Multi-account content fleet

Use this skill when two or more accounts inside one company publish on a shared content calendar: a founder plus two AEs, a leadership team, an employee advocacy roster, a set of operator voices. Produces the lane map, per-account voice sheets, cadence plan, and engagement rules the fleet runs on. Covers running multiple LinkedIn accounts without them sounding the same, splitting topics across people, staggering posting windows, avoiding reciprocal engagement patterns, and diagnosing a fleet-wide decline where every account drops at once.

SKILL.md
name:
multi-account-content-fleet
description:
Use this skill when two or more accounts inside one company publish on a shared content calendar: a founder plus two AEs, a leadership team, an employee advocacy roster, a set of operator voices. Produces the lane map, per-account voice sheets, cadence plan, and engagement rules the fleet runs on. Covers running multiple LinkedIn accounts without them sounding the same, splitting topics across people, staggering posting windows, avoiding reciprocal engagement patterns, and diagnosing a fleet-wide decline where every account drops at once.

Multi-account content fleet

Multi-account content fleet

Applies when two or more accounts in one company publish on a shared calendar. Produces the lane map, per-account voice sheets, cadence plan, and engagement rules the fleet runs on.

A fleet exists to build differentiated surface area. Every decision below either protects that asset or spends it.

Assign lanes before writing anything

Each account owns a topic lane and a point of view tied to that person's real role and lived work. The lane map decides who covers what, where lanes may touch, and which overlaps are scheduled on purpose, such as a launch week.

The mediocre version drafts a batch of posts and distributes them round-robin with light rewording. The shared audience then sees the same idea from five faces in one week, and every account pays for it.

Lane assignment happens once, up front, and gets revisited quarterly. Building the map, and the rules for when lanes may legitimately touch, is in references/lane-map.md.

Give every account a fingerprint

One page per person: rhythm, lexicon, hard rules, and a bank of their real anecdotes.

Batch production is fine. Batch sameness is what kills fleets, and it enters through shared hooks, shared structures, and shared examples, never through the calendar itself.

The anecdote bank is the part that gets skipped and the part that does the most work. A person's actual stories cannot be generated, which makes them the one component of a voice that survives scale. Collect them in an interview, not from a questionnaire. The full voice sheet spec is in references/voice-sheet.md.

Stagger the surface

Different posting windows per account. Different format mix on any given day, so the fleet never ships five text posts at 9am.

Shared proof points get a different framing per account, angled through that person's lane. A customer result belongs to whoever's lane it lands in, told from that person's vantage. One calendar, N voices.

Window assignment and format rotation are in references/cadence-and-format-mix.md.

Choreograph engagement

Fleet accounts stay off each other's posts in the first hours and never form a visible closed circle of mutual likes.

Each account engages outward in its own lane, where its audience actually is. Reciprocal fleet engagement reads as coordinated from a mile away, to people and to the feed. It also wastes the engagement, since it lands on an audience that already overlaps.

Outward engagement in-lane is the highest-return activity a fleet account does, and it is the first thing cut when the calendar gets tight. Protect it.

Watch correlation, not averages

Track a per-account impressions baseline. Independent movement is health.

All accounts rising and falling together means the fleet shares a signature, some structure, template, or timing pattern that the feed has priced in. Averages hide this completely, which is why a fleet dashboard that reports fleet-wide totals will never surface the problem.

When correlated decline hits, stop diagnosing individual posts and hunt the shared element: same hook shapes, same formatting, same production batch, same posting window. Break the shared element first, then re-baseline. The diagnostic procedure is in references/correlation-diagnostics.md.

What good looks like

The best operator checks the lane map and the correlation chart before judging any single post. The mediocre version optimizes posts one at a time while the fleet-wide signature does the damage.

The common mistake is efficiency worship. Every optimization that increases sameness across accounts spends the exact asset a fleet exists to build, which is differentiated surface area.

The output is good when each account grows on its own curve, inbound treats each person as an individual authority, and the accounts read as colleagues who think differently rather than one template wearing several names.

Rules

  • MUST assign every account a distinct lane and voice sheet before production starts.
  • MUST stagger timing and vary format mix across the fleet on any given day.
  • MUST build each voice sheet on that person's real role, work, and anecdotes.
  • NEVER publish the same idea across accounts in the same window unless the overlap is a planned campaign.
  • NEVER allow a reciprocal engagement pattern between fleet accounts.
  • NEVER judge fleet health on averages. Correlation is the signal.