Social media engagement rate:
by reach, by followers or by impressions

Why two people quote different rates for the same post

Engagement rate looks like a precise number, but it is a ratio with a choice buried in it. Total interactions, such as likes, comments, shares and saves, sit on top. What goes below is up to the person calculating: followers, reach, impressions or video views. Change the base and the rate changes, even though the post is identical.

This is why a creator claims 4% while a brand's analyst measures 2%. Neither is wrong. They are dividing the same 640 interactions by different counts of people or views.

The formula

Define engagements in writing too. Some analysts count only likes, comments and shares. Others include saves, profile visits and link clicks. A broader definition gives a higher rate.

Some teams also compute a weighted score where a share counts for more than a like. That is a useful internal tool, but it is not comparable with a plain rate, so keep it separate and label it clearly.

When you only have public information, a rough rate is still possible. Add likes and comments on a post and divide by the follower count. It will omit shares and saves that only the owner can see, so treat it as a lower bound and use it only to compare accounts measured the same way.

One post, three different rates

The follower base gives the highest number here because only a share of followers see a given post, and the post also reached non-followers. Because the three figures answer different questions, none of them is the right one in the abstract.

Choosing a base for the job

A practical rule is to use reach for your own content analysis, followers for outside comparisons where only public numbers are available, and impressions for anything that involves paid delivery.

Why the rate falls as an account grows

Smaller accounts typically have higher engagement rates than large ones. Followers are closer to the creator, the audience is more tightly knit, and the algorithm shows posts to a larger share of them. As the audience grows, the share who see and react to each post thins out. A rate of 6% at 5,000 followers and 1.5% at 500,000 is a normal pattern and not a sign of failure.

When you benchmark, therefore, compare accounts of a similar size and in the same niche. A general figure quoted across all accounts hides this difference and is of limited use.

Platform and format matter as much as size. Carousel posts, short video and stories each draw interaction differently, and the same account can show very different rates across them. Segment by format before you draw conclusions about what is working.

Reading the number for influencers and brands

• A high rate on a small audience can be real but not scalable; check whether it holds when the audience grows.
• Comments that are generic emojis or one-word replies are weaker signals than detailed questions or saves.
• A sudden spike may come from a giveaway or an engagement pod and says little about genuine interest.
• Look at the median across 10 to 20 posts, not the best post, which the account will have shown you first.

Collecting the data correctly

1. 1Choose a window, such as the 30 days after a post goes live, because engagement keeps arriving for days and early reads understate it.
2. 2Export the numbers from the platform's own analytics, not from a screenshot or the public count.
3. 3Use the same definition of engagement for every post in the sample.
4. 4Record the follower count at the time of posting if you use followers as the base, since it changes.
5. 5Compute the rate per post, then take the median across the set.

A common error is to add up all engagements across posts and divide by the follower count once. That treats a month of posts as a single event and gives a number far above any single post's rate. Compute per post and then summarise.

If you compare competitors, take a public sample of their last 10 to 20 posts, note followers and engagements, and compute a followers-based rate for each. It will be crude, but it is applied equally to everyone.

Typical mistakes in reports

• Comparing a rate by followers for one account with a rate by reach for another, which are not the same measure.
• Including promoted posts in organic averages, since paid reach changes the denominator and the audience.
• Using a single viral post to describe an account's performance rather than the median.
• Counting follower growth or profile visits as engagement without saying so.
• Quoting a rate to one decimal place from a post with a few dozen interactions, which is more precision than the data supports.

A clean report states the period, the number of posts, the engagement definition and the base at the top. Anyone can then check the arithmetic, and the same layout can be repeated next month for a fair comparison.

Beyond the rate

A rate tells you how active the audience is relative to its size, and nothing about whether it buys. For a business objective, pair engagement with downstream numbers such as link clicks, sign-ups and sales. A post with a modest 1.5% rate that drives lots of clicks may be worth more than a viral one that drives none.

Remember too that platforms change what they count and how they show it, and some hide public like counts. Record your own definitions and keep them stable over time so that trend lines mean something.