What is a good engagement rate on TikTok?

Engagement rates fall predictably as follower count grows — TikTok is no exception. Nano-creators (under 10k followers) routinely post 8%–15% engagement on individual videos because their audience is small, tightly-themed, and self-selecting. Micro (10k–100k) settles at 5%–8%. Mid-tier (100k–1M) drops to 3%–6%. Macro and mega creators (1M+) hover around 1.5%–3.5%.

When you're vetting a creator, compare them to the right peer group — not to the platform-wide average. A 2M-follower account at 3.2% is operating in the top quartile for its tier; a 50k-follower account at the same number is below the median for its tier.

The most agency-standard formula is engagement rate per video = (likes + comments + shares) / video views. Average that across the last 10–12 videos to smooth out viral outliers. Avoid the per-follower formula (likes / followers) — it punishes large accounts where most viewers aren't followers, and it ignores comments and shares entirely.

Saves were added to TikTok's public metric surface in late 2024; some calculators now include them. KOLens uses the four-signal version (likes + comments + shares + saves) when saves are exposed in the video payload, and falls back gracefully otherwise.

TikTok's For You algorithm pushes videos to non-followers — often more than to followers. Per-follower engagement rates therefore systematically overstate true audience response on small accounts and understate it on viral ones. Per-view engagement normalises for this and matches what brands actually receive per sponsored post.

If a creator quotes a per-follower engagement rate that looks dramatically high, ask for the per-view number. The two should be within 1–2 points of each other on a healthy account; a wide gap usually means the follower count is inflated or stale.

An engagement rate substantially below tier benchmark (e.g. 1% on a 100k account) is a warning sign — not a deal-breaker, but worth investigating. Common causes: a sharp content pivot that lost the original audience, a wave of purchased followers from a growth-hack agency, repeated low-effort reposts, or a niche audience that simply doesn't interact (luxury / B2B / news accounts can post 1.5% and still convert).

Pair the engagement number with comment-quality scan (real reactions vs. emoji-only spam), an audience country split, and a posting cadence. KOLens surfaces all three alongside the rate so you don't have to evaluate the number in isolation.