If you want to know what a good LinkedIn engagement rate is in 2026, you do not need another vendor survey. You need percentiles from a large dataset with a disclosed methodology. Everything below comes from one source: GTM Brigade's State of LinkedIn, a live, continuously updated report built on our own tracking corpus. This snapshot was generated on August 3, 2026.
No number in this post is borrowed from a third-party report. Where other reports' claims appear, they are labeled as exactly that — claims — in the closing section.
Use the tables the way an operator would: find your follower cohort in the first one, sanity-check the per-post distribution in the second, then adjust for your country and industry with the third and fourth. The fifth covers format. Each table is followed by one paragraph of interpretation that stands on its own if that is all you read.
Where these numbers come from
Every figure here comes from one dataset, measured one way, with the skew reported instead of averaged away. The State of LinkedIn corpus covers 45,966 posts over a rolling 60-day window, 619,520 tracked profiles, and 18,957 ranked creators. Engagement rate means total engagement over 60 days divided by follower count, computed per creator — not per post, and not divided by impressions. We publish percentile bands (p25, median, p75, p90) instead of averages, because the distribution is so heavily skewed that a mean would be dragged by a handful of viral outliers and describe nobody. One honest limitation: the corpus skews toward B2B creators and the networks around them. It is not a random sample of LinkedIn, and every number below should be read with that in mind. If your audience is B2B — founders, sales teams, marketers, consultants — that skew works in your favor, because you are being compared against accounts that look like yours rather than against the whole platform.
LinkedIn engagement-rate benchmarks by follower cohort
There is no universal good engagement rate — only a good rate for your size.
| Follower cohort | Creators | p25 | Median | p75 | p90 |
|---|---|---|---|---|---|
| Under 1,000 | 476 | 0.79% | 2.14% | 5.43% | 10.96% |
| 1,000–10,000 | 4,812 | 0.31% | 0.74% | 1.79% | 3.77% |
| 10,000–100,000 | 1,809 | 0.10% | 0.24% | 0.51% | 0.96% |
| 100,000 and up | 97 | 0.12% | 0.20% | 0.42% | 0.65% |
This table splits engagement rate — total engagement over 60 days divided by followers, computed per creator — into follower cohorts, using percentile bands because the distribution is too skewed for averages to mean anything. The drop is close to an order of magnitude: the median creator under 1,000 followers runs 2.14%, while the median creator past 100,000 followers runs 0.20%. So there is no universal good engagement rate, only a good rate for your size. If you have 5,000 followers, 0.74% is exactly average, 1.79% already puts you in the top quarter of your cohort, and 3.77% puts you in the top tenth. Notice what this does to the flat 1.5–2% industry average that circulates in vendor reports: it describes almost nobody, sitting above p75 for the 1,000–10,000 cohort and six to eight times the median for anyone larger. Benchmark against your cohort's median and p75, never a blended number.
What reactions per post actually look like
The median post gets 11 reactions, and the average post does not exist.
| Percentile | Reactions per post |
|---|---|
| Median | 16 |
| p75 | 49 |
| p90 | 146 |
| p95 | 331 |
| p99 | 1,614 |
| p99.9 | 8,233 |
These are raw reactions per post across all 45,966 posts in the 60-day window, and they show why average engagement is a broken concept on LinkedIn. The median post earns 11 reactions. Reaching the top 10 percent takes 146; the top 1 percent starts at 1,614; the top 0.1 percent at 8,233 — a five-hundredfold spread between the middle and the extreme tail. That tail is where the engagement lives: the top 1% of posts capture 44.3% of all engagement in the dataset, and the Gini coefficient across posts is 0.866, more unequal than any national income distribution. Any average-reactions-per-post figure is therefore dragged upward by a handful of viral outliers and describes no typical post. Benchmark yourself against percentiles instead: if your posts regularly clear 49 reactions, you are already beating three-quarters of everything published in this corpus.
Median engagement rate by country
Where you post from changes what normal looks like, by a factor of three.
| Country | Creators | Median ER | p75 |
|---|---|---|---|
| Australia | 112 | 1.20% | 3.24% |
| Netherlands | 150 | 0.84% | 1.89% |
| Germany | 259 | 0.81% | 1.89% |
| United Kingdom | 797 | 0.71% | 2.08% |
| Canada | 210 | 0.61% | 1.45% |
| France | 218 | 0.56% | 1.38% |
| United States | 2,463 | 0.55% | 1.35% |
| Spain | 132 | 0.52% | 1.05% |
| Romania | 878 | 0.48% | 1.29% |
| India | 299 | 0.37% | 0.85% |
Median engagement rates shift meaningfully by country, so a global benchmark quietly punishes some markets and flatters others. Australian creators post the highest median in this set at 1.20% (p75 3.24%), roughly triple India's 0.37% median and more than double the United States at 0.55%. Mid-pack sits most of Europe: Germany at 0.81%, the Netherlands at 0.84%, the United Kingdom at 0.71%, with France, Spain, and Romania between 0.48% and 0.56%. Two honest caveats apply. First, sample sizes vary widely — the United States row rests on 2,463 creators while Australia rests on 112, so smaller rows carry more noise. Second, these gaps likely reflect audience density and network maturity as much as content quality; a market where LinkedIn is newer produces different feed mechanics than a saturated one. If you sell into one market, compare yourself with that market's median and p75, not a worldwide blend.
Median engagement rate by industry
The people who sell LinkedIn engagement have some of the worst engagement in the dataset.
| Industry | Creators | Median ER | p75 |
|---|---|---|---|
| Retail | 174 | 1.51% | 3.76% |
| Computer Software | 2,013 | 0.62% | 1.45% |
| Financial Services | 217 | 0.61% | 1.56% |
| IT & Services | 778 | 0.55% | 1.35% |
| Marketing & Advertising | 541 | 0.44% | 1.05% |
| Management Consulting | 297 | 0.34% | 0.90% |
| B2B SaaS Sales | 63 | 0.31% | 0.51% |
| GTM / Go-to-Market | 64 | 0.30% | 0.53% |
| Founder-Led Sales | 85 | 0.21% | 0.35% |
| Social Selling & LinkedIn Growth | 68 | 0.21% | 0.60% |
Industry changes the baseline as much as follower count does, and the ranking is uncomfortable for the people who talk about LinkedIn the most. Retail creators lead this set with a 1.51% median engagement rate (p75 3.76%), well ahead of Computer Software at 0.62% and Financial Services at 0.61%. At the bottom sit the categories that sell LinkedIn itself: Social Selling and LinkedIn Growth at 0.21%, Founder-Led Sales at 0.21%, B2B SaaS Sales at 0.31%, and GTM at 0.30%. Read that again — the niches whose entire pitch is engagement post into the most saturated, most jaded feeds in the corpus and earn among the lowest median rates in it. Saturation is real, and audiences habituate fast. If you operate in one of those crowded categories, a 0.5% engagement rate is genuinely strong; if you are in Retail, the same number means you are underperforming your peers.
Post format vs engagement
Image posts lead on every average, but averages over a skewed distribution are a prior, not a playbook.
| Post feature | Avg reactions | Avg comments | Share of posts |
|---|---|---|---|
| Image | 138 | 25 | 51.4% |
| Text only | 79 | 11 | 43.1% |
| Carousel (document) | 58 | 18 | 2.3% |
| Native video | 51 | 13 | 3.1% |
| Long — over 1,200 chars | 131 | 27 | 30.9% |
| Medium — 300 to 1,200 chars | 101 | 17 | 52.9% |
| Short — under 300 chars | 89 | 7 | 16.2% |
| With hashtags | 75 | 8 | 35.0% |
| Without hashtags | 126 | 24 | 65.0% |
| With a link | 119 | 16 | 27.6% |
| Without a link | 104 | 19 | 72.4% |
Across the 67,577 posts we could analyze for content features in the 60-day window, image posts lead on both averages — 138 reactions and 25 comments — while making up half of everything published; text-only posts average 79 and 11. Carousels and native video trail on reactions, though carousels hold their own on comments at 18. Length correlates positively: posts over 1,200 characters average 131 reactions and 27 comments, against 89 and 7 for posts under 300 characters. Hashtags correlate negatively, at 75 average reactions with them versus 126 without. Two warnings before you rebuild your content calendar around this table. These are averages over a brutally skewed distribution, so a few viral image posts can move a cell. And they are correlations, not causation — strong creators may simply favor images and skip hashtags. Treat the table as a prior for experiments on your own account, not a formula.
The industry average is still a lie — and now we can prove it
A flat industry-average engagement rate is not a simplification of this data; it is a contradiction of it. When the top 1% of posts capture 44.3% of all engagement and the Gini coefficient across 45,966 posts is 0.866 — concentration beyond any national income distribution — a mean is not a summary, it is a distortion. When cohort medians run from 2.14% under 1,000 followers to 0.20% past 100,000, an order of magnitude apart, a number that averages those groups describes none of them.
Held against this dataset, the benchmarks that circulate look like this:
- Vendor reports cite a 1.5–2% industry average. In our cohort table, that range sits above the p75 of the 1,000–10,000-follower cohort (1.79%) and is many times the 0.24% median for creators past 10,000 followers. Whatever those reports averaged, the result is not a target that maps onto real accounts.
- Vendor reports claim video earns a multiple of other formats' engagement. In our 67,577-post feature sample, native video averages the fewest reactions of any format — 51, against 138 for image posts.
- Vendor reports claim company pages earn a small fraction of personal-profile engagement, and some circulate a specific multiple. Our dataset tracks personal creator profiles, so we cannot verify any multiple. The direction may well be right; the number is somebody's marketing until they publish percentiles.
- Some reports claim most of a post's reach is decided in its first 30 minutes, and that overall reach has collapsed versus earlier years. We do not measure time-resolved or historical reach, so we can neither confirm nor deny — and neither can most of the people repeating those figures.
The deeper problem with any rate benchmark, including ours, is that it measures how interesting you are to the internet, not to buyers. A creator can sit at p90 for their cohort while every reaction comes from peers and cheerleaders. The number a revenue team should actually watch is engagement from the accounts that can buy from them — their ICP watchlist — and no public benchmark can tell you what that should be, because it depends entirely on who is on your list. Ten reactions from decision-makers at accounts in your pipeline are worth more than a thousand from strangers, and no percentile table — ours included — can price that for you. Build the list first (how to build a LinkedIn watchlist), weight engagement by who it came from (the ICP-weighted engagement rate calculator), and run the play as a team (the for-sales-reps motion).
So the next time someone quotes you a LinkedIn benchmark, ask four questions: which percentile, which cohort, which denominator, and what sample size. If they cannot answer, it is marketing. The version that can answer — every table above, recomputed continuously on a rolling 60-day window — lives at State of LinkedIn. And if you want your own number instead of a benchmark, the free LinkedIn audit scores your motion — activity, reach, ICP-network match — in two minutes.