"Engagement rate" is used to mean at least three different numbers, and the difference between them is large enough to change a brand deal. Here is each one, when to use it, and what good looks like.
Formula 1: Engagements ÷ followers
The most common version. It is also the weakest.
(likes + comments + saves + shares) ÷ followers × 100
Use it when: a brand asks for it, which they will. Have the number ready.
Do not use it for decisions. The formula rewards having fewer followers. An account that loses 10,000 followers overnight sees its engagement rate rise, which is the definition of a broken metric.
Formula 2: Engagements ÷ reach
The formula that tells you something true.
(likes + comments + saves + shares) ÷ reach × 100
Use it when: you are deciding whether content is working.
This measures how the people who actually saw the post responded. It is comparable between a 900-follower account and a 900,000-follower account, and it predicts whether your next post gets more reach, because that is exactly what Instagram is measuring when it decides to widen the test pool.
Formula 3: Engagements ÷ impressions
Older and largely retired, because impressions counted the same viewer more than once. Some tools still report it. If someone quotes this number, ask which formula they used before comparing.
Which engagements to count
Count likes, comments, saves and shares. You can optionally weight them, because they are not equal — a send is worth more than a like — but note the weights you used if you do, because a weighted figure cannot be compared to a published benchmark.
Saves and shares are only visible to you on your own posts. On a competitor's post, you can see likes and comments and nothing else, which is why every "competitor engagement rate" tool is really measuring likes and comments only.
What counts as good
Rough bands for reach-based engagement on a well-performing post:
| Ratio | Reading |
|---|---|
| under 3% | weak — distribution likely stops |
| 3–6% | normal |
| 6–10% | strong |
| over 10% | break-out |
Adjust by niche. Highly visual consumer niches sit higher; B2B and technical accounts sit lower, because their audience is smaller and more deliberate. Compare your account to itself over time rather than to a chart.
For the follower-based formula that brands ask for, most accounts land between 1% and 5%, and anything above 5% is respectable at scale.
The three mistakes
- Reporting engagement rate for one post. A single post is noise. Use a median across the last 9–12 posts.
- Comparing across formats. Carousels earn more saves than Reels; Reels earn more reach. Compare Reels to Reels.
- Comparing across niches. A study that averages 20,000 accounts across every category describes none of them.
What to report to a brand
Lead with reach and the reach-based engagement rate, then add the follower-based figure because it will be asked for, then add the audience breakdown by country and the share of followers aged 25–34. That is a media kit that answers the real question — will this account actually move something — rather than reciting a number that can be inflated with a credit card.
Want the Numbers Without the Waiting?
Automated delivery from a panel built for it: followers, likes and Reels plays, paid anonymously in USDT, with no password and no login. Orders start within minutes.