A founder sent us a screenshot last month with the caption "best post of the year." Twenty-two thousand impressions. He was right that it was his biggest number. He was wrong about what it measured. Two lines below the impressions figure, in the same analytics panel, LinkedIn had written members reached: 6,400. The post had been shown roughly three and a half times to each person who saw it.
That is not a bad post. It's a post that found a room and then kept being served to the room. And the reason nobody reads the second number is that LinkedIn impressions vs members reached is a comparison the interface never asks you to make. It shows you two numbers and lets you pick the bigger one.
What the two numbers actually count
LinkedIn's own post analytics documentation separates the two. Impressions count every time the post renders on a screen. Members reached counts distinct people, each counted once no matter how many times the post rendered for them. On personal profiles it's labelled members reached; on Company Pages and in the API the same figure is called unique impressions.
So impressions divided by members reached is a frequency. It's the average number of times each person who saw the post saw it. A post at 3,000 impressions and 2,700 members reached was shown to almost everyone once. A post at 3,000 impressions and 900 members reached was shown to a third as many people, three times each.
Same impressions figure. Completely different event.
Why ecommerce founders get the high ratio by construction
A narrow lane produces a small room. That's the deal, and it's the right deal, because the small room contains the buyers. But a small room has a consequence in the feed: when the platform finds a post that performs with a defined audience and that audience is a few thousand people, it re-serves the post to the people who haven't engaged yet. It comes back the next morning. It comes back in the "suggested" slot four days later. Each of those is an impression and none of them is a new member.
A generalist leadership account with a broad topic profile gets a low ratio because there's always a new pool to test into. An Amazon founder writing about return rates gets a high ratio because the pool is finite and the platform is thorough.
That's not a problem to fix. It's a fact to read. The problem is founders reading the impressions line as audience size, watching it climb through a quarter, and concluding the audience is growing when the frequency is what grew.
The three readings
Run the ratio across your last ten posts and it sorts into three shapes. None of these are benchmarks. Nobody outside LinkedIn has published how the platform de-duplicates, and anyone handing you a "good" ratio invented it. Your own account is the control group, and you compare posts to each other.
Low ratio, high members reached. The post travelled. It got out of your existing room, probably through a commenter whose network doesn't overlap yours, and most of the people who saw it saw it once. This is what growth looks like, and it frequently comes with a modest impressions figure because a post that spreads wide doesn't need to be re-served.
High ratio, flat members reached. The post was popular inside the room and never left it. Impressions look strong. The number of humans who encountered you didn't move. If you publish a month of these, your dashboard says reach is up and your inbound says nothing changed, and the two are not in conflict. The same 2,000 people saw you three times.
Ratio rising across months. The room is saturating. Each post is reaching a slightly smaller set of new people and being served more often to the set it already has. This is the earliest signal we know of that a lane has been fully mined at its current depth, and it shows up in the ratio a quarter before it shows up anywhere else.
The "impressions are up and inbound is flat" complaint, explained
This is the single most common thing a founder says to us in month five. Reach is climbing. Nothing is arriving. The instinct is to blame the content, the hook, the CTA.
Pull the ratio. Most of the time impressions climbed because frequency climbed. Members reached has been flat for six weeks. The account hasn't been reaching new buyers; it's been reaching the existing buyers more often, and existing buyers who haven't bought yet don't convert on the fourth exposure of the same argument. They convert on a new argument or a new proof.
That changes the fix entirely. You don't need a better hook. You need the post that travels, which almost always means a claim specific enough that an adjacent operator disagrees with it in the comments and brings their network with them.
What it changes about how you report
Three things, and none of them require new tooling.
Grade posts on members reached, not impressions. It's the number that counts people. A post that reached 4,000 humans once outperformed a post that reached 1,500 humans three times, regardless of which one has the bigger headline figure.
Report the ratio next to both. One extra column in whatever sheet you already keep. Sorting your archive by ratio tells you which posts left the room and which ones stayed in it, and the ones that left are the ones to study for structure.
Watch the trend, not the post. A single high-ratio post is a post that landed with the core audience, which is fine. A rising ratio over eight weeks is a lane depth signal and the response is to go one level down inside the subject, not to widen it. Widening resets the room. Deepening finds the part of the existing room that hasn't been reached at that level.
The ten-minute audit
Open the analytics on your last ten posts. Two columns: impressions, members reached. Divide. Sort by the ratio.
Now look at the bottom three and the top three. The bottom three, the low-ratio posts, are your travellers. Read them for what they have in common. In our experience it's a specific number, a named mechanism, or a claim that a peer could argue with. The top three, the high-ratio posts, are your room posts. They're usually the ones that felt best to publish and got the warmest comments from people who already follow you.
Both kinds are useful. Only one kind grows the number of people who can eventually hire you.
FAQ
Is a high ratio bad? No. Frequency is how positioning works; an owned claim becomes owned by being seen more than once. A high ratio is bad only when you've been reading it as growth. Report both numbers and the ratio and it stops being misleading.
What's a normal ratio? We're not going to give you one and you should distrust anyone who does. LinkedIn hasn't published its de-duplication method and account size changes the arithmetic. Compare your posts to your own posts.
Does members reached include people outside my network? Members reached is a unique count across everyone the post rendered for. LinkedIn separately shows the share of impressions that came from your network versus beyond it in the same panel. Read that alongside the ratio; a low ratio with a high beyond-network share is the cleanest travel signal available.
Can I see this on older posts? Post analytics are available on your own posts from the post itself. Check how far back your account's analytics go; we'd start with the last quarter rather than the archive.
Should I stop caring about impressions? Keep tracking them; they're the number the platform tells you first and the number peers will ask about. Just stop reporting them as reach. Reach is the other line.
If you want the reporting rebuilt around the numbers that count people rather than renders, that's part of what we do for founders in the first month.