A LinkedIn content calendar usually dies in the second week, and it dies the same way every time.
Someone sits down on a Sunday, picks twelve themes, and assigns them to dates. Monday the 4th: thoughts on hiring. Wednesday the 6th: a lesson about pricing. It looks like a plan. Then the 4th arrives, the founder has no particular thought about hiring that day, and the choice is to write something hollow or write nothing. Most people write nothing, and by the 11th the calendar is a document nobody opens.
The failure is not discipline. It is that the calendar was built forwards, from an idea of what you might say, rather than backwards from what was actually going to happen to you.
Start with the week you already have
Your calendar app already contains next week. Not the topics, but the events: three client calls, a board prep session, two candidate interviews, a pricing discussion someone put in on Thursday.
That is a content calendar already. It just is not written as one.
The reframe is small and it changes everything: instead of asking "what should I post on Wednesday", ask "what will I know on Wednesday that I do not know today". The Thursday pricing discussion will produce a decision. The candidate interviews will produce a pattern. The client calls will produce at least one sentence you did not expect. You do not know what those will be yet, and that is fine, because you are not scheduling the content. You are scheduling the slot and letting the week fill it.
The one thing to decide in advance
Not the topics. The mix.
A post does one of three jobs, and the profiles that read as salesy are the ones where the balance drifted. Sonar's defaults are 60 percent awareness, 30 percent consideration, 10 percent conversion, with roughly a tenth of everything being off-topic, the thing you actually care about outside work.
Those numbers are advisory and they are not magic. What matters is that you set one, because the drift is invisible without it. Nobody decides to post four sales posts in a fortnight. It happens because each one individually felt justified.
So the plan for the month is three numbers and a cadence, not twelve headlines.
What a week actually looks like
Two sittings, about forty minutes total.
Monday, half an hour. Read what last week gave you. Pick the two or three moments that carry a post, write them, and schedule them across the week at your own morning. Not at the "best time to post" that some study computed across a million posts in a timezone you do not live in. Your audience's morning.
Friday, ten minutes. Reply to every comment you have not answered, and write down the two things from this week that could be next week's posts. That second half is the whole trick: Friday-you knows what happened, and Monday-you will not.
The order matters. Capture on Friday, write on Monday. Reverse it and you are back to inventing themes.
Where the calendar comes from when the week is thin
Some weeks genuinely produce nothing. A week of internal admin and rescheduled calls is not a week with three posts in it.
Two honest answers. Post less that week, which is allowed and is better than filling the slot with a generic opinion. Or reach for material that is not this week's: something from your own history, a development in your field you have a real view on, an argument in your niche you keep having. Those are not filler, they are a different supply, and a calendar that only draws on the last seven days will break the first time a quiet week arrives.
What software changes, and what it does not
Sonar builds this calendar from the week rather than from a theme list. It reads the sources you connect read-only, keeps the moments worth saying, and every week deals five drafts: some from your own moments, one from a story you told it about your career, one from a development in your niche, the rest from a pillar you chose. You keep what is true and throw the rest away.
What it does not change: the two decisions that were always yours. What happened this week that is worth saying, and whether the draft in front of you is true. Nothing publishes until you schedule it, so an unapproved draft sits on the board indefinitely, which is the point rather than a limitation.
The full version, with pillars and the funnel mix, is in a LinkedIn content strategy for founders.