How to calculate your Instagram engagement rate
ยท Ethan Parker, Editor
Divide a post's interactions by the accounts it reached, then multiply by 100. The four numbers to add up, and why every online calculator disagrees.
Divide the interactions on a post by the accounts it reached, then multiply by 100. That is the whole calculation. A post with 120 interactions that reached 3,000 accounts works out at 4 percent, and every extra step people bolt onto that sum is either a different denominator or a different idea of what an interaction is.
The carve-out arrives immediately: most engagement rate calculators online do not use reach at all. They divide by follower count, because followers are public and reach is not. The two answers can sit far apart on the very same post, so a rate means nothing until you say which denominator produced it. A rate quoted for an account somebody does not own is always the follower version, because the reach version is invisible from outside.
The mechanic, on a post of your own: open it from your profile, tap View insights underneath it, add up likes, comments, shares and saves, divide that total by the line called Accounts reached, then multiply by 100. Insights only exist on a professional account, the free Business or Creator switch inside Settings, so on a personal account the reach version of this sum is not available to you at all.
Everything after this is bookkeeping: what belongs on top of the fraction, what belongs underneath it, and what to compare the answer against.
What is the engagement rate formula?
Interactions divided by accounts reached, then multiplied by 100. Written out: likes plus comments plus shares plus saves, divided by accounts reached, times one hundred. There is no secret weighting and no hidden step.
- Add up the interactions on the post: likes, comments, shares and saves.
- Find the number of accounts the post reached. It sits on the same insights panel, usually on a line called Accounts reached.
- Divide the first number by the second.
- Multiply by 100 to turn the result into a percentage.
A worked example, with numbers invented for the arithmetic: a post picks up 180 likes, 12 comments, 20 shares and 40 saves. That is 252 interactions. It reached 4,200 accounts. 252 divided by 4,200 is 0.06, and multiplied by 100 that is 6 percent. Run the same sum on ten of your own posts and you have a baseline worth more than any figure quoted for an account you cannot see inside.
Which interactions do you count?
Likes, comments, shares and saves, on almost every version of the formula. Those four are what Instagram reports on a post, and they are the four most people mean. Profile visits and link taps sit on the same panel but are usually left out, because they measure what happened after the post rather than to it.
- Likes. Still counted when an account hides like counts from the public, because the owner can always see their own.
- Comments. Replies inside a thread count as comments too, including your own replies, which quietly inflates a busy post.
- Shares. Sends to a chat and reposts both land here. The post panel gives you one total and does not split them.
- Saves. The quietest of the four and often the most telling, since somebody saving a post intends to come back to it.
The decision that matters is not which four you choose but that you choose the same four every time. A rate with saves included, set against one without them, is not a comparison at all. Reels add plays and watch time to the panel, and those belong in a different sum.
Why do two calculators give different numbers?
Because they divide by different things. Three denominators are in common use, and each one answers a slightly different question.
- Divided by accounts reached. Answers: of the people who actually saw this, how many did something? The honest version, and it needs insights, so only the owner can work it out.
- Divided by followers. Answers: how loud is this post next to the audience I claim? Anyone can calculate it from outside, which is why public calculators use it.
- Divided by impressions or views. Answers: how many interactions per appearance? It returns the smallest of the three, because one person can see a post several times.
So one post can look strong on one measure and ordinary on another, and neither number is lying. A post that travelled well past your followers scores high on the follower version and unremarkably on the reach version. The lower two are not interchangeable either, since reach and impressions count different things, and swapping one denominator for the other is where most confusing comparisons begin.
How do you work out the rate for a whole account?
Average the posts rather than adding them together. Take your last ten or twenty posts, calculate the rate for each, and take the middle value. That middle value, the median, survives one runaway post far better than a plain average.
- Pick a fixed number of recent posts, ten or twenty, and use that same number every time.
- Work out the rate for each post on its own.
- Sort the results and take the middle one. That is your baseline.
- Repeat monthly. The direction the baseline moves is the number worth watching, not any single post.
One trap: do not use the account-level reach figure as the denominator for a month of posts. That screen counts each account once for the whole window, so it is far smaller than the sum of the post figures, and dividing a month of interactions by it produces a percentage that describes nothing.
Does the rate fall as an account grows?
Usually, and it is not a fault in your posting. A larger following holds more people who followed you for one thing two years ago and now scroll straight past. The same content, sent to a looser audience, returns a lower percentage even while the raw count of likes climbs.
This is also why a falling rate is not automatically bad news. If your reach doubled and your interactions rose by half, the rate drops while more people than ever are engaging. Read the raw counts next to the rate before concluding anything. If the rate and the raw counts are sliding together, that is the case worth digging into, and it usually starts with delivery rather than with the post itself, which is worked through in why reach falls.
What the engagement rate will not tell you
It will not tell you whether the engagement was worth having. A hundred likes from accounts that will never buy, read or return count the same in the sum as a hundred from the people you made the post for. No formula can weigh one against the other.
It will not give you an honest comparison against anyone else either. Published benchmarks use the follower denominator on accounts whose reach nobody outside can see, so measuring your reach-based rate against one compares two different sums. Your own earlier posts are the only fair comparison you have, which is argued at more length in what counts as a good rate.
If the point of the calculation is a bigger number rather than a better sum, that is a different job. ReachGrows sells one prepaid month: you set the least and the most reach a post of yours should get, and every feed post and reel you publish in the next 30 days is aimed inside that band, up to 30 posts. ReachGrows computes every price on its own server, not in your browser, so the price you are shown is the price charged. It sends reach, views, likes, shares, reposts and saves. It does not send comments and does not sell followers, so if your formula counts comments, that part of your fraction will not move.
The short version
- What is the engagement rate formula? Interactions divided by accounts reached, multiplied by 100.
- Which interactions count? Likes, comments, shares and saves. Pick your four and keep using the same four.
- Where do I find the numbers? Under your own post, behind the View insights link, on a professional account.
- Why does every calculator disagree? They divide by followers, not by reach, because reach is not public.
- Average or median? The median of your last ten or twenty posts. One runaway post ruins an average.
- Does a falling rate mean trouble? Not by itself. The rate drops whenever reach grows faster than interactions.