Most content systems are designed for the week they were invented — a calm one, with energy to spare. Then a launch happens, or a sick kid, or a board meeting, and the system dies, because it was designed for capacity you only sometimes have. The fix isn't discipline. It's designing for the worst week from the start: a system small enough that skipping it is harder than doing it.
The design constraints
Three rules shape everything below. First, one session, not daily effort — a single 30-minute block is defensible on a calendar in a way that "write every morning" never is. Second, capture is continuous, creation is scheduled — the expensive part of writing is retrieving ideas on demand, so the system never asks you to think of something and write it in the same sitting. Third, every step has a degraded mode — a version so small it survives any week, because the real enemy isn't a missed week, it's the broken streak that turns into a dead account.
During the week: capture only
Keep one running note — phone notes app, anything frictionless. When something happens that makes you think, even faintly, "that's interesting," write one line. A client question you answered twice. A mistake you caught. A strong opinion you voiced in a meeting. The bar is deliberately low: these are not drafts, not even ideas really — they're bookmarks to moments when you had something to say, and the moment is the hard part to recover later. Five to ten lines a week accumulate without any session at all, because capture costs seconds and happens inside your actual work.
The 30-minute session
Pick a fixed slot — same day, same time, calendar-blocked, treated with the seriousness of a client meeting. Then three steps, timed:
- Minutes 0–5: pick. Read your capture note and choose two or three lines with the most energy. Not the most impressive — the ones you can feel yourself explaining to someone. Delete lines that have gone cold. Choosing is the step people underestimate; done separately from writing, it's fast.
- Minutes 5–25: draft. Write two or three short posts, ugly-first. Talk each one onto the page as if explaining it to a specific colleague, then shape it: sharpen the first line, cut the wind-up paragraph, end where the point ends. A publishable LinkedIn post is 150–250 words; twenty minutes covers two or three of them once the ideas are pre-selected. If you use an AI drafting tool trained on your voice, this step compresses further — you're editing drafts toward true instead of typing from zero.
- Minutes 25–30: schedule. Queue the posts across the week and close the tab. Scheduling in the same session matters more than it seems: it converts drafts into commitments and removes the daily should-I-post-today decision, which is where most cadences actually die.
Degraded mode: the bad-week protocol
Some weeks the 30 minutes won't happen. Decide now, in advance, what the minimum is, so the decision never has to be made while exhausted:
- Ten minutes: one post, straight from the strongest capture line, published without ceremony.
- Five minutes: repost or re-angle something you published months ago. Your audience didn't see it the first time; feeds guarantee that.
- Two minutes: one genuinely useful comment on someone else's post. Presence maintained, streak psychologically intact.
The degraded modes aren't failure — they're the system working as designed. A cadence that bends survives; a cadence that can only be perfect breaks the first bad Tuesday and takes a month of restart-guilt with it. Build for the week you'll actually have, and the good weeks take care of themselves.