LinkedIn Content Strategy
A founder's LinkedIn works when it sounds like the founder, talks about what buyers are already discussing, and feeds the people who engage into the sales system. Our content strategy is built around those three things.
What it is and when to use it
A LinkedIn content strategy decides who the posts are for, what they are about, how they are shaped and what happens to the people who respond. For B2B founders the goal is conversations with buyers, not reach for its own sake.
We run it as an engine: a positioning file the model reads first, reference posts measured rather than copied, gates that stop repetitive or off-topic posts, and a pipeline that turns commenters into qualified leads.
Use this playbook when
- A founder's profile is unused, or they post when they feel like it and nothing comes back.
- The goal is qualified conversations or booked demos, not followers.
- Posts are being written, but they all look the same or drift off the audience.
Tools we use
Only tools we have run this playbook with, and what each one does in it.
- Claude Code: Ingests reference posts, measures them and drafts within the measured band
- Apify: Scrapes recent posts from reference profiles for measurement
- Figma: One visual per post, from a brief of structure and exact labels
- Notion: Content calendar and lead-magnet pages
- ClearCue: Captures commenters and engagers on the posts as signals
- Supabase: Stores captured people so they can be qualified before any outreach
The steps
The order we work in. Each step feeds the next, and every result is written to the database as it happens.
Lock the positioning
Pillars with weights, the voice, the audience and the rules the posts must follow live in one file, and an append-only decision log records every change the founder asks for.
Measure reference posts
Recent posts from chosen reference profiles are split, measured and analyzed for hooks, closers, length and structure. Image analysis covers content and layout only.
Set the length band for this audience
Length is calibrated to the audience, not to whichever reference profile writes the longest posts. Reference medians in one study ranged from 140 to 2,689 characters, so the choice matters.
Gate the topic
A topic must be in active discussion in the last 30 days, with the source noted. A small audience gets the current vocabulary first and depth later.
Draft from measured patterns
Every structural choice in a draft points to a real reference post. Every tool named is checked against what it actually does.
Audit the batch
Each batch gets a variety table across structure, angle and visual. No more than 40% of posts may be numbered lists, and every post is counted against the length band.
Brief the visual
The design brief gives structure and exact labels, never color direction.
Engage on the feed
Comments on other posts are one to three sentences, pick one specific line and end with a real question. No invented experiences.
Capture and qualify responders
Commenters and engagers are captured as signals, ranked commenters above likers, and filtered by company fit before anyone is contacted.
What good output looks like
Measured on our own profiles:
Our co-founder's last 47 posts drew 11,831 reactions and 2,513 comments.
The 5 lead-magnet posts in that set averaged 150.2 comments, against 41.95 for the other 42.
Over 90 days, the content-led system on our own names produced 159 connections from 425 requests and 76 replies from 208 messages.
A first batch of 5 posts all landed between 1,276 and 1,492 characters, inside the band set for that founder's audience.
Common mistakes, and the fix
Each of these happened on a real run. The fix is now part of how we work.
Every post in a batch was a numbered list.
- Cause
- Falling back to one skeleton.
- Fix
- Cap lists at 40% of a batch and make each post differ on at least two of structure, angle and visual.
Post after post closed with the same "DM me" line.
- Cause
- One default call to action.
- Fix
- Rotate call-to-action shapes and cap the soft DM close at one post in ten.
Posts were too long for the audience.
- Cause
- Length copied from a reference written for a more technical audience.
- Fix
- Calibrate to the audience, and cut anything more than 20% over the band.
Bullets wrapped to two or three lines on phones.
- Cause
- Drafts were previewed on desktop.
- Fix
- Keep each bullet to about 35 characters.
Copy read as machine-written.
- Cause
- Dashes and emoji habits.
- Fix
- Ban em-dashes and emoji, and use plain bullets.
Correct, technical posts underperformed.
- Cause
- Metric deep-dives for a small audience.
- Fix
- Lead with the topic people are already discussing; keep depth for later posts.
Posts spoke to the wrong mental state.
- Cause
- They said "your setup is broken" to buyers busy rolling out a platform.
- Fix
- Write about the problem the buyer is already solving.
A post promised a giveaway that did not exist.
- Cause
- No check that the asset was real.
- Fix
- Mark in every draft whether the promised asset exists.
Engagers were treated as the target list.
- Cause
- Engagement crowds are mostly not buyers.
- Fix
- Filter by company fit before any outreach.
Questions people ask about linkedin content strategy
What is the best content strategy for LinkedIn?
For B2B founders, ours is: lock the positioning and voice first, measure what good posts in your space look like instead of copying them, write about topics buyers are already discussing, vary the structure across every batch, and capture the people who comment so they can be qualified. On our co-founder's profile, lead-magnet posts averaged 150.2 comments against 41.95 for the others.
Want this playbook running on your market?
Tell us what you sell and who you sell to. We will show you how this playbook would run for you, and what we would check first.
