Why identical cross-posting reads badly
It is not that audiences resent seeing the same idea twice — most people follow you in one place. It is that a message written to satisfy every network at once satisfies none of them. Written to X’s limit, it is too thin for LinkedIn. Written for LinkedIn, it gets truncated on X. Written for Instagram, it carries a "link in bio" that means nothing anywhere else. The generic version is not a compromise between five good posts; it is a fifth good post in five wrong places.
Separate the idea from its expression
The efficient unit of work is the idea, not the post. One idea — a product update, an observation, a customer story — becomes five expressions of different lengths and registers. Thinking of it this way removes the guilt about "repeating yourself", because you are not: you are publishing one thing in the dialect each place speaks.
Start long and cut, never start short and pad
Write the full version first, at whatever length the idea naturally takes. Cutting to X’s limit is a mechanical exercise that usually improves the sentence. Padding a 40-word thought out to a 200-word LinkedIn post is a creative exercise that reliably produces filler, and everyone can tell. This one habit removes most of the friction from multi-platform writing.
What each platform actually wants
The adaptations are more predictable than they look, and after a few weeks they stop requiring thought.
- X — one sentence with the sharpest version of the point; cut every qualifier
- LinkedIn — a hook line that survives the "see more" fold, then the reasoning
- Instagram — the image carries the message; the caption adds context and tags
- TikTok — the hook is spoken in the first two seconds; the caption is a label
- YouTube — the title does the discovery work; the description carries the context and links
Decide which platforms deserve a bespoke version
Not all five need individual attention every time. For most accounts two networks are primary and the rest are secondary, and the honest allocation is to write the primary ones properly and let the others inherit something reasonable. This is a strategic decision worth making explicitly, because made implicitly it becomes "whichever one I got to before I ran out of energy".
Watch the fold, not just the limit
Character limits are the obvious constraint and the less important one. The constraint that decides whether a post works is where the feed truncates it — the point at which a reader chooses to expand or scroll past. On LinkedIn that is a couple of lines; on X, everything past the first sentence is a gamble. Writing to the fold rather than to the limit is the difference between a post that is read and one that is merely published.
Check the preview before you commit
Most cross-posting errors are visual rather than textual: an image centre-cropped into nonsense, a link card that pulled the wrong thumbnail, a line break that collapsed. These are invisible in a plain text box and obvious in a preview. If your tool renders each network’s preview, use it — the whole cost is a few seconds, and the alternative is finding out from the live feed.
Where AI genuinely helps, and where it does not
Adapting an existing, well-written post to a different length or register is exactly the kind of task a language model does well — it is transformation, with the substance already supplied by you. Generating the idea from nothing is where it produces the recognisably hollow output everyone complains about. Used for the adaptation pass rather than the thinking pass, it takes the tedium out of multi-platform posting without flattening the voice that made the post worth writing.