What is a good engagement rate on TikTok, Instagram and YouTube?

Add up the interactions on a post, then divide by one of two things:

• By followers: (likes + comments + shares + saves) ÷ followers × 100. It says how much of an account's audience reacts to what it posts.
• By views: (likes + comments + shares + saves) ÷ views × 100. It says how much of the people who actually watched reacted, whoever they were.

Take a post with 1,200 likes, 80 comments, 40 shares and 60 saves: 1,380 interactions. On an account with 50,000 followers, where the post got 40,000 views:

For an account over a month, add up the interactions on all its posts and divide by the number of posts, then by followers: that's its average engagement rate per post. Or put the numbers into the engagement rate calculator, which works out both.

Engagement by followers made sense when a post was mostly seen by the people following the account. On TikTok, Reels and Shorts, most views come from feeds that show videos to people who don't follow you, so a post can reach far more or far fewer people than you have followers.

That breaks the rate by followers for anything that takes off. Say a clip on an account with 2,000 followers gets 300,000 views and 9,000 interactions:

• By followers it's 450.00%, a number that says nothing useful.
• By views it's 3.00%, which you can compare with your other clips and with what's typical.

So use each where it fits:

• Views for clips, for comparing your own posts, and for clipping campaigns, which pay per view anyway. It tells you whether a video made the people who saw it act.
• Followers when a brand asks about your account, or to compare accounts of a similar size. It's still the number most benchmark studies and many brands use.

These are the largest recent studies. They're third-party research, mostly on brands' accounts, and each counts engagement its own way, so read the columns before comparing:

A median is the middle account: half did better and half worse. An average can be pulled up by a few accounts with very high rates, so averages usually run higher than medians. The studies also look at different accounts, so their numbers don't match exactly even when they count the same things.

So a rate above the typical one for the same app, measured the same way, is good. By followers, that means over about 2.01% on TikTok and 0.30% on Instagram; by views, over about 4.20% on TikTok and 2.59% on YouTube Shorts.

TikTok has the highest rates of the three by followers. Quid's median was 2.01% and Socialinsider's average 2.60%. By views, Socialinsider's average was 4.20% in 2025; it held between 4.20% and 4.30% from early 2025 and fell to 3.85% in the second quarter of 2026.

Instagram rates by followers are much lower: a median of 0.30% in Quid's study and an average of 0.48% in Socialinsider's, which counts only likes and comments. Socialinsider found reels did better than single images: 0.52% against 0.37% in 2025. None of the large studies gives Instagram a rate by views from its own data, so compare your reels with each other by views.

YouTube is usually measured by views, with likes and comments. Across 1.46 million channels, CreatorDB found a median of 2.59% for Shorts and 3.67% for long videos. The long-video figure was measured under YouTube's old count, when a view was generally taken to mean about 30 seconds of watching (YouTube never published the rule). Since August 24, 2026, YouTube counts a view the moment a video starts to play, on every format, so the same videos show more views and a lower rate than they did before.

Two of the studies split their results by account size, measured by views:

On TikTok and on long YouTube videos, rates mostly fall as accounts grow, most sharply for the biggest YouTube channels. Shorts didn't follow that pattern: their rate was much the same at every size. So compare your rate with accounts your size, and don't worry if it dips a little as you grow. A small account with a high rate isn't automatically doing better than a big one with a lower rate: the big one is reaching far more people.

Each app shows the numbers you need, but not all in the same place:

• TikTok shows likes, comments, shares and saves under every video, and more in TikTok Studio's analytics.
• Instagram shows likes and comments on the post, unless the poster hides likes. A professional account's insights show shares and saves too; Meta counts a reel's total interactions as its likes, saves, comments and shares.
• YouTube shows likes and comments publicly, and shares in YouTube Studio. Most YouTube benchmarks count only likes and comments.

People like, comment and share when a video gives them a reason to. The things that do that most:

• A first second that makes people stay. Nobody reacts to a video they swiped past. Open on the line that matters, not a greeting. The hook generator gives you opening lines to start from, and how to make viral clips covers the rest.
• Captions. Many people watch with the sound off; captions keep them following until the moment that makes them react. See how to caption short videos.
• One clear moment. A clip that lands one point, joke or reveal gets shared; one that wanders doesn't. End on the payoff.
• Something to answer. A take people disagree with, or a question in the caption, gives them something to comment.
• Posting when your audience is around. The first viewers' reactions decide how far a video goes. See the best time to post.

Then measure by views, a few days after posting, and compare each clip with your own recent ones. Keep what lifts the rate and change one thing at a time.