A Batching Workflow for Planning a Month of Social Content

Most content calendars do not die from a shortage of ideas. They die from context switching. Writing one post at 9:40 on a Tuesday means opening the app, remembering what you posted last week, rebuilding your sense of the account's voice, staring at an empty field, and then producing something mediocre because a meeting starts in twenty minutes. You pay that startup cost every single time.

Batching moves the cost to one place. One session, one headspace, four weeks of posts. The workflow below is not clever and it is not new. It is the version that survives contact with a real week, and the mechanics matter far more than the theory.

Start with pillars, not posts

A pillar is a recurring theme your account is actually about. Three to five of them, no more. If you cannot name yours, you do not have a content problem, you have a positioning problem, and no amount of scheduling will fix that.

For a developer tools account, pillars might be: how the product works, the problem space around it, the build log (what shipped and what broke), and the ecosystem you sit in. For a bakery they might be: what is in the case today, technique, the neighborhood, and the people behind the counter. The specifics are yours. The constraint is the count.

There are two failure modes. Too many pillars and the account has no recognizable shape, because every post is a first impression for someone. One pillar and you have a brochure that updates.

How to find yours

Open your last thirty posts and group them. Do not group by format, group by what they were about. Most accounts have four or five natural clusters and one orphan pile. The clusters you enjoyed writing are your pillars. The orphan pile is what you should stop doing.

Turn pillars into buckets

A pillar is a topic. A bucket is a shape. Keeping the two separate is what makes this workflow generative instead of decorative, because a topic plus a shape is a brief, and a brief is something you can actually write.

• The teardown: pick one thing and explain how it works, including the specific detail most people skip.
• The tip: one useful thing that fits in a screen, no preamble.
• The artifact: a screenshot, a chart, a photo, a diff. The image is the post, the text is the caption.
• The opinion: a link plus what you actually think, not a link plus a summary.
• The question: a real one you want answered, not engagement bait dressed as curiosity.
• The log: what shipped, what broke, what you learned. Forgettable on any given day, valuable over months.

Four pillars times six buckets is twenty-four distinct slots. You do not need to fill all of them and you should not try. But a slot like "build log, teardown" is a far easier prompt than a blank composer, and that difference is the entire reason batching works.

Batching does not make you write faster. It stops you paying the startup cost twenty separate times.

Run the batch as three passes

The single biggest mistake is drafting and editing in the same motion. They use different parts of your brain and they fight each other. Split the session.

Pass one: dump (about thirty minutes)

Fill the grid with one-line ideas. No writing, no judging, no checking whether you already posted it. If you have kept a capture file all month, this pass is mostly transcription, which is exactly the point. The capture file is the real secret here: a single note where every half-idea goes the moment it appears, so the dump pass is never cold.

Pass two: draft (sixty to ninety minutes)

Write every idea out in full, in the native format of the network it is for. Do not stop to fix a word. Do not stop to find the link. Leave a marker and keep moving. Some of these will be bad. Bad drafts are cheap, and you are going to throw out a third of them anyway.

Pass three: cut and schedule (about thirty minutes)

Now be ruthless. Delete the third that is filler. Fix the opening lines, which are the only lines most people read. Then place what survives on the calendar. This is the only pass where you touch the scheduler, and it should feel mechanical, because every real decision was already made.

Calendar mechanics: decide the shape once

Pick the cadence you can sustain in a bad week, not a good one. Three posts a week that actually happen beat a plan for daily that collapses in week two and takes your confidence with it.

Then assign buckets to slots, not topics. Monday is a teardown. Wednesday is a tip. Friday is the log. The topic changes every week, the shape does not. This sounds like a small thing and it removes the largest weekly decision, because you are never choosing what kind of post to write, only what to put in it.

Leave roughly a fifth of the calendar empty. A fully packed queue means you cannot react to anything real: an outage, a launch, a good question in the replies, a thing that happened. The gaps are not waste. They are the budget for being timely.

This is the unglamorous part where a scheduler earns its keep. You want to see the queue per channel and spot the holes at a glance. postqueen does that in the UI, and it also ships a CLI, which matters if your drafts already live in files.

# what is connected, and under what names
postqueen integrations:list
 
# drop a finished draft straight from a file into a slot
postqueen posts:create \
  --channel linkedin \
  --at "2026-08-03T09:00:00Z" \
  --text "$(cat drafts/teardown-week-31.md)"

If you write in a text editor, this is the shortest path from file to queue, and it means your drafts keep a home that is not inside somebody else's database.

Recycle evergreen posts on purpose

The most underused lever in content planning is that not every post has to be new. Sort every draft at write time into evergreen or perishable. It costs one second while the idea is fresh and it is nearly impossible to reconstruct three months later.

Evergreen means it will still be true in six months: no dated reference, no timely hook, no numbers that move. A teardown of how OAuth token refresh works is evergreen. A note about your launch week is not.

Recycling is not reposting the same text, which reads exactly like what it is. Rewrite the first line, swap the example for a newer one, or change the bucket entirely: a tip becomes a teardown, a teardown becomes an artifact with a caption. The idea gets reused. The post does not.

Two rules keep it honest. Nothing comes back inside ninety days on the same channel. And keep a short greatest hits file, so that when the dump pass runs dry (it will, some months are simply like that) you have somewhere to go that is not desperation.

What this actually buys you

Not virality. Batching does not make any individual post better. What it buys is a calendar that still has something in it during the week everything goes wrong, which is precisely the week most accounts go quiet and never fully recover.

It also buys editing. A post written on Tuesday for Thursday gets read once with fresh eyes before it goes out, and that single pass catches most of what you would have regretted. Consistency and a second look are boring advantages. They are also most of the game.