally. Pro tip: Subject lines that work for re-engagement: “Still interested?” and “We’re cleaning our list” consistently outperform “We miss you!” which feels manipulative.
Write a case study draft using the Problem-Solution-Result framework. Here's the raw data: Company: [name or industry], Problem: [describe], What we did: [describe approach], Results: [list metrics]. Structure: headline with the key result number, 1-paragraph executive summary, Problem section (200 words, include business context and what they tried before), Solution section (300 words, our specific approach with tools and tactics named), Results section (200 words, with before/after metrics in a table), and a client quote placeholder. Total: 700-900 words.What it produces: A structured case study draft. The Problem-Solution-Result format is the most scannable and the most commonly requested by sales teams for prospect conversations. Pro tip: Always include what the client tried before you. “They’d already spent $30,000 on [competitor approach] with no results” makes your outcome more impressive by contrast.
Generate 8 case study headlines for this result: [client type] achieved [key metric] in [timeframe] by [approach]. Use these formats: 2 with the number first ("340% increase in..."), 2 with the client industry first ("[Industry] brand achieves..."), 2 as "How [client type]..." stories, and 2 as before-and-after contrasts. Keep each under 70 characters. The headline must make a VP of Marketing want to read the full case study.
What it produces: Eight headline angles. The “VP of Marketing” audience constraint prevents generic headlines and forces specificity.
Pro tip: Test your top 2 headlines as LinkedIn post openings. Whichever gets more engagement likely works better as the case study headline too.
Write a mini case study (200 words max) for sales enablement. This needs to fit in a slide or email. Include: client industry (no name), the key challenge in 1 sentence, what we did in 2 sentences, the primary metric result in bold, and a secondary supporting metric. End with one sentence on timeline. No fluff. Every sentence must either state a problem, an action, or a result.What it produces: A compact case study for decks, emails, and proposals. Sales teams need proof that fits in a slide, not a 2,000-word narrative. Pro tip: Keep a library of 8-10 mini case studies organized by industry and challenge type. Our proposal template has a section specifically for these.
Generate 12 interview questions for a case study interview with [client role] at [client industry]. The project involved [brief description]. Split questions into: Background (3 questions about their situation before), Process (4 questions about working together and the approach), Results (3 questions about outcomes and metrics), and Recommendation (2 questions about what they'd tell others). Questions should elicit specific, quotable responses, not yes/no answers. Start each with "Can you describe," "What was," or "How did."What it produces: A case study interview guide. The question format (“can you describe” vs. “did you like”) pulls detailed, usable quotes instead of one-word answers. Pro tip: Record the interview and use an AI transcription tool, then feed the transcript back to ChatGPT: “Turn this interview transcript into a 700-word case study using Problem-Solution-Result format.”
Generate 15 FAQ questions and answers for a [business type] about [topic]. Organize by customer journey stage: pre-purchase (5 questions about evaluation and comparison), purchase (5 questions about process, pricing, and logistics), and post-purchase (5 questions about usage, support, and returns). Each answer: 2-3 sentences, under 80 words, written in first-person plural ("we" for the business). Include the target keyword "[keyword]" naturally in at least 4 answers.
What it produces: 15 structured FAQs ready for a product page, help center, or FAQ schema. The journey-stage grouping ensures coverage across the full customer experience.
Pro tip: Pull real questions from your support tickets, live chat logs, or Google Search Console’s query report. Feed those to ChatGPT: “Here are 20 questions our customers actually ask. Write clear, concise answers for each.”
Write a knowledge base article explaining "[topic/feature]" for [product] users. Structure: title (clear, not clever), 1-sentence summary (what this article covers), prerequisites (what the user needs before starting), step-by-step instructions (numbered, with expected outcome at each step), troubleshooting section (3 common issues and fixes), and related articles placeholder. Reading level: grade 8. Under 500 words. Include screenshots placeholders with descriptions of what each screenshot should show.What it produces: A structured help article with built-in troubleshooting. The screenshot placeholders ensure the final article includes visual guidance. Pro tip: Add “Write as if the user is slightly frustrated and just wants the fix. No cheerful language like ‘Great question!’ or ‘Happy to help!'”
I need 5 FAQ entries for my [type of page] about "[topic]" that will be marked up with FAQPage schema. Each answer must be: factually accurate, self-contained (makes sense without reading the rest of the page), between 50 and 150 words, and optimized for Google's featured snippets. Questions should match how real people search (conversational, specific). Include the keyword "[keyword]" naturally in 3 of the 5 answers. Format the output as: Question / Answer pairs.What it produces: Schema-ready FAQ content designed for rich results. The self-contained constraint ensures each answer works when extracted by Google and displayed independently in search results. Pro tip: Check Google’s “People Also Ask” for your keyword. If your FAQ questions match PAA queries, you can rank in both the FAQ rich result and the PAA box simultaneously.
Write FAQ-style responses to 6 common objections for [product/service]. Objections to address: price ("Too expensive"), timing ("Not the right time"), trust ("How do I know this works?"), alternatives ("Why not just use [competitor]?"), complexity ("Seems hard to implement"), and risk ("What if it doesn't work?"). Each answer: 60-100 words, acknowledge the concern (don't dismiss it), provide evidence (data, testimonials, guarantees), and end with a gentle next step.
What it produces: Sales-oriented FAQ content that handles objections before the prospect even raises them. These work on pricing pages, comparison pages, and in sales email sequences.
Pro tip: The “acknowledge first” instruction is critical. Responses that start with “That’s understandable” convert better than those that start with “Actually, we’re not expensive because…”
Turn this blog post into a 10-slide LinkedIn carousel script. Slide 1: attention-grabbing title (under 8 words, big and bold). Slides 2-9: one key point per slide (headline + 2-3 supporting sentences, under 40 words per slide). Slide 10: summary + CTA (follow, comment, or link). Each slide must make sense standalone. Use the blog post content only, don't add new information. Blog post: [paste]What it produces: A LinkedIn carousel script ready for design. Carousels get 3x more engagement than text posts on LinkedIn (Hootsuite, 2025). Pro tip: Design tip: use your brand colors, one sans-serif font at 24px minimum, and lots of white space. Carousels with more than 30 words per slide lose readers by slide 4.
Convert this 2,000-word guide into a 5-email educational series. Each email: subject line (under 45 chars), 200-word body covering one section of the guide, one actionable takeaway the reader can implement immediately, and a teaser for the next email. Email 5 should include a CTA to [specific action]. The audience is [describe]. Don't just split the guide into 5 parts. Restructure each email to deliver standalone value. Guide content: [paste]What it produces: A drip campaign from existing content. The “standalone value” constraint prevents the lazy approach of just chopping a blog post into 5 pieces. Pro tip: Use this for lead magnet sequences. Offer the full guide as a PDF download, then deliver 5 emails that expand on different aspects with additional context not in the original guide.
Here's a transcript from a [topic] webinar. Convert it into a structured blog post (1,500-2,000 words). Clean up the conversational language, remove filler phrases, organize by topic (not chronologically), add H2 headings as questions, and include key quotes from the speaker attributed by name. Add an introduction and conclusion that weren't in the webinar. Target keyword: "[keyword]". Transcript: [paste]What it produces: A publishable blog post from a webinar transcript. Organizing by topic (not chronologically) makes the biggest quality difference since webinars meander, blog posts shouldn’t. Pro tip: Always attribute quotes to the speaker. Named quotes build E-E-A-T and add human credibility that AI-written content lacks.
Convert this blog post into a 5-minute YouTube video script. Structure: hook (10 seconds, the most surprising fact from the article), intro (20 seconds, what this video covers and why the viewer should stay), 3-4 main sections (60-90 seconds each, conversational not formal), and CTA (15 seconds, subscribe + comment question). Include [B-ROLL SUGGESTION] notes for visual transitions. Write it conversational, as if talking to one person. Don't read the blog post, translate it for speaking. Blog content: [paste]What it produces: A video script with timing and visual cues. The “don’t read it, translate it” instruction prevents the robotic delivery that happens when people read blog posts on camera. Pair this with our video script template for structure. Pro tip: Add “Include 2-3 moments where the speaker directly addresses the viewer: ‘Now, here’s where most people get this wrong…'” to create engagement peaks.
This blog post was published [X months/years] ago and needs updating. Current content: [paste]. Update instructions: identify any outdated statistics, tools, or advice and flag them with [OUTDATED: reason]. Add 2-3 new sections covering [recent developments in the topic]. Update the introduction to reflect [current year] context. Suggest 3 new internal linking opportunities to [list recent content]. Keep the original structure but strengthen weak sections. Mark all changes with [NEW] or [UPDATED] so I can review.What it produces: A tracked-changes content refresh. The [OUTDATED] and [NEW] markup lets you review changes efficiently instead of comparing the old and new versions side by side. Pro tip: Content refreshes for posts older than 12 months generate an average 106% increase in organic traffic when done properly (HubSpot, 2024). Refresh your top 10 performing posts quarterly.
| Technique | Time to Apply | Quality Improvement | Best For |
|---|---|---|---|
| Role assignment | 10 seconds | 20-30% | All content types |
| Format specification | 20 seconds | 30-40% | Structured content (emails, listings) |
| Constraint stacking | 30 seconds | 40-60% | Titles, meta descriptions, short copy |
| Example provision | 1-2 minutes | 50-70% | Brand voice matching, style replication |
| Negative instructions | 15 seconds | 25-35% | Eliminating AI-sounding output |
| Iterative refinement | 3-5 minutes | 40-50% | High-stakes content (case studies, proposals) |
40+ prompts for keyword research, content optimization, technical SEO, and competitor analysis. View Prompts →
50+ prompts for PPC, email, analytics, content, and strategy planning. View Prompts →
A structured brief covering keyword targeting, search intent, competitive analysis, and content structure. Get Template →
Technically yes, but the quality drops significantly for posts over 800 words. A better approach: use one prompt for the outline, one for each section, and one for editing. This modular method produces content that reads like a human wrote it, because each section gets focused attention. Single-prompt blog posts typically need 2-3x more editing time.
GPT-4o handles most content writing tasks well and offers the best balance of quality and speed as of March 2026. GPT-4.5 produces slightly more natural-sounding prose but is slower and more expensive. For bulk tasks like meta descriptions or product listings, GPT-4o-mini works fine and costs a fraction. Use the most capable model for high-stakes content and the fastest model for repetitive tasks.
Provide a brand voice guide in your prompt. Include 2-3 examples of your actual writing, specify dos and don’ts (e.g., “Use contractions, don’t use jargon, always be direct”), and reference your tone on a scale (“professional but not stiff, 3 out of 5 on formality”). The few-shot approach with real examples works better than describing your voice in abstract terms.
There’s no legal requirement in most jurisdictions as of March 2026, but transparency builds trust. Google’s stance is that content quality matters, not the production method. If AI wrote it but a human reviewed, fact-checked, and edited it, the content meets quality standards. The FTC does require disclosure for AI-generated endorsements and testimonials specifically.
For a 2,000-word blog post, plan for 6-10 prompts: 1 for the outline, 1 per major section (4-6), 1 for the intro, 1 for the conclusion, and 1-2 for editing and refinement. This modular approach produces better results than trying to get everything in a single prompt. Budget 30-45 minutes total including review and editing.
Our content practice combines AI-assisted writing with editorial judgment and SEO engineering. We build content systems that produce consistent results across hundreds of pages. Explore Our Content Practice →
What ChatGPT prompts write social media captions and posts?
Social posts require different writing muscles than long-form content. Brevity, hooks, and platform-specific formatting all matter. These 4 prompts cover the most common social content needs. For a full collection of 40+ platform-specific social prompts, see our ChatGPT Prompts for Social Media guide.26. LinkedIn Thought Leadership Post
The prompt: What it produces: A LinkedIn post structured for the algorithm. The first line matters most because LinkedIn truncates after roughly 140 characters with a “…see more” link. Pro tip: Add “Start with a bold, declarative sentence. Not a question. Not ‘I’m excited to share.’ A statement that makes someone stop scrolling.”27. Instagram Caption with CTA
The prompt: What it produces: A platform-formatted caption with scroll-stopping hook and algorithm-friendly engagement prompt. Pro tip: Specify the CTA type. “Ask them to save this post” works better for educational content. “Ask them to comment” works better for opinion-based content. Each CTA type signals different things to Instagram’s algorithm.28. Twitter/X Thread
The prompt:Write a 7-tweet thread about "[topic]". Tweet 1: hook (bold statement or surprising fact, under 200 characters). Tweets 2-6: one key point per tweet, each under 250 characters, each starting with a number ("2/"). Tweet 7: summary + CTA (link or follow). Use line breaks within tweets for readability. No emojis in every tweet. Maximum 2 tweets with emojis. Include a "bookmark this thread" CTA in tweet 1.What it produces: A structured thread that reads well both as a sequence and when individual tweets are retweeted out of context. Pro tip: Thread hooks that start with “Most people think [X]. They’re wrong.” consistently outperform “Here’s everything you need to know about [X].”29. Cross-Platform Repurpose
The prompt: What it produces: Four distinct social posts that each stand alone. The “different insight per platform” constraint prevents the lazy approach of posting the same message everywhere. Pro tip: This pairs well with a social media calendar template for scheduling distribution across your channels.