# How to calculate engagement rate: formulas, examples and what each network counts

Canonical: https://postmesh.app/blog/social-media-engagement-rate

By PostMesh team. Published 2026-10-31; updated 2026-10-31.

Engagement rate is interactions divided by an audience number, multiplied by 100. Divide by followers to compare accounts, by reach to see how the people who saw a post responded, by impressions to see how often each showing led to an interaction, or by views for video; always say which one you used, because the same post can score 4% one way and 10% another.

Engagement rate is the most quoted number in social media reporting and one of the least consistent. Two people can look at the same post and report 4% and 10%, and both can be right, because they divided by different things. This guide gives you every common formula with worked examples, shows what Instagram, Facebook, X, LinkedIn, YouTube and Threads actually count according to their own documentation, explains why your rate might be falling when nothing is wrong, and quotes the only benchmarks we could tie to a named, dated study.

## The short answer

Add up the interactions you've decided to count, such as likes, comments, shares and saves. Divide by your chosen audience number. Multiply by 100. The audience number is the decision that matters:

| Method | Formula | Answers the question |
| --- | --- | --- |
| By followers | Interactions ÷ followers × 100 | How did this post do relative to the size of the account? |
| By reach | Interactions ÷ reach × 100 | Of the people who saw it, how many interacted? |
| By impressions | Interactions ÷ impressions × 100 | How often did a showing lead to an interaction? |
| By views | Interactions ÷ views × 100 | How many interactions came with each recorded view? |

**Which engagement rate should you use?**
- Comparing your account with other accounts → By followers. Every account's follower count is public.
- Judging how good a post was → By reach, if your analytics show it. It only counts people who saw the post.
- Matching LinkedIn or X's own reporting → By impressions. That's the denominator their analytics use.
- Measuring a video or Reel → By views, and keep to it for every video you compare.

*Pick one per comparison and write it down.*

1. Add up the interactions
2. Pick one denominator
3. Divide
4. Multiply by 100
5. Label the formula

*The last step is the one most reports skip.*

If you just want the number, the free [engagement rate calculator](/tools/engagement-rate-calculator) works out all of these from the figures you enter and shows the formula with your numbers. The rest of this guide is about choosing the right inputs, because that's where most engagement rates go wrong.

## What is engagement rate?

Engagement rate is a ratio with two parts. The top, the numerator, is the interactions a post or account received. The bottom, the denominator, is some measure of how many people or showings those interactions came from. Multiplying by 100 turns it into a percentage. Neither part is standardized across networks or tools, so an engagement rate without its formula is close to meaningless.

### What counts as an interaction?

Usually likes (or reactions), comments, shares (or reposts) and saves. Some formulas add link clicks, profile visits, quotes, sends or follows. Each network counts a different set by default, which is one reason a rate from one network's analytics doesn't match a rate from another's. For example:

- Instagram's total interactions metric is likes, saves, comments and shares, minus unlikes, unsaves and deleted comments, according to Meta's [Instagram media insights reference](https://developers.facebook.com/docs/instagram-platform/reference/instagram-media/insights).
- LinkedIn's Page analytics count clicks, reactions, comments and shares as interactions, per LinkedIn's [content analytics help page](https://www.linkedin.com/help/lms/answer/a564051).
- X reports likes, replies, reposts, quotes and bookmarks publicly, and link and profile clicks to the post's author, per X's [metrics documentation](https://docs.x.com/x-api/fundamentals/metrics).

Pick one definition for your own reports and keep it. If you decide saves count, they count every month. If you add clicks, label the result so nobody compares it with a rate that excluded them.

### Should unlikes and dislikes count?

Networks handle negative actions differently. Instagram's total interactions already subtracts unlikes, unsaves and deleted comments, so it's a net figure. YouTube reports dislikes as a separate metric. Most people leave dislikes out of engagement rate, because the rate is meant to measure positive interest, but a few include every interaction on the grounds that any reaction means the post was seen and noticed. Either is defensible; just decide once and write it in your report's notes.

## How do I calculate engagement rate by followers?

Divide a post's interactions by the account's follower count, then multiply by 100. A post with 160 interactions on an account with 4,000 followers has an engagement rate by followers of 160 ÷ 4,000 × 100 = 4%.

This is the most common formula in published benchmarks because it only needs public numbers: anyone can see an account's followers and a post's likes and comments. That makes it the right choice for comparing your account with competitors or with a study. Its weakness is that not every follower sees every post, and posts can reach people who don't follow you, so a post that travels far beyond your followers can show a rate that looks impossibly high.

### Use the follower count from the day you posted

If your account grows quickly, the follower count today can be much larger than on the day an older post went out, which makes old posts look worse. Record the follower count when you collect each post's numbers, or at least once a week, and use the closest figure.

### How to measure a competitor's engagement rate

Because followers and public interactions are visible to everyone, you can calculate a rough follower-based rate for any public account. Take its last 10 to 20 regular posts, skip anything posted in the last day or two since those are still collecting interactions, add up the likes and comments you can see, divide by the number of posts to get an average, then divide by the follower count and multiply by 100. You won't see saves, shares from private messages or reach, so compare the result only with your own rate calculated the same limited way, from likes and comments only.

## How do I calculate engagement rate by reach?

Divide interactions by reach, the number of unique accounts that saw the post, then multiply by 100. The same 160 interactions from a post that reached 1,600 accounts is 160 ÷ 1,600 × 100 = 10%.

Reach-based rates are the best measure of how good a post was, because they only count people who actually saw it. The catch is that reach is private, so you can only calculate it for your own accounts, and it's an estimate. Meta's Instagram reference describes reach as the number of unique Instagram users who saw the media at least once and notes the metric is estimated. LinkedIn calls its equivalent members reached, the number of distinct members and Pages that saw your post, and also says it's an estimate that doesn't include repeat displays.

> Facebook's reach metric has changed. Meta's Page Insights reference marks post_impressions_unique, the old count of people who had a post enter their screen, as deprecated above Graph API v25, and offers a count of unique media viewers instead. If your Facebook reports changed wording in 2026, that's why.

## How do I calculate engagement rate by impressions?

Divide interactions by impressions, the number of times the post was shown, then multiply by 100. If the post was shown 2,400 times, the rate is 160 ÷ 2,400 × 100 = about 6.7%.

Impressions count repeat showings, so one person scrolling past twice counts twice. That makes impressions larger than reach and impression-based rates lower than reach-based ones. It's the denominator LinkedIn and X use in their own reporting:

- LinkedIn defines engagement rate on Page posts as the ratio of interactions per impressions, with interactions including clicks, reactions, comments and shares. It counts an impression when content is at least 50% on screen for 300 milliseconds or more, or when it's clicked.
- X's impression count is the number of times a post appeared on a screen, with repeat views counted again.

If you want your numbers to match what a LinkedIn Page admin sees, include clicks in the interactions and divide by impressions. If you leave clicks out, your rate will be lower than LinkedIn's, and that's fine as long as you say so.

## How do I calculate engagement rate by views?

Divide interactions by views, then multiply by 100. It's the natural choice for video, where views are the headline number. But a view means something different on each network, so view-based rates are only comparable within one network:

| Network | What a view is, per its documentation |
| --- | --- |
| Instagram | Total number of times the media has been played on Instagram (Meta's media insights reference) |
| Facebook | The number of times your content was played or displayed, including videos, posts, stories and ads (Meta's Page Insights reference) |
| Threads | The number of times your post was played or displayed (Threads insights reference) |
| YouTube | From August 24, 2026, counted the moment a video starts to play, across all formats (YouTube Help) |

The YouTube change matters if you track view-based rates over time. YouTube's [page on how engagement metrics are counted](https://support.google.com/youtube/answer/2991785) says that starting August 24, 2026, views are counted the moment a video starts to play across Shorts, long-form videos and live streams. YouTube says Partner Program earnings and eligibility aren't affected. But if views are counted earlier than before, the same interactions spread over more views give a lower rate, so a drop in your view-based rate around that date may be a counting change rather than a change in your audience.

### Pair a view-based rate with watch time

For video, interactions are only half the story, because many viewers watch without tapping anything. Networks report watch metrics you can put next to the rate. The YouTube Analytics reference defines average view duration as the average length of video playbacks in seconds, and lists engaged views, which it describes as views past the first frame or when the user clicks or taps to play. Meta's Instagram reference includes the average time spent playing a Reel. A video with a modest engagement rate but long average watch time is still doing its job.

## Worked example: one post, four engagement rates

Here's one post measured every way. It received 120 likes, 18 comments, 7 shares and 15 saves, which is 160 interactions. The account had 4,000 followers, the post reached 1,600 accounts, and it was shown 2,400 times.

| Method | Calculation | Engagement rate |
| --- | --- | --- |
| By followers | 160 ÷ 4,000 × 100 | 4.0% |
| By impressions | 160 ÷ 2,400 × 100 | 6.7% |
| By reach | 160 ÷ 1,600 × 100 | 10.0% |
| Likes and comments only, by followers | 138 ÷ 4,000 × 100 | 3.5% |

- 4.0%: By followers (160 ÷ 4,000)
- 6.7%: By impressions (160 ÷ 2,400)
- 10.0%: By reach (160 ÷ 1,600)
- 3.5%: Likes and comments only, by followers

*Same post, same week. Only the formula changed.*

The post didn't change; the formula did. That's why every engagement rate in a report needs its formula and its raw numbers next to it. A reader who sees 10% next to a competitor's 4% might think you're more than twice as engaging, when you may simply have divided by reach while the competitor's figure was by followers.

### Converting one rate into another

The formulas are linked. Engagement rate by followers equals engagement rate by reach multiplied by reach divided by followers. In the example, 10% × (1,600 ÷ 4,000) = 4%. That relationship is handy when a report gives you one rate and you need another: if you know roughly what share of your followers a typical post reaches, you can translate a reach-based rate into a follower-based one for comparison with a benchmark, and say so in a footnote.

## What does each network's analytics count?

Here's what each network's documentation lists for a single post, so you know what you can put into each formula. Menu names inside the apps change often, so this lists the metrics rather than tap paths.

| Network | Interactions you can count | Audience numbers available |
| --- | --- | --- |
| Instagram | Likes, comments, saves, shares, reposts, total interactions | Reach (estimated), views |
| Facebook Pages | Reactions by type, comments, shares, clicks | Views, unique viewers |
| X | Likes, replies, reposts, quotes, bookmarks; link and profile clicks for your own posts | Impressions |
| LinkedIn | Reactions, comments, reposts, saves, sends; clicks on Page posts | Impressions, members reached (estimated) |
| YouTube | Likes, dislikes, comments, shares, subscribers gained | Views, impressions |
| Threads | Likes, replies, reposts, quotes, shares | Views |

### Instagram

Meta's media insights reference lists comments, likes, saved, shares, reposts and total interactions, plus reach and views. Impressions are deprecated for media created after July 2, 2024, so for newer posts use reach or views as the denominator. Reach is estimated. To see these numbers in the app you need Insights on your account; [how to turn on Instagram Insights](/blog/how-to-turn-on-instagram-insights) covers that, and [how to see who liked your Instagram post](/blog/how-to-see-who-liked-your-instagram-post) covers the like list. The [Instagram engagement rate calculator](/tools/instagram-engagement-rate-calculator) is set up for reach or followers.

### Facebook

Meta's [Page Insights reference](https://developers.facebook.com/docs/graph-api/reference/insights) defines post views as the number of times your content was played or displayed, and offers unique media viewers, reactions by type, clicks, and Page post engagements, which counts reactions, comments, shares and more. It also says a number of Page Insights metrics will be deprecated for all API versions by June 15, 2026, and marks the old post impressions metrics as deprecated above v25. So for Facebook, views and unique viewers are the audience numbers still listed in Meta's reference.

### X

X's metrics documentation lists impressions, likes, replies, reposts (excluding quotes), quotes and bookmarks as public metrics, and link clicks, profile clicks and total engagements as private metrics only the author can see. It also notes that private and organic metrics are available only for posts created in the last 30 days, so collect them within a month if you want them in a report. If your impressions look unusually low, [the X shadowban test guide](/blog/x-shadowban-test) covers how to check whether your posts are being limited.

### LinkedIn

LinkedIn's [post analytics page](https://www.linkedin.com/help/linkedin/answer/a516971) lists impressions, members reached, reactions, comments, reposts, saves, sends, profile viewers and followers gained for posts on personal profiles. LinkedIn keeps discovery and engagement counts for 1,000 days but members reached for only 400 days, so export reach figures before they age out. Page analytics add clicks and an engagement rate by impressions, as covered above.

### YouTube

The [YouTube Analytics metrics reference](https://developers.google.com/youtube/analytics/metrics) defines likes, dislikes, comments, shares (through the Share button) and subscribers gained, alongside views and engaged views. YouTube Studio also reports impressions and impressions click-through rate; its [impressions page](https://support.google.com/youtube/answer/7628154) notes that not every thumbnail sighting counts as an impression, and impressions on external websites and end screens aren't included. YouTube Analytics reports dislikes too; [how to see dislikes on YouTube](/blog/how-to-see-dislikes-on-youtube) explains where you can find them.

### Threads

Meta's [Threads insights reference](https://developers.facebook.com/docs/threads/insights) lists views, likes, replies, reposts, quotes and shares for each post, with views and shares marked as in development. There's no reach figure, so Threads rates are usually by views or by followers.

> TikTok isn't in the table because we couldn't load TikTok's own help or developer pages to confirm its metric definitions for this guide. Use the metric descriptions shown inside TikTok's analytics, and [why your TikTok isn't getting views](/blog/why-is-my-tiktok-not-getting-views) for help with low numbers.

## How do I calculate engagement rate for a whole month?

There are two ways, and they give different answers. Pooling adds up all the interactions and divides by the total audience number. Averaging calculates each post's rate, then takes the average of those rates. Pooling weights big posts more; averaging treats every post equally.

| Post | Interactions | Reach | Rate |
| --- | --- | --- | --- |
| Post A | 50 | 500 | 10.0% |
| Post B | 300 | 10,000 | 3.0% |
| Pooled: 350 ÷ 10,500 | 350 | 10,500 | 3.3% |
| Average of the two rates |  |  | 6.5% |

**Pooled rate**
- Total interactions ÷ total reach
- 3.3% in the example
- Big posts count more
- Best for overall account performance

**Average of post rates**
- Mean of each post's rate
- 6.5% in the example
- Every post counts equally
- Best for typical post quality

*Both are valid. Label which one you report.*

Neither is wrong, but they answer different questions and must not be swapped between reports. For a follower-based monthly figure, a common approach is the average interactions per post divided by followers: if 12 posts earned 1,920 interactions in total and you have 4,000 followers, that's 160 interactions per post, or 4% per post.

### Don't pool across networks

Adding Instagram, LinkedIn and X interactions together and dividing by combined impressions or views produces a number that doesn't describe anything, because each network defines both parts differently. Report a rate per network, by that network's best available denominator, and if you need one overall figure for a summary, show the per-network rates beside it so readers can see what's driving it.

### Small numbers swing wildly

A post that reached 40 people and got 6 likes has a 15% rate by reach. One extra like makes it 17.5%. With small accounts or small samples, report the raw interactions alongside the rate, and compare groups of posts rather than single posts, so one lucky post doesn't steer your plans.

## What is a good engagement rate?

There's no single good number, because it depends on the network, the formula, the industry, the account's size and whether posts were paid. The most useful published reference we could tie to a named, dated source is Quid's [2026 Social Media Industry Benchmark Report](https://www.quid.com/knowledge-hub/resource-library/blog/2026-social-media-industry-benchmark-report), published March 5, 2026 (the report formerly published by Rival IQ). It sampled 150 companies at random from each of 18 industries, included organic and paid posts, and used the median per industry. Its engagement rate is total interactions divided by follower count.

| Network | All-industry median engagement rate by followers (Quid, 2026) |
| --- | --- |
| TikTok | 2.01% |
| Instagram | 0.30% (down from 0.36%) |
| YouTube | 0.21% |
| X | 0.03% |

Read those numbers with care. They're medians across 18 industries (one of which is influencers), they're by followers, and they mix organic and paid posts. A creator's account, or a rate by reach, will look very different and shouldn't be compared with them. The most reliable benchmark is your own history: your median rate over the last three months, by the same formula, for the same kind of post.

### How to use a benchmark without being misled

- Match the formula. Quid's figures are by followers, so compare them only with your own rate by followers.
- Match the network. A TikTok rate and an X rate come from different counting rules and audiences.
- Match the account type. The all-industry medians blend 18 industries, from aviation to influencers, so your own kind of account may sit well above or below them.
- Match the post mix. The study includes paid posts, so if yours are all organic, expect a different number.
- Prefer your own trend. A rate that rises over three months by the same formula tells you more than any outside median.

## Why is my engagement rate going down?

Before you change your content, check whether the denominator changed. A falling rate is often a measurement effect:

- Your followers grew faster than interactions. Rates by followers fall naturally as an account grows, even when each post gets more interactions.
- A network changed how it counts. YouTube began counting views when playback starts on August 24, 2026, and Meta deprecated impressions in its Instagram and Facebook insights data in favor of views. A larger denominator means a lower rate.
- A post was boosted or promoted. Paid distribution adds reach and impressions from outside your usual audience, which can move the rate either way, so mixing paid and organic posts muddies the comparison.
- Your format mix changed. Each network counts views differently from reach, so shifting from photos to video can move a rate without any change in how good the posts are.
- You're comparing different formulas or interaction sets between periods.

**Your engagement rate dropped. What changed?**
- Interactions are steady, followers or reach grew → A bigger denominator, not worse posts. Compare raw interactions too.
- The drop lines up with a network's counting change → Note the date in your report and compare like with like from then on.
- Some posts were boosted → Separate paid and organic posts before comparing.
- Interactions fell on similar posts → A real change. Test one variable at a time: hook, format, topic or time.

*Rule out the measurement causes first.*

- July 2, 2024: Instagram impressions deprecated for new media in Meta's API
- By June 15, 2026: Meta deprecates a number of Facebook Page Insights metrics
- August 24, 2026: YouTube counts a view the moment a video starts to play

*Dates from Meta's developer references and YouTube Help. Mark them in long-running reports.*

If interactions really did fall on comparable posts, look at timing and format before anything else. The best-time guides for [Instagram](/blog/best-time-to-post-on-instagram), [TikTok](/blog/best-time-to-post-on-tiktok) and [LinkedIn](/blog/best-time-to-post-on-linkedin) give starting windows to test, and [Instagram post vs Story vs Reel](/blog/instagram-post-vs-story-vs-reel) explains how the formats differ. If you post the same thing everywhere, [does cross-posting hurt engagement](/blog/does-cross-posting-hurt-engagement) looks at whether that's part of the problem.

## How do I report engagement rate consistently?

**Engagement rate reporting rules**
- [ ] Name the denominator: followers, reach, impressions or views
- [ ] List which interactions are included
- [ ] Use the same window for every post, such as 7 days after publishing
- [ ] Keep paid and organic posts separate
- [ ] Say whether a monthly figure is pooled or averaged
- [ ] Show raw interactions and the denominator next to the rate
- [ ] Note any network counting changes in the period

*Write these at the top of every report.*

The observation window is the rule people skip most often. A post measured one day after publishing will have fewer interactions than one measured after a week, so collecting numbers whenever you remember makes newer posts look worse. Pick a fixed window, such as seven days, and collect every post's numbers at that point. Where a network only keeps some data for a limited time, like X's private metrics for 30 days or LinkedIn's reach for 400 days, collect before it disappears.

Finally, remember what engagement rate can't tell you. It doesn't say whether the comments were from potential customers, whether anyone clicked through, or whether the post sold anything. If the goal is traffic, tag your links with the [UTM builder](/tools/utm-builder) and measure visits; if the goal is sign-ups, measure sign-ups. Use engagement rate to compare posts, not to judge the whole strategy.

## How do I use the engagement rate calculator?

1. **Collect the numbers**: From your analytics, note the post's likes, comments, shares and saves, and its reach, impressions or views. Note your follower count too.
2. **Open the calculator**: Go to the [engagement rate calculator](/tools/engagement-rate-calculator). It runs in your browser, so nothing you enter is sent anywhere.
3. **Enter interactions and denominators**: Fill in the interactions, then followers, reach or views, or all three.
4. **Read the rate and the formula**: The calculator shows the rate for each denominator with the formula and your numbers, so you can copy the working into a report.
5. **Record it with its formula**: Log the rate, the denominator and the raw numbers in your tracking sheet, using the same window every time.

The calculator uses only what you type. It doesn't fetch private analytics, and it can't check that two networks define reach or views the same way, which, as the tables above show, they don't.

## What should I do with my engagement rate?

Use it to run small experiments. Pick one variable, such as posting time, the first line of the caption, or video versus photo, and change only that for a few weeks while keeping your formula fixed. Compare groups of similar posts, not single posts, and look at the raw interactions too.

Consistent publishing makes those tests possible, because a gap of two weeks wrecks any comparison. [PostMesh](/pricing) schedules posts up to 30 days ahead to eight networks, and it shows the metrics each network returns for posts sent from PostMesh: reactions, comments and shares on Facebook Pages; impressions, likes, replies and reposts on X; views, likes, replies and reposts on Threads, when Threads returns them; views, likes and comments on YouTube; score and comments on Reddit. On Instagram it shows likes and comments for the posts it sends. LinkedIn post metrics aren't available yet. For everything else, including reach, use each network's own analytics as described above.

## Questions

### What is the simplest engagement rate formula?

Interactions divided by followers, multiplied by 100. It only needs public numbers, so you can calculate it for any account.

### Is engagement rate by reach or by followers better?

By reach is better for judging your own posts, because it only counts people who saw them. By followers is better for comparing accounts, since reach is private.

### Should I include saves and shares?

Yes, if your analytics show them, because they're strong signals of interest. Whatever you choose, include the same interactions in every report.

### Why does LinkedIn show a different engagement rate from mine?

LinkedIn's Page engagement rate includes clicks and divides by impressions. If you leave out clicks or divide by followers, your number will differ.

### Can engagement rate be over 100%?

By followers, yes, if a post reaches far beyond your followers. By reach it would need more interactions than viewers, which can happen when people interact several times.

### What's a good engagement rate on Instagram?

Quid's 2026 benchmark report put the median for company accounts at 0.30% by followers. Your own three-month median by the same formula is a better yardstick.

### How do I calculate engagement rate for a video?

Divide interactions by views and multiply by 100. Compare only within one network, since each defines a view differently.

### Why did my YouTube engagement rate drop in late August 2026?

From August 24, 2026, YouTube counts views the moment a video starts to play, so the same interactions can be spread over more views.

### Does the calculator save my numbers?

No. It works in your browser and doesn't send what you enter anywhere.

## Sources

- [Meta for Developers: Instagram media insights reference](https://developers.facebook.com/docs/instagram-platform/reference/instagram-media/insights)

- [Meta for Developers: Page Insights reference](https://developers.facebook.com/docs/graph-api/reference/insights)

- [Meta for Developers: Threads insights](https://developers.facebook.com/docs/threads/insights)

- [X Developer Platform: Metrics](https://docs.x.com/x-api/fundamentals/metrics)

- [LinkedIn Help: Content analytics for your LinkedIn Page](https://www.linkedin.com/help/lms/answer/a564051)

- [LinkedIn Help: Post analytics for your content](https://www.linkedin.com/help/linkedin/answer/a516971)

- [YouTube Help: How engagement metrics are counted](https://support.google.com/youtube/answer/2991785)

- [YouTube Help: Impressions and click-through rate](https://support.google.com/youtube/answer/7628154)

- [YouTube Analytics API: Metrics](https://developers.google.com/youtube/analytics/metrics)

- [Quid: 2026 Social Media Industry Benchmark Report (March 5, 2026)](https://www.quid.com/knowledge-hub/resource-library/blog/2026-social-media-industry-benchmark-report)