The hardest part of posting consistently was never the posting. It was the blank box on a Tuesday morning when you have nothing to say, a client waiting, and a nagging feeling that you are falling behind everyone who apparently posts three times a day.

So you skip Tuesday. Then Wednesday, because now you feel behind. Then you vanish for two weeks and come back with a post that starts "hey, it has been a minute!" (We have all written that post.)

Here is what actually fixed it for me. I stopped trying to write posts daily and started writing them monthly, in one sitting, with a stack of prompts I run in the same order every time. Roughly two hours, once a month, and the content problem stops being a daily problem.

These are the twelve prompts I run. They are grouped in three passes: figure out what you are even talking about, turn that into real posts, then make it sound like a human wrote it. Copy them, swap in your own details, and delete the ones that do not fit your business.

Quick tip before you start

None of this works well if Claude does not know you. Before you paste prompt one, spend three minutes telling it what you sell, who buys it, what you charge, what you sound like, and what you absolutely will not say. Rambly is fine. Messy is fine.

Better yet, save that in a Claude Project so it is waiting for you every month. Then you are not re-explaining your business every time, and the drafts come back closer to usable on the first pass.

Pass one: figure out what you are actually talking about

Most people skip this part and go straight to "write me 30 Instagram captions." That is why the output feels like it could belong to anybody. The first four prompts exist so the next eight have something real to work with.

1. The brain dump sort

Here is everything I know how to do, everything clients keep asking me, and everything I have opinions about: [dump it all, unsorted, however it comes out of your head]. Sort this into four content pillars I could post about for a year without repeating myself. For each pillar, tell me who it speaks to and what they are worried about.

Do the dump badly on purpose. Fragments, half sentences, that thing a client said last week. You are not writing, you are emptying a drawer.

2. The question mining prompt

Based on those pillars, list 40 real questions my ideal client would type into Google or ask a friend at dinner. Use their words, not industry words. Sort them from most beginner to most advanced.

Questions are the single best content source you have, because every question is already a post title. Forty gives you more than a month with room to throw some away.

3. The story inventory

Ask me 10 questions designed to pull real stories out of me: mistakes I made, clients I turned around, things I believed five years ago that I no longer believe. Ask them one at a time and wait for my answer before the next one.

This is the prompt that makes your month not sound like everyone else's month. Advice is everywhere. Your Tuesday in 2019 is not.

4. The angle spread

Take my top pillar and give me 12 genuinely different angles on it: a myth I want to bust, a mistake I see constantly, a before and after, a hot take, a beginner walkthrough, a behind the scenes, and so on. Do not just reword the same point 12 times. If two angles overlap, replace one.

That last instruction matters more than it looks. Without it you get twelve versions of the same sentence with the adjectives shuffled.

Pass two: turn the ideas into actual posts

Now you have raw material. These four prompts turn it into things you could genuinely paste into a phone.

5. The one idea, five formats pass

Here is one idea from my list: [paste it]. Turn it into five pieces: a 30-second video script, an Instagram caption, a LinkedIn post, a short email to my list, and three story prompts. Same idea, real differences in tone and length for each platform. The LinkedIn one can be longer and calmer. The video script should sound like talking, not writing.

Run this on four or five ideas and you have most of a month. One idea genuinely does stretch that far when the formats are actually different.

6. The hook stack

Here is my post: [paste it]. Write 10 different opening lines for it. Two curiosity, two contrarian, two story-based, two direct promise, two question. No clickbait I would be embarrassed by. Then tell me which one you would pick and why.

The first line decides whether the other forty get read. Ten options takes thirty seconds and the third or fourth is usually better than what you would have written cold.

7. The carousel breakdown

Turn this post into a 7-slide carousel. Slide 1 is the hook, slides 2 through 6 are one idea each with a maximum of 20 words, slide 7 is the takeaway plus a soft call to action. Keep it readable on a phone at arm's length.

Carousels are where a good idea gets a second life. Same content, different shape, and people who scrolled past the caption will stop for slide one.

8. The comment bait pass

For each of these posts, write a closing line that gives people something specific and easy to answer in the comments. Not "thoughts?" Something they can answer in four words while standing in line at the store.

Nobody replies to "let me know what you think." Everybody replies to "which one of these are you guilty of, one or three?"

Pass three: make it sound like you, then get it scheduled

This is the part people skip, and it is the difference between a month of content and a month of content you are willing to put your name on.

9. The voice check

Here are three things I have actually written: [paste real emails, texts, old captions, anything in your voice]. Now rewrite these drafts to match how I actually talk. Keep my sentence rhythm and my weird phrases. Cut anything I would never say out loud.

Real writing samples beat any amount of describing your voice with adjectives. Even your text messages work.

10. The AI tell scrub

Go through every one of these posts and delete the AI giveaways: "in today's fast-paced world," "let's dive in," "game-changer," "unlock," rule-of-three lists that exist for rhythm rather than meaning, and any sentence that could be about any business. Flag anything vague and ask me for the real detail instead of inventing one.

That final clause is the important one. Left alone, AI will happily fill a gap with a made-up client result. Told to flag instead, it asks you a question you can actually answer.

11. The calendar layout

Take these finished posts and lay them out across four weeks. Mix the formats so I am never posting two carousels in a row. Put the strongest three where they will do the most good. Give it to me as a simple table: date, platform, format, hook, and the first line of the post.

A pile of drafts is not a plan. A table with dates on it is a plan, and it is what stops you from posting your four best ideas in the same week.

12. The next month starter

Based on what I wrote this month, what did I not say that I should have? What questions did these posts raise that I never answered? Give me 15 starting ideas for next month, and tell me which two I should build something bigger around.

Run this before you close the chat. Next month's session starts from a list instead of from nothing, which is what makes the second month take half as long as the first.

How to actually run this in one sitting

Block two hours. Not four, and not fifteen minutes. Two.

Prompts one through four take about thirty minutes and feel like talking, not working. Five through eight take about an hour and produce the bulk of the drafts. Nine through twelve take thirty minutes and are mostly you approving, cutting, and occasionally saying "no, here is what actually happened."

One rule that makes the whole thing hold together: keep it all in one chat, or one Claude Project. The prompts build on each other. If you open a fresh window at prompt nine, you have thrown away everything it learned about you in the first eight.

And do not aim for perfect. Aim for scheduled. A good post that goes out on Tuesday beats a great post still sitting in a draft folder in November.

Frequently asked questions

Do I need a paid Claude account for this?

No. All twelve prompts are normal conversation, so a free account will run them. You will feel the ceiling in a two-hour session with a lot of pasting, and that is the honest moment to consider upgrading. Start free, upgrade when the limits actually get in your way rather than because someone told you to.

Will a month of posts written in one sitting all sound the same?

They will if you skip pass one. That is exactly what the pillar and angle prompts are for: you are pulling from four different buckets and twelve different angles instead of one mood on one afternoon. Prompt nine catches whatever sameness slips through.

How is this different from just asking for 30 captions?

Asking for 30 captions gives you 30 captions about nothing in particular, because the model has nothing to work from except the topic. This gives it your pillars, your clients' real questions, and your actual stories first. Same tool, wildly different output, and the difference is entirely in the setup.

What if I do not have any good stories to tell?

You do, you just do not think of them as stories because you lived them. That is why prompt three has Claude interview you one question at a time instead of asking you to "share a story." The thing you consider boring and obvious is usually the thing your audience has never heard anyone explain.

Should I post the drafts exactly as written?

Read every one before it goes out, and change at least something in most of them. Not because the drafts are bad, but because the small edit you make is usually the specific detail that makes the post yours. Treat it as a strong first draft from a teammate who does not know how your Thursday went.

What if I miss a month?

Then you run it whenever you next get two hours, and prompt twelve means you are starting from a list rather than a blank page. This is not a streak and there is no penalty. Systems that survive real life are the only ones worth building.

Pick one pass and start there

If two hours feels like too much right now, do not do all twelve. Run prompts one through four this week and stop there. Having pillars and forty real client questions sitting in a document will make your next posting week noticeably easier on its own.

Then next month, add pass two. The whole thing gets faster every time because the setup work is already done.

If you would rather build your month next to people doing the same thing, that is what happens every day inside Club Jam on Skool. It is $47 a month with a 7-day free trial, 220+ on-demand modules, and a room full of everyday business owners who are not developers either. We share the prompts that are working, fix the ones that are not, and nobody gets told to go read the documentation.

Come fill your calendar with a little backup. 🧡