LinkedIn analytics beyond vanity metrics
· Updated · Zybble editorial
Impressions are the sugar of social analytics: instant, addictive and nutritionally empty. A post can reach tens of thousands of feeds and change nothing. If you are building a business on LinkedIn, the question analytics must answer is narrower and harder: is the right audience paying attention, and is that attention compounding?
Vanity metrics are not wrong because they are numbers; they are wrong because they are inputs you do not control and outputs you cannot spend. An impression is a feed rendering, not a person reading. Optimising for it produces posts that interrupt many strangers instead of persuading the few who can buy, refer or follow your work. The metrics below are the ones that survive contact with that test.
The metrics, defined honestly
- Impressions — how many times your post was displayed in feeds. Cheap to earn, easy to inflate, no guarantee anyone looked. Useful mainly as the denominator for everything else.
- Reach — the number of unique people who saw it. Closer to 'audience', still short of 'attention', and harder to get at precisely for member posts, which is why most tools work in impressions.
- Engagement rate — reactions, comments and reposts divided by impressions. The first number that measures response rather than exposure.
- Follower movement — whether your posts are causing people to subscribe to your thinking. The compounding mechanism of the whole channel.
For the precise relationships between the first three — and why they get conflated so often — see impressions vs reach vs engagement on LinkedIn. The rest of this article is about the two numbers worth managing.
Engagement rate over reach
Reactions, comments and reposts divided by impressions is the first honest number. It strips out the lottery of distribution and tells you how the people who saw the post responded. Track it per post, then look at the median across a month — single posts lie, medians do not.
A worked example with round numbers. Say a post earned 2,000 impressions, 40 reactions, 12 comments and 6 reposts. Engagement rate is (40 + 12 + 6) ÷ 2,000 = 2.9%. On its own, 2.9% means nothing — there is no universal grade for engagement rate, and anyone selling you one is guessing. What matters is your own baseline: compute the median of your last ten posts, then ask whether this one is above or below the typical post to your audience. Relative movement is the signal.
The full arithmetic, common pitfalls and the levers that genuinely move this number are covered in how to calculate and improve your LinkedIn engagement rate.
Comments are the closest thing to revenue
For consultants and founders, the metric closest to pipeline is the quality of comment section you generate. Comments from people in your target market are leads wearing a disguise. Count them. Read them. Reply to every one — the reply doubles the comment count and often starts the real conversation in DMs.
Go one step further and keep a simple tally: of the comments on your last ten posts, how many came from people who plausibly fit your customer profile? A post that draws twenty comments from fellow creators teaches you less about demand than one that draws three from buyers. This number never appears in any dashboard by default — which is exactly why it is worth an hour of your attention.
Benchmarks are folklore; your median is a fact
The internet is full of 'average LinkedIn engagement rate' figures, and almost all of them are useless for one of three reasons: they mix personal profiles with company pages, they define engagement rate differently than you do, or they describe an audience that is not yours. Benchmarks answer 'how do I compare to everyone?' — a question with no bearing on any decision you will actually make.
Your own median answers the only question that matters: 'is my next post better or worse than my typical post?' Improve against that line for a year and the external benchmarks will take care of themselves.
A simple weekly review
- Which post had the highest engagement rate — and what did it have in common with your other winners?
- Who commented: strangers, peers, or the people you actually sell to?
- Did follower growth accelerate after any specific topic or format?
- What one variable will you deliberately change next week?
That last question is the point of analytics. If the review does not change what you do next, you are not measuring — you are spectating. Fifteen minutes a week is enough, provided the numbers are already in one place rather than scattered across LinkedIn's per-post screens.
This review loop is the design brief for Zybble's analytics: per-post performance first, then the trend, then the decision. And when you want to connect the numbers to commercial outcomes rather than content quality, what LinkedIn content ROI actually looks like for B2B picks up exactly where this article stops.
What is a good engagement rate on LinkedIn?
There is no trustworthy universal number — definitions, populations and time frames vary too much between sources. The honest benchmark is your own trailing median: the median engagement rate of your last ten posts. A post is 'good' when it beats that line, and your presence is improving when the line itself drifts upward over the months.
Are impressions or reach more important?
Neither is a goal; both are denominators. Impressions tell you how widely the feed displayed the post and reach tells you how many unique people were involved — but the number you manage is the response per unit of exposure. Fixating on the numerator metrics (impressions, reach) produces broad, shallow content; managing engagement rate produces content that works on the audience you actually have.
How do I see analytics for my LinkedIn posts?
LinkedIn shows basic per-post numbers (impressions, reactions, comments, reposts) beneath each post and in the analytics section of your profile. The limitation is not access but assembly: comparing dozens of posts over months means exporting and joining the data yourself. That is the specific problem a dedicated tool solves — Zybble computes per-post performance and trends automatically for every post it publishes.
Can I see who viewed my LinkedIn post?
LinkedIn shows who reacted and commented, but it does not reveal who viewed a post, and profile-view identities are only partially visible depending on the other person's settings and your account. Treat 'who engaged' as the usable signal — it is public by design and it is the group most likely to become buyers or referrals.
Reach is what the algorithm did. Engagement is what your audience did. Confuse the two and you will optimise for the wrong boss.
