How to calculate engagement rate on X, with a worked example

Published By

Engagement rate on X is the number of reactions a post gets, divided by the number of times it was seen, times 100. Using public figures only, that is likes plus reposts plus replies plus quotes, over views. For a whole account, work the rate out post by post and take the median, so that one post that went everywhere does not decide the result.

Engagement rate is a simple fraction that people manage to calculate in three incompatible ways. Here is the arithmetic, one example done by hand, and the choices that change the answer.

The formula

  • By views: (likes + reposts + replies + quotes) ÷ views × 100
  • By followers: (likes + reposts + replies + quotes) ÷ followers × 100

The top of the fraction is the same in both. What changes is what you divide by: the people who saw the post, or the people who follow the account.

A worked example

Take one post with 4,000 views, 60 likes, 12 reposts, 9 replies and 3 quotes. The reactions add up to 84.

RateSumResult
By views84 ÷ 4,000 × 1002.1%
By followers, with 10,000 followers84 ÷ 10,000 × 1000.84%

Same post, same reactions, two numbers that differ by more than double. Neither is wrong. They answer different questions, which is why a rate without its method next to it says very little.

By views or by followers?

By views is the more honest of the two today. A post is shown to people who do not follow the account, and plenty of followers never see it, so the follower count is a poor stand-in for the audience of one post. Views are that audience, counted.

The rate by followers is the older way to measure. It is still useful to compare with older figures, and for posts that carry no view count.

From one post to a whole account

An account has many posts, and they do not behave alike. Suppose five posts have rates of 1.2%, 1.5%, 1.8%, 2.1% and 9%. The last one was picked up by a large account.

  • The average is 3.12%. Four of the five posts are below it.
  • The median, the value in the middle, is 1.8%. That is what a normal post from this account does.

Use the median. One lucky post should not rewrite the description of an account, and with the average it always does.

Which posts to count

  • Posts and quotes written by the account. Threads included.
  • Not replies to other people. A reply lives in someone else's conversation and is seen mostly by the people in it.
  • Not reposts. They are another account's posts.
  • Nothing younger than 24 hours. A post that just went out has not collected its views or its reactions yet.
  • At least 5 posts. Below that, leave the figure out.
Example Mootmap account card: engagement rate 2.40%, median views 1.8K, 14 posts a week, and a timetable of the week with the busiest hour and the hours with the most views marked.
An example card, for an invented account: three figures on the left, the week it keeps on the right.

Why it will not match X Analytics

X uses the same words for a wider count. X's own definitions describe engagements as every interaction with a post, including clicks anywhere on it: links, media, the username, the profile photo, opening the post. Its engagement rate is those engagements divided by impressions. Clicks are not public, so a rate built from the outside is lower than the one an account owner sees. The two are compared side by side in X Analytics vs the public numbers.

Common questions

What counts as an engagement on X?

From outside an account, four things can be counted: likes, reposts, replies and quotes. X's own analytics, which only the account owner sees, also count clicks on the post, its links, its media and the profile.

Are bookmarks part of the engagement rate?

Not in the public rate described here. X shows a bookmark count under a post, but a bookmark is a private save, not a reaction other people see. It is worth reading next to the rate, not inside it.

How many posts do I need for a rate that means something?

At least 5, each more than 24 hours old, and more is better. With three or four posts the result is an anecdote about those posts, not a description of the account.

Is there a good engagement rate on X?

There is no single figure that holds for every account, and this site does not print one: it has no dataset to base it on. Compare an account with itself over time, or with accounts of a similar size measured the same way on the same day.

Skip the arithmetic: the X Account Analyzer works the rate out post by post for any public account. It is free and needs no login.