The four common formulas
| Formula | Calculation | Best when |
|---|---|---|
| By followers | Interactions ÷ followers × 100 | You only have public data. Most comparable across accounts. |
| By reach | Interactions ÷ people reached × 100 | You own the account and can see reach. |
| By impressions | Interactions ÷ impressions × 100 | Measuring paid or repeated exposure. |
| By views | Interactions ÷ video views × 100 | Short video, where views can far exceed followers. |
Reach and impressions are private to the account owner. For researching other creators or competitors, by followers and by views are the formulas you can actually calculate.
A worked example
A creator with 50,000 followers posts a video that gets 1,800 likes, 120 comments and 80 shares.
interactions = 1,800 + 120 + 80 = 2,000
by followers = 2,000 ÷ 50,000 × 100 = 4.0%
if the video had 200,000 views:
by views = 2,000 ÷ 200,000 × 100 = 1.0%For an account, average the rate over a sample of recent posts rather than judging one viral post.
What counts as good in 2026
Engagement varies hugely by platform and falls as accounts grow. Rival IQ’s 2026 benchmark report gives these medians per post by followers, across industries:
| Platform | Median engagement per post |
|---|---|
| TikTok | 2.01% |
| 0.30% | |
| YouTube | 0.21% |
| X | 0.03% |
Treat benchmarks as a rough guide. Compare accounts of similar size, in the same niche, on the same platform.
Pitfalls
- Mixing formulas: a 4% rate by followers and a 1% rate by views can describe the same post.
- Hidden counts: some platforms hide likes or shares, so interactions are undercounted.
- Bought engagement: sudden spikes, generic comments and a mismatch between views and comments can point to fake activity.
- Different meanings: a Reddit score is upvotes minus downvotes, not a like count.
In Monocrawl
Engagement from public data in Monocrawl
Monocrawl returns the public counts you need and, on some endpoints, computes the rate for you:
instagram/engagementreturnsengagement_rate_percent: 100 × average likes plus comments on sampled posts ÷ current followers.panorama/creator-vettingreturns a creator’s profile, a recent-post sample and observed interaction metrics on TikTok, Instagram, YouTube, X or Threads.youtube/profile/fullincludes computed engagement across recent uploads as a ratio of interactions to views.
When a platform does not publish a counter, the value stays null instead of being guessed.
Common questions
What is a good engagement rate on Instagram?
Around the 2026 cross-industry median of 0.30% per post by followers is typical for brands. Smaller creators often run far higher.
Do views count as engagement?
Usually not. A view is exposure, not interaction. Views are the denominator in the by-views formula.
Why is engagement lower on big accounts?
Followers grow faster than the share who actively interact, so rates by followers fall as accounts grow.
Sources