Notes on growing
with LinkedIn content
Playbooks, product news, and honest lessons from two founders who treat LinkedIn as a system, not a slot machine.
Run your LinkedIn content pipeline from Claude
Connect the LinkedMash MCP server to Claude, Cursor or any MCP client, and your assistant can finally read the library you built.
Karthikeyan Kumar11 min read
The 30-day saved-posts system: from graveyard to content engine
Label on save, review on Friday, draft two posts from your best finds, schedule for the week ahead. Four steps, one month, no blank pages.
Karthikeyan Kumar9 min read
Your first line is 80% of the post
Hooks are not a talent, they are a small set of shapes. Name the shape and you can borrow it without copying anyone.
Ramya Chinnadurai6 min read
From saved post to Notion database, automatically
Six columns, one sync, and a filter recipe. After that every LinkedIn post you save files itself, media and links intact.
Karthikeyan Kumar8 min read
A labeling system you will actually keep
Label by what you would do with a post, not what it is about. Search already covers topics; it cannot cover judgement.
Ramya Chinnadurai7 min read
Posting times are overrated. Consistency is not.
Posting time is a small effect people optimise because it is easy to change. Publishing more often is the large one they avoid.
Ramya Chinnadurai5 min readLinkedIn saved posts are a dead end by design
Saving is one tap. Retrieval is unbounded scrolling. The two sides of the loop are mismatched, so the pile only ever grows.
Ramya Chinnadurai6 min readOne saved-posts playbook per week
A new playbook, a product note, or one tactic worth stealing from what we have been saving.
One email a week, nothing else. Unsubscribe in one click.