Key finding
Across 211 LinkedIn posts published by one team of about 10 accounts, the median matured post earned 1,778 impressions and 35.5 reactions, while the average matured post earned 5,516 impressions, 3.1 times the median, because a small number of outlier posts carry most of the reach.
What we measured
Most LinkedIn benchmark posts quote averages from surveys. This one measures every post one team actually published, with the impression and engagement counts LinkedIn itself reports.
The dataset is a single workspace: our own. About 10 LinkedIn accounts run by the Reachium team, publishing through the platform's content engine. It is the same engine we sell, pointed at ourselves, and it is how we fill our own calendar. A companion study measures that funnel end to end, from post comments to booked meetings.
That framing matters for reading the numbers. This is not a random sample of LinkedIn. It is one B2B team posting deliberately, most of it in a lead-magnet format that asks readers to comment. Where that format distorts a metric, the study says so.
Methodology
Dataset window. 211 published posts, 12 November 2025 to 5 August 2026, pulled on 6 August 2026. Ten active accounts published them; 10 early posts belong to one further account that has since been removed.
Maturity. Impressions accrue for weeks. In this dataset, posts 0 to 7 days old at the pull showed a median of 39.5 impressions and posts 8 to 14 days old showed 821, far below settled levels. So every headline figure uses the matured cohort: the 132 posts published at least 30 days before the pull. The full 211-post series is quoted only where it is labelled as such.
Metrics. Impressions, reactions, comments and reposts are LinkedIn's lifetime totals for each post, as refreshed between 21 May 2026 and the pull date. Engagement rate is reactions plus comments plus reposts, divided by impressions, computed per post, and only for posts with at least 100 impressions so tiny denominators cannot manufacture extreme rates. That leaves 169 of the 211 posts.
One format note. 182 of the 211 posts are lead-magnet posts, which ask readers to comment a keyword. That inflates comment counts far above what a standard post sees, so this study reports comment figures with that context attached rather than as a general benchmark.
The findings
Impressions per post, matured cohort (n = 132)
| Percentile | Impressions |
|---|---|
| 10th | 31 |
| 25th | 369 |
| Median | 1,778 |
| 75th | 5,237 |
| 90th | 11,826 |
| Maximum | 90,777 |
The average is 5,516, which is 3.1 times the median. Anyone benchmarking against an average LinkedIn number is benchmarking against outliers. Half of this team's matured posts earned fewer than 1,778 impressions, and a quarter earned fewer than 369.
The matured cohort earned 728,127 impressions in total; the full 211-post series earned 919,385. For reference, the full series' median is 1,401, lower than the matured median because it includes posts still accruing.
Engagement per post, matured cohort
| Metric | Median | Average | Maximum |
|---|---|---|---|
| Reactions | 35.5 | 45.4 | 262 |
| Comments | 80 | 156.5 | 1,778 |
| Reposts | 0 | n/a | 23 |
The comment figures are not a normal-post benchmark. The dominant lead-magnet format asks for comments, and the medians reflect that. A separate study in this series compares the two formats directly; standard posts in this dataset show a median of 1 comment. The repost median of 0 is worth its own sentence: more than half of all matured posts were never reposted once, and the best post managed 23.
Engagement rates (full series, posts with 100+ impressions, n = 169)
| Rate | Median |
|---|---|
| Engagement rate | 5.05% |
| Reaction rate | 1.45% |
| Comment rate | 3.76% |
The comment rate is format-driven, which the split confirms: lead-magnet posts show a median engagement rate of 5.84% against 1.65% for standard posts. These are medians of per-post rates. A different basis exists and produces a different number: summing all engagements across the whole series and dividing by total impressions gives 3.81%. The sum basis runs lower because high-impression posts engage a smaller share of their larger audiences, so a rate quoted without its basis is not comparable to either figure.
Day of week, matured cohort, descriptive only
| Day | Posts | Median impressions |
|---|---|---|
| Mon | 17 | 1,845 |
| Tue | 24 | 1,655.5 |
| Wed | 25 | 794 |
| Thu | 19 | 558 |
| Fri | 13 | 2,710 |
| Sat | 17 | 3,386 |
| Sun | 17 | 1,824 |
Read this as a description of one team's calendar, not as scheduling advice. Which authors posted on which days is not controlled here, the per-day samples are 13 to 25 posts, and author identity moves medians in this dataset far more than weekday plausibly does.
By month, matured cohort, descriptive only
| Month | Posts | Median impressions |
|---|---|---|
| Nov 2025 | 1 | 0 |
| Feb 2026 | 4 | 31,905 |
| Mar 2026 | 10 | 7,824.5 |
| May 2026 | 35 | 2,639 |
| Jun 2026 | 51 | 1,602 |
| Jul 2026 (1st to 6th, maturity cutoff) | 31 | 607 |
No posts were published in December, January or April. The table moves with posting cadence, author mix and season all at once, so it supports no causal claim in either direction. The early months are a handful of posts that landed big; the later months are systematic daily output.
What this means
The distance between the median and the average is the single most useful thing here. LinkedIn reach is not normally distributed, and any content plan built on average numbers will overestimate what a typical post does by a factor of three.
A median of 1,778 impressions with about 10 accounts posting is also a genuinely strong result to benchmark against, not a humble one. The 25th percentile of 369 is what a below-par week looks like for the same team with the same process.
The engagement rates travel better than the raw counts. Impressions depend on account strength; a median reaction rate of 1.45% of impressions is closer to a property of the content itself, and it is the number we would compare across teams first.
In practice
A founder judging their own content should compare a post against the median, not the average, and against a month of posts, not a single one. In this dataset a quarter of matured posts landed under 369 impressions with the same team, the same tools and the same process that produced the 90,777-impression outlier.
A marketer setting expectations for a 10-account employee advocacy program has a defensible baseline here: roughly 728,000 impressions from 132 matured posts over the measured period. The median engagement rate to expect alongside that, measured across the full series' posts with at least 100 impressions, is 5.05%.
Limitations
One team, one workspace. All 211 posts come from Reachium's own accounts. Our niche is B2B lead generation; other niches will see different levels, and nothing here is a random sample of LinkedIn.
The lead-magnet format dominates. 182 of 211 posts ask for a comment, which inflates comment counts and, through them, the engagement rate. The reaction rate is the cleanest cross-team metric in the study.
Account strength varies widely inside the team. Median matured impressions per account range from 167.5 to 6,627, so the pooled medians blend strong and weak accounts. Per-account claims are deliberately absent because most accounts have too few matured posts to carry one.
Impressions are LinkedIn's own counts. They are lifetime totals as of the refresh date, they include repeat views by the same person, and LinkedIn does not document their edge cases. The single November 2025 post records 0 impressions, most likely because the post no longer exists on LinkedIn.
Maturity is a 30-day cutoff, not a guarantee. Posts keep accruing slowly after 30 days, so matured figures are near-final rather than final.