LinkedIn rewards founders
who show up like people.

Every move on LinkedIn advertises one page: posts, comments, and invites all funnel curiosity to your profile, and most founders format it for a recruiter who is never coming. The rebuild: a banner that states outcome, audience, and mechanism in under ten words; a headline that says who you help and what changes for them (it travels under every comment you write, so "CEO at Company" wastes the most visible line you own); a custom button pointed at one action, not your homepage; a featured section stacked with proof; and an about section written like a short sales letter, first person, reader named in line one.

Ten minutes of rebuild changes the yield on every hour that follows, because everything else in this playbook multiplies through the profile.

Readers arrive in three states: they don't know they have the problem, they know the problem but not the solutions, or they know solutions exist and need a reason to trust yours. A working week covers all three deliberately: belief-shift posts (contrarian takes, diagnostics, mistake callouts) cast the widest net, methodology posts show the problem is solvable, and proof posts (case studies, numbers, system reveals) close the smallest, warmest slice. Every post gets one named target state; "who is this for" kills the mediocre post before you write it.

Ideas come from collection, not inspiration: your buyers' recurring phrasings, your own operations and mistakes, what you shipped. The first line is the whole game (everything below it hides behind "...more"), and consistency beats brilliance.

People tell you they're shopping without meaning to: re-evaluating tooling, complaining about the tool they pay for, asking their network for recommendations. A comment section is a self-selected list, and not just the author: everyone who commented on a post about your problem just told you the topic is live for them.

LinkedIn has no real keyword search, so intent surfaces around accounts: the company pages, creators, and operators your buyers follow. Watch those, reply usefully in public first (your headline rides along), work the commenters, then connect warm: "enjoyed the thread on X" is the request that gets accepted and the DM that gets read.

"Comment KEYWORD and I'll send it over" is simultaneously the most abused format on LinkedIn and, run honestly, one of the few plays that turns one post into a self-qualified prospect list: commenters identify themselves in public, the asset's topic qualifies them, and the comment volume funds its own distribution.

The fork between operator and engagement-farmer is the asset. A two-page PDF of bullet points converts curiosity into disappointment at scale. The bar: a document that feels like something you could have charged for, complete enough that a stranger could act on it without ever talking to you. Deliver fast and complete, reply like a person while the post runs, and don't run it weekly; the format's power is proportional to its scarcity on your feed.

Every founder who posts for a year knows the graph: the post that took off in March, the near-identical one that sank in July. The algorithm owns your reach and can reprice it without notice. The compounding founders run one quiet conversion underneath everything: each unit of rented attention gets a clean chance to become an owned relationship, meaning an email address, given voluntarily, attached to a person who knows why they gave it.

The capture points already exist in this playbook: route giveaway delivery through a simple name-and-email page instead of a raw file, point the profile button at your best asset, and front-load a short welcome sequence while attention is peak. Gate the delivery, never the promise: asking for a phone number after promising a free doc burns the room you just assembled.

You can only message first-degree connections, which is doctrine, not limitation: by the time you can DM someone, a real interaction created the permission, and the DM's job is to continue that thread, never open a colder one. The arc that works is small messages across a couple of weeks: deliver or reference the shared context, check in lightly days later, drop an unprompted insight specific to them, bridge to a concrete result story, then make one small, easy-to-decline ask.

Days between touches, and two consecutive silences means stop. A pitch is a withdrawal against accumulated trust; the value-first touches are the deposits that make the eventual ask affordable.

Every invite is a bet on acceptance, and losing too many bets closes the table: ignored and rejected invites accumulate against your account until friction arrives. That flips the question from "how many can I send" to "how many will be accepted," which is a targeting question. The note rule: reference real shared context or say nothing; a manufactured note ("impressed by your journey") is worse than a bare invite.

Pace steadily, ramp with account maturity, withdraw stale pending invites, and let inbound carry most of the load: the posture this playbook builds generates incoming requests, which cost nothing and tell the platform your account attracts rather than chases.

LinkedIn polices automation harder than any major platform for a structural reason: paid access to professional attention is the business, so unpaid automated access is competition. What gets accounts caught is rarely content: it's where the session runs from, the absence of human incidental behavior, inhuman rhythm, and recipient-side votes like "I don't know this person."

The division of labor that survives: automate the watching, the scoring, and the drafting from real context; keep the judgment and the human-shaped sending. A restriction ladder looks survivable until the rung that isn't, and the account it takes is the one GTM asset you cannot re-spin in an afternoon.

Worked example · This split is Thread Otter's architecture: watching, intent scoring, and voice-true drafting run on autopilot; sends stay human-shaped and approval-gated until you decide otherwise.