LinkedIn maxxing with Claude: 12 prompts for a content system that does the hard parts

LinkedIn maxxing with Claude means running your LinkedIn writing as a system inside one Claude Project: 12 prompts cover research, a voice guide, drafting, editorial review and reporting, while publishing, comments and messages stay in your own hands.

Hand AI a topic and ask for a LinkedIn post, and it skips the work that makes a post worth reading: knowing your customers, finding an interesting angle, collecting evidence and knowing what you can credibly claim.

The 12 prompts below turn that work into a repeatable routine, from researching your market to planning a content calendar and seeing which posts lead to real business conversations.

What you'll end up with:

• A voice guide built from how you really write and speak.
• A research routine that finds good ideas before you need them.
• A drafting and review process with clear quality standards.
• A reporting loop that tells you what to write next.

Claude fits this kind of system because one Project can hold your files, instructions and memory, and a single conversation can search the web, read your uploads and return finished documents and spreadsheets.

Bookmark this page for your next planning session.

What should you load into Claude before the first prompt?

Claude needs more than your job title and a request to "sound human". Give it enough to understand who you sell to, what you believe and which claims you can back up.

Paste this first:

Treat everything below as the standing context for my LinkedIn writing. Store it as a reusable reference document and tell me where you saved it. Business: [company name, website, product or service, usual length of a sale]. Audience: [decision roles, company sizes, industries, regions]. They mostly struggle with [problems], push back with [objections], and start looking for help when [triggers]. Positioning: we help [audience] reach [outcome] through [approach]. What sets us apart: [differences]. We are the wrong choice for [disqualifiers]. Opinions: I hold these three views about my industry: [views]. I disagree with [common advice] because [reason]. Evidence: you may use these approved case studies, results, examples and personal experiences: [sources]. Never make up customers, revenue numbers, stories or quotes. Voice: these [10-20 posts] and [transcripts or writing samples] sound like me. Steer clear of [phrases, tones, formatting habits]. Goals: my main aim on LinkedIn is [qualified conversations / deal support / hiring / awareness]. My offer is [offer], and I want interested readers to [action]. Rules: use connected sources only where you have authorized access, and if you can't open a source, ask me for an export or pasted text. Publishing, comments, connection requests and messages on LinkedIn stay with me. Put drafts in [review location], keep evidence apart from inference, and flag anything that needs my input.

Update this context whenever your offer, audience or opinions change. In Claude, replace the file in the Project each time, so new conversations never start from an outdated version.

Where does Claude keep this context?

A brief pasted into one chat shouldn't depend on that chat, so give it a permanent home before prompt 1.

Quick reference:

• One Project for LinkedIn. Set up a single Project for the whole workflow and run every prompt inside it. A Project has its own files, links, instructions and memory, available in every conversation you open there. Conversations started outside it can't see those files.
• Project files for long material. Ask Claude to return the reference document from the prompt above as a file, then add it (and the voice guide from prompt 1, once it exists) to the Project's files.
• Project instructions for short rules. Copy the "Rules" part of the context into the Project instructions, and it will apply to every conversation in the Project without being pasted again.
• Google Docs that stay up to date. If your positioning lives in Google Drive, add the doc from Files -> "+" -> "Drive". It syncs from Drive, so editing the doc changes the context. The option exists only for private Projects; shared Projects don't get it.
• Instructions for Claude. Preferences that hold for everything you do, like spelling or "no emojis", go in Settings > Your Claude (Settings > Account in the older layout).
• Memory. Claude draws on what it remembers from your chats. If you also write for a client whose details mustn't mix in, switch "Memory" off in the "+" menu before the first message of that conversation.

Part 1: research and content production

Prompts 1 to 8. Run each one inside the LinkedIn Project, so every conversation starts from the same files and instructions.

1. How do you build a voice guide?

A voice guide earns its keep when it explains the choices behind your writing: how you introduce an argument, where examples go, how you move from one idea to the next.

A label like "conversational and authoritative" won't take you far.

Study the posts and transcripts I've supplied and write a practical voice guide for my LinkedIn writing. Cover how I open a post, explain an idea and close an argument; my usual vocabulary, sentence and paragraph lengths, and how often I use lists; how I show disagreement, doubt, humour and confidence; phrases I use without thinking, with word-for-word examples and where each comes from; and habits that show up in my weaker pieces and should not become rules. Add a list of phrases and patterns to avoid, including needless repetition, invented drama, certainty I can't back up and too many one-line paragraphs. End with three before-and-after edits that turn generic copy into my voice while keeping the meaning. Make the guide concrete enough that another writer could follow it.

In Claude: upload your posts as a single text file or paste them in, and attach transcripts as files so they aren't spread across messages. Ask for the guide as a Markdown document. When a line sounds wrong, select it, click "Edit with Claude" and say what to change; the fix goes exactly where you marked. Once the guide sounds like you, add it to the Project's files. When it stops changing from week to week, turn it into a skill: skills teach Claude how you want a job done, and Claude follows them whenever that job comes up. If some of your best posts exist only on LinkedIn, Claude in Chrome can open the pages you name and read them. Choose posts you'd still publish today, since weaker ones teach the wrong habits.

Why it matters: a good voice guide makes feedback reusable. You stop correcting the same phrasing every week and improve the instructions behind every draft.

2. Which customer questions are worth answering?

Sales calls and customer conversations surface questions no keyword tool will show you. The useful ones usually sit around a decision someone is struggling with.

Go through the sales calls, support threads, customer interviews and closed-lost notes I've supplied from [period]. Sort the questions that keep coming up into: problems customers are trying to understand; worries that slow a decision; alternatives they weigh against us; expectations that end in disappointment; and questions asked right before they decide. For each theme, give an anonymized quote, where it came from, how many separate conversations mention it, the person's role and their stage in the decision. Suggest two LinkedIn angles per theme: one that helps a reader diagnose the problem and one that helps them judge a solution. Rank the top 10 by how often they recur, how much they matter commercially and how strong our evidence is. Keep what customers said apart from your reading of it.

In Claude: if your call notes live in Google Drive, switch on the Google Drive connector and use "+" -> "Add from Google Drive" to attach several documents to one conversation; together they have to fit in the conversation's context. Claude's answer says which documents it used and links back to them, which fills the "where it came from" column. In Claude Desktop you can give Claude access to a local folder of transcripts instead. Ask for the ranked themes as a spreadsheet with a tab per theme. Drive files are read as text only, so screenshots inside those docs get skipped. Keep recurring call notes in one Drive folder and add the newest each month, so the counts grow with your pipeline and never restart from zero.

Why it matters: you get a backlog of topics tied to conversations your business already has, in words you'll know and with evidence you can check.

3. How do you find topics moving from X into your market?

A discussion picking up on X can turn into a useful LinkedIn post, as long as you explain why it matters to your customers and add something of your own beyond a rewrite.

Use web search and public sources you can open to review the past [seven days] of discussion about [topics] and from these accounts: [accounts]. Find up to 10 developments or arguments that matter to [ideal customer]. For each one, return the original source, author, date and link; a neutral summary of what happened or what is being argued; the strongest opposing readings; what it changes for our audience; and an angle our own experience or evidence could support. Mark like and view counts as a snapshot, don't treat popularity as proof, and check factual claims against the original source where you can. Recommend three ideas worth developing and say what I would need to add before each could become a credible LinkedIn post.

In Claude: the current Claude app has no web search switch; Claude searches the web when a request calls for it. For a broader sweep, type /deep-research or choose "+" -> "Research". Claude has no built-in X search, so list the account and thread URLs you care about, and if a page can't be read, paste the posts into the chat. Research can call your connected tools without asking each time, so turn off connector tools that write or send before you start. Save the three chosen ideas in a Project file so prompt 6 can use them. When Research is done, ask Claude which sources it actually opened, so you can tell a checked claim from one lifted out of a search snippet.

Why it matters: research hands you a starting place, and your judgment gives readers a reason to follow you. A useful adaptation ties the discussion to a decision your readers really face.

4. Which posts beat their authors' usual results, and why?

Measuring every post against a celebrity account's biggest hit will skew your research. Compare a post with its own author's typical results first, then look at what might explain the gap.

Analyse this LinkedIn swipe file and the performance data that comes with it: [file]. Where there is enough comparable data, compare each post's reactions and comments with the author's median for a similar format and period. Show the calculation, and flag small samples, missing numbers and very recent posts. For the clearest outliers, describe the topic and the reader it was meant for; the opening, the argument, the evidence and the close; the format and rough length; whether the comments suggest real interest from potential customers; and other possible explanations, such as a launch, a news event or unusually wide distribution. Turn this into five reusable structures, then propose an original idea for each based on what I know. Don't copy distinctive wording, and don't present an observed pattern as a rule of the algorithm.

In Claude: upload the swipe file as a spreadsheet or CSV with one row per post. Claude runs code on uploaded data, so it can calculate each author's median, show the formula and return a sortable spreadsheet in place of a paragraph of numbers. If all you have is post links, Claude in Chrome or the browser built into Claude Desktop can open each page you list and read it; you still choose which posts go in the file. To see where your own posts sit against your usual results, run them through the LinkedIn engagement rate calculator first. Ask for one tab with the raw rows and one with the medians, so you can check a single author without redoing the whole sheet.

Why it matters: you come away with hypotheses worth testing, and you avoid assuming that copying a successful post's layout will repeat its results.

5. How do you turn recorded know-how into original posts?

A transcript can hide a strong post inside a story, a passing explanation, or a disagreement that deserved more airtime than it got in the original conversation.

Read this interview, podcast, webinar or meeting transcript: [source]. Find eight moments that could carry an original LinkedIn post, putting decisions, mistakes, surprising observations, practical explanations and opinions backed by experience first. For each moment give the exact quote with its timestamp or location; the main idea and who would gain from it; any context needed so the speaker isn't misrepresented; facts already available to support it; and one follow-up question that would make the idea stronger. Pick the three best moments and draft a post from each using my voice guide. Keep the speaker's level of certainty, strip out confidential details, and leave a placeholder wherever a result or example still needs confirming.

In Claude: run this in the LinkedIn Project so the voice guide is already loaded, then attach the transcript file. Timestamps come back only if your transcript has them, so export it from your recorder with timestamps on. If the transcript is a Google Doc your guest commented on, Claude can read those comments and suggestions as well, which helps with the context column. Ask for the three drafts in one Markdown document with placeholders in square brackets, so you can see at a glance what still needs confirming. For a long webinar, have Claude work through the transcript section by section and report candidate moments for each part before choosing three; that keeps the quotes exact.

Why it matters: you can produce posts from knowledge that already exists, without asking someone to come up with a fresh opinion every time the calendar has a gap.

6. How do you build a four-week content calendar?

A calendar should weigh what customers need to understand against what you have enough evidence to say. Filling every slot with another version of your offer makes the feed repetitive.

Plan four weeks of LinkedIn posts at [three posts a week] from the approved research ideas. Mix in practical explanations that help customers solve a problem; opinions backed by experience or evidence; customer examples we have permission to share; answers to objections and decision questions; and the occasional offer post with a relevant next step. For each post, state the reader, one objective, the main argument, the source material, a suggested format and what the reader should do next. Check for repeated topics, overlapping arguments, unsupported claims and too many sales-heavy posts. Write full briefs for week one and shorter outlines for weeks two to four. Keep one slot a week open for something timely, since the plan will need to change.

In Claude: say "Google Sheet" by name if you want the calendar in Google Sheets. With the Sheets connector on, the file opens in a pane beside the chat and you can edit it while Claude works, whereas asking for a plain "spreadsheet" gets you a downloadable file. Since the research from prompts 2 to 4 sits in the same Project, ask Claude to name the file each idea came from. Choose posting days with our best time to post on LinkedIn tool, and when a brief calls for a carousel, lay out the slides in the LinkedIn carousel maker. Keep the finished calendar in the Project too, so prompt 7 can pull the week-one briefs by name.

Why it matters: a good calendar cuts weekly planning time and still leaves room for better ideas, new evidence and shifts in the market.

7. How do you draft posts that have something to say?

Let the research settle the argument before Claude starts polishing sentences. Skip that order and you get clean prose that takes several paragraphs to reach an obvious idea.

Write three LinkedIn drafts from these approved briefs: [briefs]. Use my business context and voice guide. In each draft, make one clear argument for one reader; open with a specific observation, tension, result or question that the brief supports; explain the reasoning with a concrete example; use bullets where they make parallel items easier to scan; vary the length of sentences and paragraphs, with the odd one-line paragraph for weight; and finish with a useful consequence or a natural next step. Aim for [180-300 words] unless the idea needs a different length. Avoid repeated wrap-ups, invented personal stories, inflated promises and formula contrasts. Give two alternative openings and say briefly how each shifts the emphasis. List missing evidence apart from the copy that is ready to publish.

In Claude: paste the week-one briefs from the calendar, or name the Project file that holds them. Ask for the drafts as a Google Doc if a colleague or client edits with you, or as a Claude Docs living document if your team works inside Claude. For each opening, set Claude's two alternatives next to a few from the LinkedIn hook generator, and if a draft feels flat for its length, run it through the LinkedIn post booster before editing by hand. Ask Claude to put the missing-evidence notes under each draft and keep them out of the body, so the ready copy can be pasted straight out once you fill the gaps.

Why it matters: clear limits give you drafts that are easier to judge. Your review can go into the quality of the argument, with no time lost cutting filler.

8. How do you set up a separate editorial review?

Give the reviewer one defined job: find the weak spots in the draft and explain how to fix them. A comforting "This is strong" doesn't count as an editorial process.

Review these LinkedIn drafts against the brief, the source material, my business context and my voice guide. Score each out of 100: relevance to the customer (25), specificity and supporting evidence (25), strength and clarity of the argument (20), match with my voice (20), readability and formatting (10). Show what each score is based on. Check every factual claim, mark generic passages, and call out repetition, abrupt one-line fragments and paragraphs that would read better as lists. Return a revision ready to publish, a short note on the changes that matter, and any open questions. Treat invented evidence, exposed confidential information or a misquoted source as a blocker whatever the score. Keep everything in the review document for a human to approve.

In Claude: open a new conversation in the same Project for the review, so the reviewer starts from the Project files and the drafts and never sees the reasoning that produced them. The new Claude app doesn't offer branching from an earlier conversation, so a fresh conversation is the clean way to keep the two roles apart. If the drafts are in a Google Doc, Claude can read the comments and suggestions already there. Before approving, paste the revision into the LinkedIn post analyzer and clean up line breaks with the LinkedIn text formatter. Put the rubric in the Project instructions only if every review should use it; otherwise keep it in this prompt, so drafting conversations stay about writing.

Why it matters: a rubric puts your standards in writing. The score organises the feedback, while the source checks and an editor's judgment decide whether the post is ready.

What should happen in the first hour after you publish?

Prompt 8 ends with a revision you approve. You publish it yourself, then write the publishing time next to the post in its Project file, because prompt 11 needs it to compare posts at the same age. Every other maxxing technique comes back to one thing: get as many reactions as you can in the first hour after the post goes live.

Ask the people you can reach. Colleagues, friends, relatives and clients are the best people to ask, since you can actually get hold of them. Send them the link as soon as the post is up.

If you don't have those people, use a community where members act by hand. Pick one that removes automated accounts and lets you report actions that weren't done.

Use the groups below. The LinkedIn pods and groups in the table were checked before they were listed.

Use LinkedInLikes. It's our own free community:

• members like, comment, repost and save by hand from their own LinkedIn accounts;
• comments are written in each member's own words, and the post owner can't dictate them;
• if someone confirms an action they didn't do, you can report it within 72 hours, and we remove accounts we detect as automated;
• we never ask for your LinkedIn password; you only share the public link to the post.

Once the post is live, open LinkedInLikes, paste the post link, choose the action (likes, comments, reposts or saves) and set how many you want; the 100 starter points cover your first requests. Publishing, comments, connection invitations and messages stay with you. Get comments through LinkedInLikes.

Which LinkedIn pods and groups can you join?

Free communities, checked on Oct 21, 2026. Member counts are what each platform showed that day. Every group has its own admins, so read the pinned message before your first post.

To get likes and comments on your own posts without searching through groups, add a request on LinkedInLikes.

Part 2: pipeline and performance

Prompts 9 to 12. These draw on sales and CRM material, so check which connectors are on before each conversation.

9. How do you write content around deals that stopped moving?

Some of your most useful posts may answer a worry that several prospects raised in private. The aim is to explain the issue well enough that future prospects arrive better informed.

Go through the CRM records and call notes I've supplied for qualified deals that stopped moving during [period]. Find the objections that recur and separate objections stated outright in conversations, possible explanations nobody has confirmed, and operational delays that content won't fix. For each repeated objection, note the contact's role, the deal stage, the evidence available and the question a useful LinkedIn post should answer. Recommend five content briefs. Each should present the concern fairly, show where our approach fits, admit where it doesn't, and say what proof is needed. Draft the two strongest posts without naming prospects or revealing anything about the deals. Separately, list approved resources a salesperson could share in a relevant conversation.

In Claude: look for your CRM under Customize > Connectors first. If it isn't in the directory but offers its own remote server, add it as a custom connector. Without a connector, export the deals to a CSV and upload it. Connected tools can change records, so for this prompt keep Claude in Manual mode, or block the CRM's write tools under Customize > Connectors, and let it read only. Ask for the five briefs in a separate file from the two drafts, so deal notes never end up in a document you share. If you work from an export, delete the name and company columns before uploading; it's simpler to keep from Claude anything the post must never mention.

Why it matters: this links marketing with the questions salespeople have to answer again and again. It also keeps you from writing persuasive posts about problems your customers don't care about.

10. How do you prepare briefs for conversations under your posts?

When someone responds to a post, the right next step depends on what they said and whether a business conversation would be welcome. A reaction on its own says nothing about business interest.

Review these LinkedIn conversations I've supplied: [comments, inbound messages or interactions I logged by hand]. Sort them into questions that deserve a useful public answer; conversations where sharing a resource would help; clear interest in our offer; and general discussion with no commercial next step. For each relevant conversation, sum up the person's stated need, the business context we have and what we still don't know. Draft a reply that answers their actual message before bringing up anything else. Where it fits, suggest one follow-up question or a resource we already have. Don't read business interest into a like, don't invent familiarity, and don't force a meeting request into a conversation that doesn't call for one. Keep every reply as a draft for me to review and send.

In Claude: copy the comments and messages into the chat or a text file, because Claude doesn't read your LinkedIn inbox. If a conversation moved to email, the Gmail connector lets Claude find the thread and draft a reply there, and by default Claude asks for your approval before anything is sent. For short public replies under your post, compare Claude's draft with options from the LinkedIn comment generator, then rewrite in your own words before posting. Group what you paste by post, so the summary also shows which posts started useful conversations; prompt 11 can use that list.

Why it matters: preparation makes replies more thoughtful without turning every exchange into a sales sequence. Sometimes the right result is just a useful answer.

11. How do you report on business results with honest attribution?

Impressions tell you about distribution. They can't tell you whether the right people got interested or whether a post helped close a deal.

Combine the LinkedIn analytics, tracked website visits, inbound inquiry records and relevant CRM data I've supplied for [period]. Report on reach, reactions and comments by post, topic and format; qualified inbound conversations; resource sign-ups and demo requests where tracking exists; opportunities with a documented LinkedIn contact; and time spent producing and reviewing content. Keep directly tracked conversions, self-reported discovery and possible influence in separate groups. Don't attribute a whole deal to LinkedIn because one contact reacted to a post. Mark posts that reached relevant prospects, posts that reached a wide but less relevant audience, and posts with too little data to judge. Close with three findings, three limitations and one recommended experiment. Where you can, compare posts at the same age so new posts aren't judged against older ones.

In Claude: upload the LinkedIn analytics export together with your web and CRM exports, and ask for a spreadsheet with working formulas, so you can check every number yourself. Claude can also chart the data in the chat, or turn the findings into a short deck with Claude Slides if you report to someone. Ask for a separate tab of posts too new to judge. If tracking is missing for part of the period, tell Claude which weeks, so it labels those rows and leaves them unfilled.

Why it matters: better reporting gives you a reason to repeat or change an approach. It also keeps uncertainty in view, so incomplete tracking never turns into a confident revenue claim.

12. How do you turn feedback into a better system?

Every edited draft says something about your standards. Write down the lessons that keep recurring, and you spend less time fixing familiar problems.

Compare the original drafts, my edits, the final published versions and the performance data from [period]. Find voice corrections I keep making; claims that needed more evidence; ideas that got clearer after a change of framing; formatting changes that made posts easier to read; and topics that started useful sales conversations. Propose updates to the voice guide, the research briefs and the editorial checklist. For every proposed rule, show an example, explain what it's for and note when an exception might make sense. Keep style preferences apart from performance hypotheses. Don't make a permanent rule out of one post that did well or badly. Return a short change log for my approval and suggest one controlled experiment for next month.

In Claude: keep the original drafts and the published versions as two files in the Project, so the comparison has something exact to read. When you approve the change log, ask for an updated voice guide file and swap it for the old one in the Project's files; a rule that covers everything you write belongs in the Project instructions. If you built a skill in prompt 1, update it with the same approved changes so the next draft starts from them. Memory also picks up corrections you make in chat, though the Project files are what you can read and edit, so treat those as the source of truth.

Why it matters: this is how the workflow gets easier to run over time. Your corrections become guidance you can reuse, and ideas about performance stay testable and never harden into superstition.

How should you roll out all 12 prompts?

Add the workflow in stages so you can check each output before taking on another recurring job.

• Week 1, the base. Load your business context, run prompt 1 and check the voice guide against writing you'd really publish. In Claude, that means the Project exists and the context file sits in its files.
• Week 2, the research backlog. Run prompts 2 to 4 on a manageable set of customer conversations, market sources and LinkedIn examples. Connect Google Drive first if your call notes are there.
• Week 3, the first batch. Use prompts 5 to 7 to turn approved material into a calendar and a few drafts. Save each output as a Project file so the next prompt can read it.
• Week 4, the review. Run prompt 8, publish the posts you approve, run the first hour on each one, and note the changes you make. Open a new Project conversation for every review.
• Weeks 5 and 6, conversations. Add prompts 9 and 10 where you have enough sales context and inbound activity to make them useful. Add a CRM connector only after setting it to read-only.
• Then every month. Use prompts 11 and 12 to review results, improve the instructions and pick the next experiment. This monthly pass is the first job worth putting on /schedule.

For any recurring job, decide its inputs, where the output goes, how often it runs and who owns it. Test one run before relying on the schedule, and have Claude report missing access or unfinished work.

Running it on a schedule in Claude. Repeating jobs can run on their own: type /schedule in a Cowork conversation, or open "Scheduled" in the left sidebar to create, view and edit them. Scheduled runs happen in the cloud, so they continue while your computer is off. Good candidates are a Monday sweep with prompt 3 and a monthly run of prompt 11 on the exports you dropped into Drive. Keep the permission setting on Manual for anything that touches connectors with write access. Scheduled runs prepare material and never post to LinkedIn, so publishing stays a manual step.

You don't need 12 separate agents on day one. Start with the job that eats the most time and produces an output you can judge reliably.

What can't Claude do for you?

Claude can organise research, find useful evidence, prepare drafts and learn from your edits. The quality still depends on the experience and judgment you bring to it.

That means holding a real opinion, being willing to explain it, and noticing when a polished draft says something you'd never defend in a conversation.

When a post from this workflow goes live, give it a first hour on LinkedInLikes: sign up free, confirm your email and put the 100 starter points toward likes or comments.

More articles

• Building an AI content system for LinkedIn from one empty folder
• LinkedIn maxxing with Claude Code
• LinkedIn maxxing with ChatGPT
• The 2027 LinkedIn marketing playbook

Ready to begin: open a new Claude Project, paste your business context, and run prompt 1. Better inputs make every later step more useful.