Posting is an operations problem.

The plan is two posts a week, and it is a good plan. It holds for three weeks. Then a large job lands, and the gap is nine days, and then it is a month, and the account acquires that particular shape: a burst of activity, a silence, a burst, a longer silence. Anyone scrolling it can read the history of the company's busy season in the timestamps.

The diagnosis offered is almost always one of three: we got too busy, we need more discipline, we need better ideas. All three describe a failure of character. None of them is what happened. What happened is that a recurring output was expected to appear without anybody designing the process that would produce it, and undesigned processes fail in predictable, boring, entirely mechanical ways.

Look at it as a line, not a habit

Every business runs recurring outputs. Invoices go out. Stock gets reordered. Equipment gets serviced. Payroll runs. Nobody expects payroll to happen because somebody felt inspired on the right Friday. It happens because the inputs arrive on a schedule, the steps are the same every time, and the whole thing is arranged so that no single person's week can stop it.

Every other recurring output in the business has a process. This one is expected to happen because somebody feels like it. That is the entire defect, and it is upstream of every symptom the owner is looking at.

Where the line actually breaks

The input has no collection step. The raw material of anything worth posting is a moment on the job: the state of it before, the thing found behind the wall, the finished work in good light. That material exists for about an hour, and it exists at the job, which is not where publishing happens. If nothing collects it there, then at publishing time the only material left is invention, and invention is slow, expensive, and worse.

Production and publication are welded together. When the post gets made on the day it goes out, every single publication depends on that day having room in it. One crowded day is one missed post. Any process with no buffer between making and delivering stops the instant demand spikes, and demand spikes are exactly when the business has the most worth showing.

There is no standard unit. If every post is an open creative question, the cost of producing one is unbounded, and unbounded cost is what gets postponed when a day gets tight. A line runs on repeatable units. Without them, the work restarts from nothing every time and is priced accordingly.

What designing it looks like

Separate capture from composition. Capture is a two minute step that happens where the work happens, by whoever is standing there, and it produces raw material rather than finished posts. Composition is an entirely different task, done at a different time, in a different place, by whoever composes. Fused together, they fail at the point of the busiest person's worst day.

Publish from inventory, not from today. Working ahead is not about efficiency. It is about decoupling. When there is a buffer of finished work, the schedule survives a bad week because the output was made during a good one. A process that delivers from stock does not care what today looks like, which is the only property that matters over a year.

Batch the like work. Most of the cost of producing one post is setup: getting into it, finding the material, remembering the conventions. Making six in one sitting does not cost six times one, it costs closer to twice. Batching is the standard answer to any operation with high setup and low run time, and this is one.

Fix the shapes. A short list of standing formats converts an open question into a fill-in. What should we make becomes which of these five is this, and a fill-in can be completed on an ordinary Tuesday by somebody who is not feeling remotely creative. That is the point. The process has to run at the average level of enthusiasm, not the peak.

None of this makes the work exciting. It makes it survivable, which is the actual requirement, because the only cadence worth designing is one that will still be running when nobody involved is excited about it any more. Which raises the next problem, and it is not audience fatigue. It is that the people producing get tired of the message long before anyone outside the building has noticed it.