TikTok Engagement Rate: Formula, Benchmarks and How to Use It

When you do content research or competitor analysis, one of the most common mistakes is looking only at view counts. A video with several hundred thousand views looks impressive, but if likes, comments and shares are barely there, it may simply have been pushed to a lot of people without really moving any of them. Engagement rate is the metric that fills that gap. This guide explains how it is calculated, how the different definitions differ, roughly what public benchmark data looks like, and, most importantly, how to use it to reach reliable conclusions rather than fixating on a single number.

The short version

Engagement rate measures how many of the people who saw a video did something about it.
There are three common formulas. The denominators differ, so the numbers can't be compared directly. Always compare using the same definition.
There is no universal "good" threshold. What works is comparing against videos in the same niche, same period and similar size.

Three common formulas

1. By views (most common, best for analyzing a single video)

Engagement rate = (likes + comments + shares + saves) ÷ views × 100%

This answers: "Of the people who watched, what share interacted?" It doesn't depend on follower count, which makes it the best choice for comparing videos across different accounts and the most practical one for competitor analysis.

2. By followers (often used to evaluate an account)

Engagement rate = (likes + comments + shares + saves) ÷ followers × 100%

You often see this in influencer vetting. The catch is that a large part of any video's audience isn't followers (TikTok's recommendation system serves videos to non-followers), so comparability between big and small accounts, or between hits and ordinary videos, is weaker.

3. By reach

Engagement rate = total interactions ÷ people reached × 100%

Reach is only visible in an account's own analytics. You can't get it for someone else's video, so it won't help in competitor analysis, but it's good to know it exists.

Our Engagement Rate Checker uses a simplified version of the first formula: (likes + comments + shares) ÷ views. Save counts aren't available from public data, so the result runs slightly lower than definitions that include saves. It is sufficient for side-by-side comparison.

A worked example

Say a video has 120,000 views, 4,800 likes, 150 comments and 420 shares.

Total interactions = 4,800 + 150 + 420 = 5,370
Engagement rate = 5,370 ÷ 120,000 × 100% ≈ 4.48%

Looking at the mix: likes make up about 89%, comments about 3% and shares about 8%. That composition says more than the total, as covered below.

Where industry benchmarks sit

A useful public reference is Socialinsider's TikTok benchmark study. It calculates engagement by views, counting likes, comments, shares and saves, across roughly 2 million videos from more than 210,000 profiles, covering 2024 through 2025. According to its published figures:

Average engagement rate was about 3.85% for 2024 and about 4.20% for 2025.
By follower tier, the 2025 averages fall roughly between 3.75% and 4.4%, with smaller accounts slightly higher, though the gap is modest.

Two cautions. First, these are cross-industry averages, and individual niches can differ a lot. Second, they come from a particular sample and definition, so don't treat 4% as a pass mark. Their real value is giving you an order of magnitude: if a video works out to 0.3% or 15%, that's worth a closer look.

Why you shouldn't fixate on a number

Several factors explain why comparison beats fixed thresholds:

Video type: tutorials, reviews, skits and comedy invite different kinds of interaction. Tutorials tend to get more shares and saves, skits more comments.
Niche differences: some audiences quietly watch to the end, others love debating in the comments.
Distribution stage: early on, views are low and the rate tends to be high. Once a video is pushed to a broad audience, views surge and the rate is often diluted. The same video can show different numbers at different times.
Watch time and prompts: whether the ending invites a reaction, and whether the video holds back the payoff until the end, both affect the number.

How to use it: same-niche comparison

Here is a simple process for competitor analysis, topic research and product research.

Step 1: Define the niche and sample. Choose keywords or a category and collect videos published in the last 2 to 3 months, at least 15 to 20 per niche. Too few and chance skews the picture.

Step 2: Calculate each video with the same definition. Paste each link into the Engagement Rate Checker and record plays, likes, comments, shares and the rate in a sheet, plus the posting date and the account's follower tier.

Step 3: Look at the distribution, not the average. Sort by engagement rate and find the videos clearly above the group's median. Those are worth breaking down. Also pull out videos with very high views but low engagement, which often owe their numbers to recommendation traffic rather than the content itself.

Step 4: Look at the interaction mix. The same 5% can hide very different compositions:

Pattern	What it often suggests (rule-of-thumb)
Higher share of shares	Useful or highly relatable content people want to pass along
Higher share of comments	A debatable point, a question, or a clear prompt to respond
Many likes, few comments or shares	Pleasant to watch, but not enough pull to take another step

This table summarizes common patterns, not laws. Use it to form a hypothesis, then verify it by watching the video.

Step 5: Break down the top performers. Once you have the best few, analyze them by hook, pacing, copy and music. Our earlier article on breaking down viral videos and building a swipe file covers that part.

Using it for product research and export work

If you run overseas short-video marketing for exporters or factories, engagement rate has another practical use: checking whether a product direction has real audience interest.

If a dozen videos in a category show generally low engagement, even when a few have huge views, be careful. They may have been lifted by paid promotion or trend-riding, which isn't the same as demand for the product.
If several videos from different accounts all show clearly higher engagement, and viewers in the comments ask about price and where to buy, that's a far more reliable signal than any single hit.
Read the numbers together with the comments: the rate shows how many people reacted, and the comments show what they said.

Limits to keep in mind

No save counts: they aren't available from public data, so figures run slightly lower than save-inclusive definitions. Fine for side-by-side comparison, but don't compare directly against numbers in industry reports.
Snapshots in time: if a video is still growing, the numbers will change. To track a trend, note the date and parse again after a few days.
Don't conclude from one or two videos: sample size matters more than calculating to many decimal places.

FAQ

Is a higher engagement rate always better? Usually, but read it alongside views. A video with a few hundred views can reach a very high rate from just a handful of likes, which limits its value as evidence. A sound approach is to check that the view level is meaningful first, then look at the rate.

Does a low rate mean the video is bad? Not necessarily. It may have been pushed to a very broad audience, or the content type may simply have a high bar for interaction. Compare against similar videos.

Can I freely analyze other people's video data? Analyzing publicly visible data for study and research is standard practice. Reposting or reusing someone else's video, however, raises copyright issues and needs permission first.

Why doesn't my result match an industry report? Most likely the definitions differ: whether saves are included, whether the denominator is views or followers, and whether the time window matches can each shift the number noticeably.

The simplest first step: pick a niche you care about, collect around 15 videos, and run each through the Engagement Rate Checker. You'll quickly see which videos are genuinely popular and which were just pushed hard.