Free tool

Engagement rate calculator

Two rates, both shown: interactions per view and interactions per follower — plus honest benchmarks by account size. 100% in your browser.

Engagement rate by views

(likes + replies + reposts) ÷ views

Engagement rate by followers

(likes + replies + reposts) ÷ followers

Rough benchmarks
Follower bandRate by followersRate by views
Under 1k followers5–10%+ is strong3–6%+ is strong
1k – 10k2–5% is strong2–4% is strong
10k – 100k1–2% is strong1.5–3% is strong
100k+0.5–1% is strong1–2% is strong

These are rules of thumb from practitioner experience, not published study data. Rates fall as accounts grow because follower counts inflate faster than active-reader counts. Judge your trend against yourself first.

Two rates, two questions

"Engagement rate" sounds like one number, but the two common formulas answer different questions. Interactions ÷ views asks: of the people who saw this, how many cared? It's a pure content-quality signal, unaffected by account size, and it's the number to watch when you're testing formats, hooks, or topics. Interactions ÷ followers asks: how much of my audience does a typical post activate? It's noisier — inflated by viral reach beyond your followers, deflated by dead followers — but it's comparable across accounts because follower counts are public and view counts aren't always. Serious practitioners track both and refuse to average them into one number, because when they diverge, the divergence is the insight: high by-views but low by-followers means your content is good and your distribution is the bottleneck. The reverse means your loyal audience carries posts the algorithm won't.

How to read the benchmarks honestly

The table above is labeled "rough benchmarks" deliberately. There is no authoritative, audited dataset of X engagement rates by follower band — anyone quoting precise figures is quoting someone's marketing content. What practitioners consistently observe is the shape: rates fall as accounts grow, roughly by a factor of 2–4 per order of magnitude of followers. A 500-follower account where each post pulls interactions worth 8% of the follower count is doing genuinely well; a 50,000-follower account at 8% would be extraordinary. So use the bands as sanity checks, not grades. The comparison that actually drives decisions is you-versus-you: this month's average against last month's, this format against that one.

What moves each component

The three interaction types respond to different things, and diagnosing which one is low tells you what to fix. Likes are the cheapest signal — they track agreement and polish, and they rise when your take is clean and relatable. Replies track conversation — they rise when you ask real questions, take positions people can push back on, or leave a deliberate gap for others to fill. A post that's completely airtight gets likes but no replies; there's nothing left to say. Reposts track identity — people repost what they want to be seen endorsing, which is why useful reference material (checklists, frameworks, hard-won numbers) and sharp opinions travel while diary entries don't. If your rate is low overall, look at the mix first: an account with likes but no replies needs more open questions, not better writing.

The failure modes of chasing the number

Engagement rate is a useful thermometer and a terrible boss. Optimizing it directly produces predictable pathologies: engagement-bait questions ("What's one book that changed your life?") that spike replies from people who will never buy, read, or remember you; hot takes you don't hold; and deleting honest posts that "underperformed." The number also punishes exactly the posts that grow an account — experimental formats and new topics almost always engage below your average at first, because your existing audience didn't follow you for them. Treat below-average engagement on a deliberate experiment as tuition, not failure.

A simple weekly practice

Once a week, pull your last 10–20 posts and compute the two averages. Flag the top two and bottom two posts, and write one sentence about each: what did the winners share, what did the losers share? Over a month, those sentences converge into an actual content strategy — one derived from your audience's behavior rather than someone else's advice thread. That loop — post, measure, compare, adjust — is the whole game; the calculator just makes the measuring step take thirty seconds instead of a spreadsheet session.

Frequently asked questions

Which engagement rate should I use — by views or by followers?

By views, when you can get it. It measures how compelling your content was to the people who actually saw it, independent of how big your account is. The by-followers rate is easier to compare across accounts (views aren't always public) but is distorted by inactive followers. Track both; trust by-views for content decisions.

What counts as an interaction?

This calculator uses likes + replies + reposts, the three universally visible signals. Broader definitions add bookmarks, link clicks, profile clicks, and quote posts — if you have access to those in X Analytics, they're worth watching, especially bookmarks, which many practitioners treat as the strongest quality signal.

What's a good engagement rate on X?

It depends heavily on account size. As rough rules of thumb: under 1k followers, interactions worth 5–10% of your follower count per post is strong; at 1k–10k, 2–5%; at 10k–100k, 1–2%; above 100k, 0.5–1%. By views, 2–4% is solid for most accounts. These are practitioner heuristics, not audited studies — your own trend line matters more.

Why do engagement rates fall as accounts grow?

Because follower counts only go up while attention doesn't. Old followers go inactive, interests drift, and the algorithm shows each post to a shrinking fraction of a growing list. A falling rate with rising absolute engagement is normal growth; a falling rate with flat absolute engagement is the signal to investigate.

Should I average over one post or many?

Many. Single posts are dominated by luck — time of day, one big repost, algorithm mood. Average your last 10–20 posts, exclude the outliers in both directions, and recompute every couple of weeks. The moving average is the metric; individual posts are noise.

Does this tool send my numbers anywhere?

No. The math runs entirely in your browser. Nothing you enter is transmitted or stored.

Stop doing this by hand.

DeckPost drafts in your voice, schedules at your peak windows, and reviews the numbers with you every week.

DeckPost does this automatically, daily →