Listicle
May 202615 min read

Best ChatGPT Prompts for LinkedIn in 2026: 17 Prompts Solopreneurs Actually Use

Copy-and-paste task prompts for LinkedIn hooks, full posts, comments, profile rewrites, repurposing and batching. Each one is built to be paired with a voice prompt prefix, because that is the only way ChatGPT produces output that sounds like you.

Two-part rule for every prompt below: voice prompt at the top, task prompt at the bottom. The 17 prompts in this article are task prompts. They produce generic content on their own. They produce on-voice content when prefixed with a 500-800 word voice prompt that defines how you write. Build the voice prompt once. Reuse it across all 17 task prompts forever.

Why most "best ChatGPT prompts for LinkedIn" lists fail

The standard listicle gives you a clever-sounding prompt and tells you to paste it into ChatGPT. The output is the same generic LinkedIn voice everyone else gets. The reader concludes that ChatGPT is the problem. ChatGPT is not the problem. The missing voice prompt is the problem.

A task prompt asks ChatGPT to do something. A voice prompt tells it how to do it. Without the voice prompt, ChatGPT defaults to its training average, which is exactly what makes generic AI content recognisable. With a voice prompt, the same task prompt produces output that sounds like you 70-85% of the time on first draft. How to build a voice prompt covers the structure.

How to use the 17 prompts in this article

Recommended workflow:

  1. Build a voice prompt using the five-section structure.
  2. Create a Custom GPT and paste the voice prompt into the instructions field.
  3. Add 4-6 of the task prompts below as conversation starters.
  4. Use the Custom GPT for every LinkedIn task. The voice prompt loads automatically every time.

Alternative for occasional users: paste the voice prompt as the first message in a new ChatGPT conversation, then paste the task prompt below it. Slower but works without a Plus subscription.

Hooks

PROMPT 1

20-hook generator from one topic

USE: HOOK BATCHING · OUTPUT: 20 HOOK OPTIONS · BEST FOR: WEEKLY PLANNING

Generate 20 LinkedIn hooks for the topic: [TOPIC]. Constraints: - Each hook is one or two sentences, max 22 words. - Five hooks open with a specific number, dollar amount, or named scenario. - Five hooks name the reader's pain before naming any solution. - Five hooks take a clear point of view (no hedging). - Five hooks use a contrarian or counterintuitive angle. No hashtags. No emojis. No openers like "I've been thinking about" or "Let me tell you about". Output as a numbered list.

Why it works: forces variation across the four hook patterns that consistently outperform on LinkedIn (number, pain, point of view, contrarian). Twenty options is enough to find three you actually like.

PROMPT 2

Hook rewrite from a generic opener

USE: SAVING A POST WITH A WEAK OPENER · OUTPUT: 5 ALTERNATIVES · BEST FOR: POST EDITING

Here is a LinkedIn post that loses readers in the first line: [PASTE POST] Rewrite the first sentence five different ways. Keep the body of the post unchanged. Each rewrite must: - Be under 18 words - Open with a specific detail (number, named scenario, or concrete observation) - Avoid the words "I", "we", "you" in the first three words - Feel like a continuation of the existing voice Output the five hooks as a numbered list with a one-sentence note explaining what each hook is doing.

Why it works: most LinkedIn posts have a strong middle and a weak opener. Replacing the first line is the highest-leverage edit you can make.

PROMPT 3

Hook library extraction from your past wins

USE: BUILDING A REUSABLE LIBRARY · OUTPUT: STRUCTURED HOOK PATTERNS · BEST FOR: ONE-TIME EXTRACTION

Here are five LinkedIn posts of mine that performed above my average: [PASTE 5 POSTS] Extract the hook from each post. Then identify the structural pattern of each hook (e.g. "specific number plus contrarian observation", "named scenario plus question", "personal admission plus reframe"). Output a table with three columns: post number, hook, hook pattern. Then propose 10 new hooks following the same patterns but on different topics: [LIST 3-5 TOPICS YOU WANT TO COVER].

Why it works: builds a hook library from patterns that already worked for you specifically, rather than from generic best-practice lists. Detail in how to build a LinkedIn hook library with AI.

Full posts

PROMPT 4

Story-to-insight post

USE: PERSONAL EXPERIENCE POSTS · OUTPUT: 180-220 WORD POST · BEST FOR: TUESDAY/THURSDAY POSTS

Write a LinkedIn post in my voice based on this story: [2-3 SENTENCE STORY OR SITUATION] Structure: - Hook (1-2 sentences, specific detail) - Setup (2-3 short sentences) - Turning point or insight (1-2 sentences) - Implication for the reader (1-2 sentences) - Closing line (1 line, provocative or open) Word count: 180-220. No hashtags. No call to action that asks the reader to comment or DM. Vary sentence length deliberately, with at least three sentences under 8 words.

Why it works: forces the post into the structure that performs on LinkedIn (story plus insight plus implication) and limits the AI's tendency to balloon into 400-word essays.

PROMPT 5

Contrarian opinion post

USE: POSITIONING POSTS · OUTPUT: 150-200 WORD POST · BEST FOR: STAKING A CLEAR POSITION

Write a contrarian LinkedIn post in my voice on the topic: [TOPIC]. Constraints: - Open with the contrarian claim in the first sentence (under 16 words). - One paragraph of evidence or reasoning. - One paragraph naming the standard view and why it is wrong. - One closing line. - Word count: 150-200. The post must take a clear side. No hedging language ("it could be argued", "in many cases"). No safe-middle-ground conclusions. The reader should know where I stand by sentence one and still know it at the end.

Why it works: a clear point of view is one of the seven LinkedIn patterns that consistently outperforms. Detail in LinkedIn AI posts without suppression.

PROMPT 6

List-format post (without sounding like a corporate listicle)

USE: TACTICAL CONTENT · OUTPUT: 200-250 WORD POST · BEST FOR: HOW-TO ANGLES

Write a LinkedIn post in my voice that lists [N] things related to [TOPIC]. The post must read as a tight observation, not a corporate listicle. Structure: - Hook (one line setting up the list) - The list itself: each item is one or two sentences max, with a concrete detail - One closing line that reframes or provokes Banned: bullet points formatted with emojis or asterisks. Banned: "in this fast-paced world" or any version of it. Banned: a closing line that asks the reader to comment. Word count: 200-250.

Why it works: lists perform on LinkedIn but the default AI list format is exactly the corporate-listicle output that gets scrolled past. The constraint forces a tighter, more reader-aware version.

PROMPT 7

Lesson-from-failure post

USE: VULNERABILITY-DRIVEN POSTS · OUTPUT: 180-220 WORD POST · BEST FOR: BUILDING CREDIBILITY

Write a LinkedIn post in my voice about a specific time something I tried did not work. Use this raw material: [2-3 SENTENCE FAILURE] Structure: - Hook: one specific line about the failure (no abstractions) - Body: what I tried, what happened, what I learned - Reframe: one line connecting the lesson to a wider principle - Closing: one line that does not ask for engagement Banned: "humble brags" disguised as failures. Banned: lessons that conclude with "trust the process". The post should leave the reader with a specific, transferable observation, not a platitude. Word count: 180-220.

Why it works: vulnerability builds credibility but only when the failure is specific. The constraints filter out the humble-brag pattern.

Comments

PROMPT 8

High-signal comment generator

USE: COMMENTING ON OTHERS' POSTS · OUTPUT: 3-5 SENTENCE COMMENT · BEST FOR: NETWORK BUILDING

Here is a LinkedIn post I want to comment on: [PASTE POST] Write a comment in my voice that: - Adds a specific observation, example, or counterpoint - Is 3-5 sentences (no longer) - Does not start with "Great post" or any version of agreement - Does not pitch my services or ask the author to DM me - Includes one concrete detail (number, named example, or specific situation) The comment should leave the original poster wanting to reply.

Why it works: agreement comments get ignored. Comments with a specific addition or counterpoint earn replies and visibility from the original poster's audience.

PROMPT 9

Comment-to-DM bridge

USE: WARMING UP A PROSPECT · OUTPUT: COMMENT + DM · BEST FOR: SALES PIPELINE

I want to comment on this LinkedIn post and then DM the author: [PASTE POST] Author context: [1-2 SENTENCES ABOUT WHO THEY ARE AND WHY I CARE] Write: 1. A 3-4 sentence comment in my voice that engages with their content (not their service or product) 2. A DM I would send 24-48 hours later that references the comment naturally and asks one specific question (not a sales pitch) The DM must feel like a continuation of a conversation, not the start of a sales sequence.

Why it works: replaces the "engage publicly then sell privately" pattern with "engage publicly then continue privately". Lower conversion rate per touch but higher reply rate.

Profile

PROMPT 10

Headline rewrite (5 versions)

USE: PROFILE OPTIMISATION · OUTPUT: 5 HEADLINES · BEST FOR: ONE-TIME REWRITE

Here is my current LinkedIn headline: [CURRENT HEADLINE] Here is what I do (longer description): [2-3 SENTENCES] Here is the audience I want to reach: [1-2 SENTENCES ON YOUR ICP] Rewrite the headline 5 different ways. Each version must: - Be under 220 characters - Lead with what I do for the audience, not my job title - Include one concrete signal of credibility (years, specific result, named outcome) - Avoid "passionate about", "thought leader", "helping businesses unlock" Output as a numbered list with a one-line note on what each version emphasises.

Why it works: the headline is the highest-leverage piece of profile real estate. Five versions lets you compare angles before committing.

PROMPT 11

About section that reads like a person, not a CV

USE: PROFILE REWRITE · OUTPUT: 3-PARAGRAPH ABOUT SECTION · BEST FOR: REPOSITIONING

Write a LinkedIn About section in my voice with this raw material: [3-5 SENTENCES ABOUT YOUR BACKGROUND, WHAT YOU DO, WHO YOU SERVE, AND ONE SPECIFIC RESULT OR STORY] Constraints: - 3 paragraphs, 80-110 words each - Open with a hook line (no "I am a..." opener) - Middle paragraph names the audience and their specific situation - Closing paragraph names what to do next (visit a site, follow for content, etc.) - No bullet points - No emojis - No "I'm passionate about" or "I help businesses..." - Read it back: it should sound like the person in paragraph 1 wrote paragraphs 2 and 3

Why it works: most LinkedIn About sections fail because they switch into corporate-CV register halfway through. The constraint enforces consistent voice across all three paragraphs.

Repurposing

PROMPT 12

Blog-to-LinkedIn series

USE: REPURPOSING WRITTEN CONTENT · OUTPUT: 5 LINKEDIN POSTS · BEST FOR: WEEKLY DISTRIBUTION

Here is a blog post I published: [PASTE BLOG POST OR LINK CONTEXT] Extract 5 distinct angles from it. For each angle: - Write a 150-180 word LinkedIn post in my voice - Each post must stand alone (a reader who has not read the blog should still get value) - Each post must end with a soft pointer to the full blog post (not a hard CTA) - Each post takes a different angle: contrarian, story, list, observation, question Output the five posts in sequence, each labelled with the angle it uses.

Why it works: blog content is too dense for LinkedIn. Extracting five angles produces five posts that can be staggered across two to three weeks. More detail in how to repurpose one post into ten.

PROMPT 13

Podcast or interview to LinkedIn posts

USE: PODCAST REPURPOSING · OUTPUT: 5-10 POSTS · BEST FOR: PODCASTERS AND INTERVIEW GUESTS

Here is a podcast transcript or interview I did: [PASTE TRANSCRIPT OR KEY QUOTES] Extract 5-10 LinkedIn post ideas, each anchored on a specific moment in the conversation. For each post: - Identify the specific quote or moment that anchors it - Write a 150-200 word post in my voice that builds out the idea - Include the original framing (e.g. "On a recent podcast I said..." only if it adds context, not as a default opener) Output as: post number, anchor moment, post body.

Why it works: podcasts are dense. Anchoring each post on a specific moment is the difference between a useful post and a generic summary. Detail in repurpose a podcast into 30 LinkedIn posts.

PROMPT 14

Long thread to single post

USE: SQUEEZING A LARGE IDEA · OUTPUT: ONE 200-WORD POST · BEST FOR: ARGUMENTATIVE CONTENT

I have a long argument or thread I want to compress into one LinkedIn post. Here is the raw material: [PASTE LONG-FORM ARGUMENT] Write a single LinkedIn post in my voice that: - Captures the core point in the first sentence - Includes the strongest single piece of evidence - Closes with an implication for the reader Word count: 180-220. The post should make the reader feel they got the whole argument, even though most of the long-form material was cut.

Why it works: compression forces clarity. Most argumentative posts fail because they include three points instead of one strong one.

Batching and planning

PROMPT 15

Weekly content plan from one theme

USE: WEEKLY PLANNING · OUTPUT: 5-POST PLAN · BEST FOR: BATCHING SUNDAY EVENING

My weekly LinkedIn theme is: [THEME]. My audience is: [AUDIENCE]. Plan 5 LinkedIn posts for the week. Each post must be a different angle on the theme: - Monday: tactical (a how-to or framework) - Tuesday: story (a specific personal example) - Wednesday: contrarian (a clear opinion) - Thursday: observation (something niche I noticed) - Friday: question (something that invites comments without begging) For each, output: post type, hook (one line), and three-sentence outline of the body.

Why it works: a five-post plan with deliberate angle variation prevents the "same post in different words" problem that kills LinkedIn engagement.

PROMPT 16

Month-of-content audit and gap finder

USE: MONTHLY REVIEW · OUTPUT: AUDIT + 5 NEW IDEAS · BEST FOR: AVOIDING REPETITION

Here are my last 20 LinkedIn posts: [PASTE OR LIST POSTS] Audit them. Identify: - The 3 themes I have leaned on most - The 3 themes my audience would expect me to cover but I have not - Patterns I am repeating (hook structures, post structures, closing lines) - Voice elements that are working (specific phrases, signature moves) Then propose 5 new LinkedIn post ideas that fill the gaps without contradicting the patterns that are working.

Why it works: AI is better at pattern detection across 20 posts than humans are. Use it for the audit, then write the gap-filling content yourself.

PROMPT 17

Voice match check on a draft

USE: PRE-PUBLISH SANITY CHECK · OUTPUT: AUDIT REPORT · BEST FOR: CATCHING DRIFT

Here is a draft LinkedIn post I want to publish: [PASTE DRAFT] Run a voice match check against the voice prompt above. Report: - Sentences that drift from the voice (quote them and explain why) - Banned words or phrases that slipped in - Hook strength (rate 1-5 and explain) - Closing line strength (rate 1-5 and explain) - One specific edit that would lift the post the most Be honest. Generic praise is useless.

Why it works: AI is willing to critique its own output if you ask it to. Used pre-publish, this catches voice drift that the writer no longer notices. More on this in AI content that doesn't sound like AI.

What ties all 17 prompts together

Notice the pattern across the list. Every prompt:

That structure is what differentiates a prompt that works from a prompt that produces generic output. The voice prompt prefix carries the "how it should sound" layer. The task prompt carries the "what to produce" layer. Together, they produce content that sounds like you and serves a specific purpose.

Where to go from here

If you do not have a voice prompt yet, start there. How to build a voice prompt walks through the five-section structure used at Syxo. How to train ChatGPT on your writing style covers the Custom GPT setup.

If you already have a voice prompt and you want a deeper understanding of why these prompts produce different output: the complete guide to AI voice prompts covers the full system.

Related reading

All 17 prompts, paired with a voice prompt that sounds like you

The DFY Voice System builds your voice prompt, sets up the Custom GPT, and includes a starter library of task prompts already calibrated to your voice. The Voice Build methodology, applied to your existing writing. £497 founder pricing. Delivered in 2-3 working days.

See The Voice Build

Frequently Asked Questions

What is the best ChatGPT prompt for LinkedIn posts?

There is no single best prompt. The strongest output comes from layering a voice prompt (defines how you write) with a task prompt (defines what to produce). The 17 prompts above are task prompts.

Should I use a Custom GPT or paste prompts each time?

For weekly or daily LinkedIn content, build a Custom GPT and store the voice prompt in the instructions. For occasional use, copy-paste works.

Why do my ChatGPT LinkedIn posts sound generic?

No voice context, default vocabulary, structural sameness. Fix all three by adding a voice prompt prefix and banning the default words explicitly.

How long should a ChatGPT prompt for LinkedIn be?

Task prompts: 50-150 words. Voice prompts: 500-800 words and live separately in Custom GPT instructions.

Which version of ChatGPT works best?

GPT-4o and the GPT-4 family follow voice prompts more reliably than GPT-3.5. Plus subscription pays back inside the first week of serious use.

Do these prompts work in Claude or Gemini?

Yes. The task prompts are model-agnostic. The voice prompt prefix travels across tools.