Do not index
Why is your LinkedIn reach dropping even though you are posting more than ever? Because the feed stopped counting how many people saw your post and started counting whether anyone cared enough to finish it. That is the shift, and most people are still optimizing for the old number. If your distribution is collapsing, it is almost never a format problem or a timing problem. It is a depth problem wearing a format problem's clothes.
LinkedIn's newer LLM-based feed does not detect AI, and it does not punish you for using it. What it measures is dwell time, saves, and real discussion. Generic content gets near-zero dwell time, no saves, and no conversation, so the system simply stops distributing it. It looks like a penalty. It is not. It is the absence of a signal. Nobody finished reading, so the algorithm concluded there was nothing worth showing to the next person. As ZoomSphere's breakdown of the 2026 feed put it, the advice that actually moves the needle is blunt: "Find one thing the client specifically knows from direct professional experience that is not available anywhere else in exactly this form. Build the post around that."
This is for agency owners running client content at scale and for ghostwriters charging $5k to $30k per month who live or die by whether their clients' posts perform. It is for founders of $200k to $2M agencies who have a team of three to ten people producing posts daily and watching the numbers slide anyway. If reach is your product, this is the mechanic underneath your product.
This is not for someone posting once a month about a job change. Skip this if you do not depend on distribution for revenue. If you are still chasing the perfect posting time or the ideal hook formula while ignoring whether the post says anything, this article will not fix your reach, because the timing was never the variable.
What the algorithm is actually reading
Think about what finishing a post requires. A reader gives you their attention only if the post is giving something back fast enough to justify the next line. Generic content does not. It front-loads a clever hook, then pays it off with advice the reader has seen two hundred times. They bounce at line three. Multiply that across an audience and your dwell time average tells the system this content is filler. No save, because nobody references filler later. No comment, because there is nothing specific enough to react to.
I run this at an agency, and the most valuable deliverable in our workflow is no longer the post. It is the brief. A weak brief, the kind that says "write something about leadership for the CEO," produces a post that any model would produce, because the input was generic. A strong brief contains one thing the client knows from direct experience that exists nowhere else in that exact form. Same writer, same tools, same account. Completely different raw material, and completely different result on the feed.
The Brief Spine
So the method we use is what I call the Brief Spine. Every post is built around a single load-bearing piece of first-hand knowledge, and everything else in the post hangs off that spine. No spine, no post. The spine is not a hook and it is not an opinion anyone could hold. It is a specific number the client watched happen, a decision they made under pressure, a result they can name, a mistake that cost them something real. You build the post around that one thing, and the post earns its dwell time because the reader cannot get that information from the other forty posts in their feed.
When a post underperforms, we do not adjust the hook first. We check the spine. Nine times out of ten the spine was missing, and the writer compensated with style. Style does not survive contact with the new feed. This is the same discipline behind a real quality-control system, the kind that catches a weak post before it ships rather than after a client churns, and it is worth reading how that quality-control system prevents client churn because the failure mode is identical. Thin input, polished output, dead reach.
The strategic implication is that the locus of value in content moved upstream. It used to sit with the writer who could turn a phrase. Now it sits with whoever can extract the one true thing worth building around. If your operation is organized so that writing is the expensive step and sourcing the insight is an afterthought, you have it backward for the feed you are actually posting into. The teams that reorganize around extracting depth, even at the cost of producing fewer posts, will hold their distribution while everyone optimizing for volume watches it quietly disappear. Reach is no longer something you win by showing up more. It is something you earn by being worth finishing.
