Ask three people what a "good" engagement rate is and you will get three different numbers — and all three can be right. That is because engagement rate is not one metric; it is a family of formulas measured against wildly different baselines. TikTok's average engagement is roughly seven times Instagram's, a 2,000-follower creator and a 2-million-follower creator live in different numeric worlds, and even the same account scores differently depending on whether you divide by followers, views, or reach. The benchmarks in this post give you realistic 2026 targets for Instagram, TikTok, YouTube, LinkedIn, and X — by follower tier, with the exact formula each one assumes, so you can tell whether you are actually underperforming or just comparing apples to oranges.
TL;DR: Judge engagement only within the same platform, tier, and formula. Healthy 2026 floors by followers: Instagram — nano (under 10K) 5%+, micro (10K–100K) 3%+, mid (100K–500K) 1.5%+, macro (500K–1M) 1.2%+. TikTok runs 2–3x higher structurally: nano 10%+, micro 7%+, mid 4%+, macro 3%+. YouTube is different entirely — measure likes-per-view (4%+ healthy), comments-per-view (0.3%+), and views-per-subscriber (15%+ over 30 days) instead of a follower-based rate. Median reality is lower than the "healthy" floors: half of creators sit under ~1.7–2.2% on Instagram and ~0.8–2.9% on TikTok depending on tier, per 2026 creator databases. Never compare across tiers, and never compare an average to a median. Below the floor for your tier means something is wrong; far above it can mean bought engagement.
Why your engagement rate is probably not broken
Before the numbers, the context that stops creators from spiraling: engagement rates have been falling for years on every platform, and 2026 is no exception. Reach-based rates on Instagram now sit around 1–3% for healthy accounts, and follower-based averages are even lower — large industry studies of 70 million posts put Instagram's average engagement near 0.5% and X near 0.1%. That is not because audiences stopped caring. It is because platforms push content to far more non-followers than they used to, which inflates the denominator of any rate and drags the percentage down. The environment changed; you did not fail.
The practical consequence: compare yourself to tier-and-format-matched peers, not to a single magic number, and never to a number from 2019. The rest of this post gives you the current reference points.
The three formulas (use the right one)
Engagement rate is one idea with three common implementations, and the benchmark tables below are only valid for the formula they cite:
| Formula | What it measures | When to use it |
|---|---|---|
| Followers-based: (likes + comments + shares) ÷ followers × 100 | How your existing audience responds | Comparing creators of similar size; brand outreach |
| Reach-based: (likes + comments + shares) ÷ reach × 100 | How people who actually saw the post responded | Your own content performance; post-by-post iteration |
| Views-based: (likes + comments) ÷ views × 100 | Engagement among actual viewers | YouTube long-form; TikTok when view counts are available |
The followers-based version is the industry default because it only needs public data — but it is the most misleading one, because a viral post seen mostly by non-followers produces a deceptively low rate while a quiet post seen only by superfans produces a deceptively high one. When you analyze your own content, prefer reach-based or views-based numbers. When brands analyze you, they will use followers-based — so know what your followers-based number looks like, but do not run your creative life by it.
Instagram benchmarks (2026)
Healthy follower-based floors for feed posts, per tier — these are the numbers brands and agencies use to vet creators:
| Follower tier | Healthy floor (2026) | Median reality |
|---|---|---|
| Nano (under 10K) | 5%+ | ~2.2% |
| Micro (10K–100K) | 3%+ | ~1.7% |
| Mid (100K–500K) | 1.5%+ | ~1.5% |
| Macro (500K–1M) | 1.2%+ | ~1.2% |
| Celebrity (1M+) | 0.8%+ | ~0.9% |
Notes that matter: engagement falls as accounts grow because bigger audiences are less intimate — a macro creator at 1.2% is outperforming a nano creator at 3% relative to their size, so never compare across tiers. Reels reach far more non-followers than static posts, so their followers-based rate runs lower while their actual distribution is higher — judge Reels by reach-based engagement or saves, not the follower formula. Carousels pull the strongest save rates; static single images the strongest like-per-reach. Format and niche shift the numbers more than effort does: in 2026, cars, fitness, and comedy niches run above the platform median, while skincare, finance, and home decor run below it.
TikTok benchmarks (2026)
TikTok's engagement runs 2–3x Instagram's at every tier, structurally — the For You feed distributes most content to non-followers, so dividing by followers inflates the rate. That is normal, not fraud. Healthy follower-based floors (including shares, which TikTok treats as a primary action):
| Follower tier | Healthy floor (2026) | Median reality |
|---|---|---|
| Nano (under 10K) | 10%+ | ~2.9% |
| Micro (10K–100K) | 7%+ | ~0.8–2% |
| Mid (100K–500K) | 4%+ | ~0.4% |
| Macro (500K–1M) | 3%+ | ~0.3% |
Two things make TikTok numbers slippery. First, the median is far below the healthy floor because most videos flop and a few explode — a creator might see 25% on one video and 0.3% on the next purely because the algorithm decided which to push. Compute the median engagement across your last 9–12 videos, not the average, and you get an honest picture. Second, if you can see view counts, use the views-based formula — it strips out the algorithm-distribution effect and gives a much steadier signal. In 2026 the views-based healthy range lands around 3–6% for most tiers, roughly half the followers-based numbers.
YouTube benchmarks (2026)
YouTube breaks the formula hardest, because subscribers are a vanity denominator — most of your subscribers do not watch any given video. The metrics that matter are all view-based:
| Metric | Healthy floor | Excellent |
|---|---|---|
| Likes per view (long-form) | 4%+ | 6%+ |
| Comments per view (long-form) | 0.3%+ | 1%+ |
| Views per subscriber (30-day rolling) | 15%+ | 30%+ |
| Average view duration | 40–50%+ of video length | 60%+ |
A channel with 800,000 subscribers whose recent videos get 25,000 views has an audience-fit problem — the subscribers were earned in a previous content era. Views-per-subscriber exposes it: under 15% over 30 days means your subscriber base is stale and you should be priced and pitched on views, not subs. Shorts behave like TikTok, not long-form: benchmark them separately against views-based ranges, since a Short's audience is mostly non-subscribers.
LinkedIn and X benchmarks (2026)
LinkedIn engagement is the highest per impression of any major platform, because content is distributed mostly within your own network — reach-based rates of 2%+ are healthy for creators, and strong posts clear 5%. X is the harshest environment: average engagement sits near 0.1–0.2% in large studies, and a "good" post lands in the low single digits. X is also the platform where replies and quote-posts matter more than likes — a post with 200 replies but 800 likes is performing better than one with 2,000 likes and no conversation. Judge X by conversation depth, not the rate.
What to do with your number
Benchmarks tell you when something is wrong, not what to do next. The escalation path:
- Compare like-for-like. Same platform, same formula, same tier, same format. A Reels rate is not comparable to a carousel rate.
- If you are at or above the floor: your problem is distribution, not content quality — work on hooks, timing, and formats that travel.
- If you are below the floor: check whether your audience is actually seeing your content (reach collapsing?) or just not responding (saves and shares near zero?). Low reach is an algorithm problem; low saves is a content problem.
- If you are suspiciously far above the floor: check for bought followers or engagement pods — a nano account at 40% is usually paying for engagement, and brands will spot it.
- Track the conversion layer. Engagement is a means, not the end. The number that pays your bills is how many of those engaged people click through to your newsletter, your shop, or your link-in-bio — measure profile visits and link clicks next to engagement so you know whether your content is earning attention or just collecting it. A link-in-bio tool like Biolinky gives you click data on every destination so you can see which posts actually move people toward a follow, a signup, or a sale.
How niches shift the numbers
Your niche moves your benchmark more than almost anything you control. Audiences in some topics simply engage more — the content is aspirational, opinionated, or community-driven. Others scroll past quietly. 2026 data across creator databases shows the spread clearly:
| Platform | Above-median niches | Below-median niches |
|---|---|---|
| Cars, fitness, comedy, pets | Skincare, finance, home decor | |
| TikTok | Cars, beauty, fashion | Education, how-to (higher saves, lower likes) |
| YouTube | Gaming, commentary, vlogs | Tutorials, finance (high watch time, low like rate) |
The practical rule: if your niche runs below the platform median, hold your content to the median of your niche, not the platform-wide floor — a finance creator at 1% on Instagram is outperforming a fitness creator at 1.5%, because finance audiences engage less but convert more. That last part is why engagement is a proxy, not the prize. Finance and B2B audiences leave few likes but click links, book calls, and buy. When you benchmark, filter for your niche first; when you judge success, look past the rate to the actions.
How to actually move your number
Benchmarks diagnose; these levers fix. If you are below your tier floor, work them in this order:
- Ask for the action. A direct question in the caption or on-screen ("Which of these should I try first? Comment 1 or 2") reliably lifts comments 2–3x over a passive caption. Engagement responds to invitations.
- Reply to every comment for the first hour. Comments beget comments — the algorithm treats early conversation as a quality signal, and early repliers pull in more. Batch your publishing time so you can sit with the post for 60 minutes.
- Design for saves and shares, not likes. Saves and shares are the actions platforms weight most in 2026 — Instagram's own signal hierarchy is watch time first, then likes per reach, then sends and shares per reach. Make content people bookmark ("5 templates to steal") or send to a friend ("text this to your business partner"), and your distribution improves even if your like count does not.
- Cut the formats dragging you down. If your carousels average 4% and your static images 1%, the format is the problem, not your audience. Double the winners and stop posting the losers for 30 days.
- Post for the audience you have, not the one you want. Content that assumes inside knowledge confuses casual viewers and they scroll; content a newcomer can grasp earns the engagement that grows you toward the audience you want.
Expect change over 4–6 weeks, not days — engagement is a lagging indicator, and the algorithm needs repeated signals before it re-classifies your content.
Red flags that have nothing to do with your content
- Sudden rate spike with flat reach: bought engagement or a pod — brands will audit you and walk away.
- High likes, zero link clicks: entertainment without intent. Fine if you are building reach; fatal if you are selling.
- High saves, low follows: people value the post but not you. Add a reason to follow (series, POV, weekly format).
- Rising reach, falling rate: normal — new viewers engage less than your core. Watch absolute engagement, not just the percentage.
Your monthly 15-minute check
- Pull the last 9–12 posts per platform.
- Compute the median followers-based rate (for brand comparisons) and the median views/reach-based rate (for your own learning).
- Compare each against the tier floor above, and against your own trailing 90 days — your own baseline is the most meaningful benchmark you have.
- Log saves, shares, and link clicks, not just likes. Those are the actions that indicate an audience willing to buy.
A "good" engagement rate is not a number — it is a range that depends on platform, tier, formula, and format. Learn your platform's healthy floors, compare yourself only to like-for-like peers, and treat engagement as the middle of your funnel rather than the finish line. The creators who stop chasing a magic percentage and start watching what their engaged audience actually clicks are the ones whose numbers quietly stop mattering — because the money is already following.