What is the average engagement rate on Twitter?

There is no single site-wide average worth quoting, because the by-followers rate falls by roughly an order of magnitude from nano to mega accounts — an "average" across all sizes describes nobody. Compare against the band for the account’s own tier instead: the table on this page gives below-average, good and excellent ranges for each. If you want a like-for-like number across sizes, use the by-impressions rate, where 0.5–3% is typical.

How do you calculate engagement rate on X?

By followers: (likes + replies + reposts + quotes) ÷ followers × 100, averaged over the last 10–15 non-viral posts. By impressions: the same numerator ÷ impressions × 100. The two answer different questions — by-followers is what brands compare across creators, by-impressions is what tells you whether the people who actually saw a post cared. X shows impressions publicly on every post, which makes the second formula easier here than on any other platform. The free calculator runs it either way.

Why is my Twitter engagement rate so low?

Three common causes before you assume the audience is fake. Follower counts on X accumulate over many years and a large share of old accounts are dormant, so the denominator inflates while the real audience does not. X Premium changes algorithmic distribution, which means you may not be in the same comparison pool as the accounts you are measuring against. And reply-heavy accounts get engagement that never shows on their own posts. A rate far below the tier band with almost no replies at all is the pattern that actually suggests inflated followers.

What Is a Good Engagement Rate on X (Twitter) in 2026?

Q: What is a good engagement rate on X (Twitter)?

It depends almost entirely on size. By followers: 3–6% is good for a nano account under 10K, 1.5–3% for micro (10K–100K), 0.8–2% for mid (100K–500K), 0.5–1.5% for macro (500K–1M) and 0.3–0.8% for mega accounts over 1M. A nano creator at 5% is doing exactly what you would expect; a mega account at 5% would be extraordinary. By impressions, 0.5–3% is a good benchmark at most sizes.

X engagement rate is the most misread metric in influencer marketing — mostly because the platform exposes two completely different numbers depending on which formula you use, and because X’s historically high bot count means even a healthy-looking rate deserves a second look. Before you benchmark a creator or pitch your own rate to a brand, you need to know which calculation you’re running.

The short version: by followers, the typical X account in 2026 lands at about 1.1%, ranging from a 2.18% median for nano accounts (1K–10K) down to 0.35% for accounts over 500K, according to SociaVault Labs’ 2026 engagement benchmarks (40,000 X accounts, February–March 2026, fraud-flagged accounts excluded). Brand accounts run far lower — Socialinsider’s 2026 benchmarks put the brand average at 0.10% — so be sure which population a benchmark describes before you compare against it.

Two formulas — and why they matter more on X than anywhere else

By followers (most common for cross-platform brand comparisons):

(Average likes + replies + reposts) ÷ followers × 100

By impressions (X’s native metric — surfaced in creator analytics):

(Average likes + replies + reposts) ÷ impressions × 100

On X the two formulas can give very different answers for the same account. X surfaces content heavily through algorithmic feeds, replies, and quote posts — so the gap between who sees a post and who follows the account is enormous. A creator with 80,000 followers might get 600,000 impressions on a breakout thread, or 4,000 impressions on a quiet post. Neither figure tells you much about the other.

What this means practically: when a brand quotes your “X engagement rate,” assume by-followers unless they say otherwise — that’s what third-party tools default to. When you’re diagnosing your own content or deciding whether a post landed, use by-impressions — it tells you whether the people who actually saw it cared.

X engagement rate benchmarks by follower tier (2026)

The table below is from SociaVault Labs’ 2026 engagement benchmarks. It uses the by-followers formula — (likes + replies + reposts) ÷ followers × 100 — on 40,000 X accounts measured February–March 2026, with accounts flagged as fraudulent excluded. Read the 25th percentile as “below average” and the 75th as “excellent” for that tier. Note SociaVault’s tier boundaries, which differ from the usual 10K/100K/1M cut-offs.

Follower tier	25th percentile	Median	75th percentile
Nano (1K–10K)	1.32%	2.18%	3.41%
Micro (10K–50K)	0.87%	1.42%	2.24%
Mid (50K–100K)	0.58%	0.97%	1.54%
Macro (100K–500K)	0.37%	0.63%	1.02%
Mega (500K+)	0.19%	0.35%	0.58%

The same study puts the overall X median at 1.38% in 2024, 1.22% in 2025 and 1.11% in 2026 — a 9% drop in the latest year.

The same pattern holds here as on every other platform: smaller, tighter accounts engage at higher percentages. A nano creator at 2% is about at the median for the tier; a mega account at 2% would be more than three times its tier’s 75th percentile.

Other sources measure brands, not creators, and come out much lower. Socialinsider’s 2026 benchmarks (Jan 2026; 10.3 million organic X posts from brand profiles, likes + reposts ÷ followers) put the average at 0.10%, down from 0.15% in 2024 and 0.12% in 2025. The Rival IQ/Quid 2026 Social Media Industry Benchmark Report (March 2026; 150 companies per industry) puts the median X brand post at 0.03% of followers, up from 0.015% the year before. These are not in conflict with the creator table; they describe company accounts, which engage far less than people do. Use the creator figures to judge a creator.

By impressions there is no tier table we could verify. Ordinal’s study of X posts (January 2024–May 2026, sample skewed toward B2B and startup accounts) uses (likes + replies + reposts + quotes) ÷ impressions and reports roughly 0.79% for image posts, 0.68% for text-only posts and 0.54% for video. Metricool’s X statistics (Feb 2026; 1.1 million posts from 15,116 accounts) reports average engagement rising from 1.32 to 1.58 between 2024 and 2025, but does not state its formula, so do not compare it to either of the above. Viral threads and breaking-news reactions can spike far above any of these — those outliers skew your average and shouldn’t be included in a stable benchmark you’d show a brand.

Worked: why X is the one platform where you can check both

X shows a view (impression) count publicly on every post, which most platforms don’t. So you can run both formulas on an account that is not yours — and the gap between them is the most informative number on the platform.

An account with 34,000 followers averages, over 12 posts: 210 likes, 18 replies, 46 reposts, 9 quotes — 283 interactions — against 21,500 impressions per post.

By followers: 283 ÷ 34,000 × 100 = 0.83% By impressions: 283 ÷ 21,500 × 100 = 1.32%

By followers, 0.83% sits just under the 0.87% 25th percentile for SociaVault’s 10K–50K tier. By impressions, 1.32% is above every per-format average in Ordinal’s sample. Both are true, and together they say something neither says alone: the content engages the people who see it, but it is only reaching about 63% of the follower count per post.

That is a distribution problem, not an audience-quality one — and the fix is different. A dead audience shows low numbers on both formulas. A distribution problem shows exactly this split: healthy by impressions, weak by followers.

The pattern that should actually worry you is the inverse — high by followers, low by impressions — which usually means a small number of accounts engaging very heavily on posts almost nobody saw. That is the shape of a pod.

Niche changes the picture significantly

The only niche figures we could verify for X come from SociaVault Labs’ 2026 engagement benchmarks: politics is the highest-engaging niche at a 1.84% median, fashion the lowest at 0.68% (by followers). The rest of this list is qualitative — our read of how these niches behave, not measured benchmarks:

Finance and crypto: A dense, opinionated, reply-heavy community — a good engagement signal, but also a space where coordinated boosting is common.
Politics and news commentary: The top niche by median, but bursty rather than consistent. One viral political take inflates the average massively — always look at the median, not the mean.
B2B and marketing: Lower per-post engagement numbers but high-intent audiences. Saves and link-clicks matter more than likes here.
Entertainment and celebrity: Broad audiences with lower per-follower engagement — expect large accounts here to sit toward the bottom of their tier.
Tech and developer community: Thread-heavy content performs well. Long-form threads consistently outperform single tweets on every engagement metric.

If you’re comparing two creators on X, compare within the same niche first. Cross-niche comparisons routinely lead to the wrong conclusion.

What low rates signal on X — and why it matters more here

A below-average X engagement rate has a few possible explanations:

Natural niche effect — the account operates in a category where lower rates are structural (entertainment, sports commentary on large accounts).
Inactive or stale audience — the creator was active in an earlier era of Twitter, accumulated followers, then shifted content or posting cadence. Old followers stop engaging; the denominator stays large.
Bought or inauthentic followers — X has one of the highest concentrations of bot accounts of any social platform. A large following that doesn’t engage is a stronger red flag on X than the same pattern on Instagram.

The difference between option 2 and 3 isn’t always visible in the engagement rate alone. What gives away purchased followers is the composition: accounts with no bio, no profile photo, no posting history, or accounts following 5,000 people while having two posts and zero followers of their own. You can see this in a follower audit — or use our authenticity checker to surface it quickly.

X Premium and algorithmic distribution distort the comparison pool

Paid X Premium accounts are widely believed to get extra algorithmic distribution. We could not verify a published measurement of whether, or by how much, that raises impressions, so we won’t put a number on it. If it does, the direction is what matters: if paid distribution adds impressions without adding proportionate interactions, a paying creator’s by-impressions rate goes down on the same content quality.

When you compare two creators on X, treat Premium status as noise. By-followers still provides the most stable cross-account comparison because follower count doesn’t fluctuate based on subscription status.

Engagement quality signals that matter

Two X accounts at 2% can be worth entirely different amounts. The signals that separate quality engagement:

Reply quality. X comment sections (replies) are highly visible and searchable. Thoughtful replies that engage with the creator’s actual argument signal a real, invested audience. Generic “great post!” or emoji replies can indicate a pod or a bought engagement pattern.
Repost vs. quote-post ratio. Quote posts (commenting while reposting) require more effort and signal real opinion. A creator whose reposts are mostly quote posts has an audience engaging with their ideas, not just amplifying them reflexively.
Link-click rate. If you have access to creator analytics, link CTR is a strong predictor of whether the audience acts on recommendations — which is what brands are actually paying for.

When pitching to a brand, frame your engagement rate with context: “2.4% engagement by followers, with [average replies] replies per post, mostly substantive” says far more than a bare percentage.

What to actually aim for

Calculate your rate over your last 10–15 non-viral posts using the by-followers formula — that’s what brands will compare.
Find your tier in the table, then adjust for niche.
If you’re at or above your tier’s median, lead with that number in pitches and media kits.
If you’re below, investigate before adding followers. More followers at a low rate makes the ratio worse and flags you harder in an audit.

On X in 2026, engagement quality matters more than on most other platforms — precisely because the platform’s history of bots has trained brands and tools to scrutinize it harder. A 2% rate with genuine, readable replies and a follower list that holds up to a spot check is worth far more than a 4% rate built on inflated numbers. Make sure yours can stand the audit.