Twitter Engagement Rate: How to Calculate and Improve It

Twitter engagement rate is the share of people who interacted with a post, and on X it is usually calculated as engagements divided by impressions. That’s the number X shows in its analytics. Benchmark reports often use a different formula, engagements divided by followers, so the two numbers can’t be compared with each other. For your own tweet performance, the useful comparison is each post against your own average.

Below: both formulas, what the published benchmarks say and don’t say, and how to raise the rate on your next posts.

How is engagement rate calculated on X?

There are two common formulas, and they answer different questions:

Formula	Calculation	What it tells you	Who uses it
By impressions	Engagements ÷ impressions × 100	Of the people who saw the post, how many reacted	X’s own analytics
By followers	Engagements ÷ followers × 100	How much of your audience reacts, whatever the reach	Benchmark reports, many brand tools

An example with made-up numbers: a post gets 2,000 impressions and 60 engagements, from an account with 5,000 followers.

By impressions: 60 ÷ 2,000 = 3%
By followers: 60 ÷ 5,000 = 1.2%

Same post, two different numbers. Always check which formula a tool or report uses before you compare.

What counts as an engagement on X?

In X’s analytics, engagements cover every interaction with a post, not only likes. That includes replies, reposts, likes, bookmarks, clicks on your profile, clicks on links, and clicks to expand the post or its media. Tools that calculate engagement themselves sometimes count only likes, replies and reposts, which gives a lower number for the same post.

That’s one more reason to compare like with like: your posts, measured the same way, over time. X analytics shows where to find each metric.

What is a good engagement rate on Twitter?

It depends on your niche, your account size and the formula, so treat any single number with care. A widely cited public benchmark is Rival IQ’s yearly report:

In its 2025 Social Media Industry Benchmark Report, the median engagement rate per tweet across all industries was 0.015%, measured by followers.
The sample was brand accounts: 150 companies from each of 14 industries, each with at least 5,000 followers on X.
Rival IQ reported that engagement fell on every platform it tracks, and that X saw the largest drop, 48%, compared with the year before.
Rival IQ also notes it can’t measure engagement by impressions for these accounts, because of platform API restrictions.

What that means for you:

The 0.015% figure describes brands, measured by followers. A creator or founder posting in their own voice is a different kind of account.
It doesn’t apply to X’s analytics number, which divides by impressions and will look much higher.
Your own average is the benchmark that matters. A post well above your average is a pattern to repeat. A post well below it is a lesson.

We haven’t found a public benchmark for engagement by impressions that we trust, so we won’t give you one.

How do you calculate your own average?

Fifteen minutes, once a month:

Export or copy the stats for your last 30 posts (impressions and engagements). The full dashboard needs X Premium; without it, open each post’s stats (X Premium tiers).
Calculate the rate for each post, by impressions.
Take the median, not the average, so one viral post doesn’t skew it.
Mark posts at least twice your median as winners and posts under half of it as misses.

Then look for what the winners share: topic, format, first line, time of day. Best time to post on X covers the timing part.

How do you improve your Twitter engagement rate?

Engagement rate goes up when more of the people who see a post react to it. The levers you control:

1. The first line. Readers decide fast whether to stop. Lead with the specific part: the number, the claim, the result. See these Twitter hooks examples.

2. Topics your niche already cares about. Posts about what your readers are discussing this week get more reactions than posts from a blank page. Find what works in your niche on X shows how to check.

3. Posts that invite a reply. A clear opinion or a specific question gets replies, and replies are a strong signal for reach (how the X algorithm works).

4. One idea per post. Two points in one post often get neither. Split them.

5. Links in the first reply. Keep the main post readable on its own and put the link in the reply. Why the first reply works for links.

6. Repeat what already worked. Your winners from last month are a proven base for next month. Reuse your best tweets shows how to bring them back without copying.

7. Your voice, not a template. Generic posts get scrolled past. If you write with AI, feed it your own best posts first (make AI tweets sound less generic).

What not to do: buy engagement, join engagement groups, or automate likes and replies. They inflate the number without growing the account, and X’s rules prohibit artificial amplification.

Where does Kalen fit?

Kalen is an X assistant that connects to Claude, ChatGPT, Cursor or Claude Code through MCP. He is not an analytics dashboard. He uses performance data for one job: helping your AI write posts like the ones that already work.

Every day he researches your niche: the day’s top posts for your keywords and X trends. Once a week, the most popular posts of your reference accounts and of your own account.
Your AI writes the posts from that briefing. Kalen runs no writing model.
Every draft is scored from 0 to 100 against what works in your niche and your own best posts. Weak drafts go back for up to 3 rewrites, and a weak post is shown as weak, never hidden.
Nothing is scheduled until you say yes in the chat. Posts go out through the official X API. No auto-likes, auto-replies or engagement automation.

A post costs 1 credit, 2 with a first reply or a link. Starter is $9/month (100 credits), Pro $29/month (700), Ultra $49/month (1,500), each starting with 14 days free and 50 credits. Cancel before the trial ends and nothing is charged. See pricing.

FAQ

How do I calculate engagement rate on Twitter?

Divide a post’s engagements by its impressions and multiply by 100. That’s the rate X’s analytics uses. Some tools divide by followers instead.

Why is my engagement rate high but my followers aren’t growing?

Engagement rate measures reactions, not follows. Check new follows and profile visits per post too, and fix your bio and pinned post if visits don’t turn into follows.

Does a bigger account have a lower engagement rate?

By followers it often looks lower, because not every follower sees every post. Dividing by impressions only counts people who saw the post, which is another reason to use it for your own posts.

Is engagement rate the best way to measure tweet performance?

It’s the best single number for comparing posts. Pair it with new follows per post to see which posts grow the account. For the bigger plan, see how to grow on Twitter.