The TikTok engagement rate benchmark you can actually defend.

The quick answer

Searches for a TikTok engagement rate benchmark are usually looking for one number: a percentage to compare an account against so the result can be called good or bad. That number does not exist in a form worth trusting, and the pages that publish one are almost always describing a population that has nothing to do with the account you care about. A benchmark drawn from a curated creator roster or a self-selected sample of agency clients is a real average of a real group, and still the wrong yardstick for a handle picked at random.

The useful benchmark is one you build. Choose five to ten accounts that genuinely resemble yours, measure all of them the same way on the same day, and compare inside that set. The checker at the top of this page returns a real engagement rate for any public handle in seconds and costs nothing, which makes a peer set a twenty-minute job rather than a research project. That is the whole argument of this page, and the rest of it is the evidence and the method.

Why the spread makes a single benchmark useless

This is not a stylistic objection. Our own live short-video pulls across ten public brand accounts say the benchmark does not exist. Across those ten windows, likes as a share of views ranged from 5.15% to 27.10%, a better-than-fivefold spread. Comments as a share of views ranged from 0.026% to 0.292%, an elevenfold spread. Any single good engagement rate number would misjudge most of the accounts in that set.

Read those two ranges together, because they behave differently. Likes are cheap and near-automatic, so the like share stays inside one order of magnitude even across very different brands. Comments cost the viewer real effort, so the comment share swings wider and separates accounts that provoke a response from accounts that are merely watched. An account near the top of the like range and near the bottom of the comment range is not the same business as one placed the other way around, and a single blended rate flattens that distinction into one figure that describes neither.

TikTok also makes the denominator question harder than other platforms do. Distribution runs through the For You feed rather than the follower graph, so a video reaches people who do not follow the account, and views on two competitors of different sizes are not directly comparable at all. A rate measured per follower and a rate measured per view answer different questions, and mixing published figures that quietly use different denominators is how two sources disagree by an order of magnitude while both stay arithmetically correct.

What the free check returns

The check on this page reads the profile fields TikTok publishes to logged-out visitors: the follower count, the following count, the public video count, and the verification badge. Alongside those it derives an engagement snapshot from the videos in the sampled window: average public likes, average public comments, and an engagement rate, calculated as average likes plus comments on recent videos divided by followers.

Three fields rather than one is a deliberate choice. The rate is the headline, but the averages behind it are what make it readable. Two accounts can land on the same rate with completely different mixes, one carried by likes on a video that travelled and one carried by a comment section that argues with every post, and only the second is a community you could brief a campaign around. Keep all three columns when you record a peer set.

How to build your own peer-set benchmark

This is the method the rest of the page is arguing for. It takes about twenty minutes for a set of eight accounts and it fits inside the free daily allowance.

1. Choose five to ten close peers. Pick public accounts that match yours on size band, niche, and posting style. Closeness matters more than fame, and an account ten times your size belongs in a different set. If you cannot name five, widen the niche before you widen the size band.
2. Check each handle with the free checker. Run every account through the checker on this page in one sitting, so each rate is drawn from a comparable moment rather than months apart. A set assembled over six weeks measures the calendar as much as the accounts.
3. Record the rate, average likes, and average comments. Write down all three fields per account, plus the follower count, because the rate on its own hides whether the response came from likes or from conversation. A spreadsheet with five columns is enough.
4. Compare inside the set, never against a published average. Read your account against the middle of your own peer set and against its own history. That middle is your benchmark, and it is the only one with a definition you can show a client. Note the spread in your set too, because a wide one is itself the finding.
5. Re-check on a cadence. Repeat the pass monthly or quarterly and keep the same accounts in the set. The direction of travel across repeated pulls is the finding, not any single number, and a set that has drifted down together is telling you something about the category rather than about your content.

Two habits keep the set honest. Never swap peers between passes just because the comparison got uncomfortable, since the value of the set is that it holds still while your account moves. And write the selection rule down next to the numbers, so the next person to open the sheet can see why those accounts and not others. A documented benchmark can be argued with, which is what makes it more useful than a percentage lifted from a blog post.

What the rate cannot tell you

Everything on this page works from what TikTok shows a logged-out visitor: the profile's follower count, following count, public video count, and verification badge, plus the public videos themselves with their captions, dates, views, likes, and comments. Watch time, completion rate, traffic sources, audience demographics, follower growth history, follower lists, and anything inside a competitor's TikTok Analytics are private. No external product can retrieve them, whatever its pricing page implies.

Two consequences follow. Private accounts and follower-only videos expose nothing at all, so no amount of software reaches them. And fake-follower audits are not possible from the outside either: TikTok does not publish a profile's follower list to logged-out visitors, so nothing external can inspect individual followers, and engagement rate is not a substitute because the variation between legitimate accounts of different size, niche, and format mix is far larger than the effect that buying followers has on the number. A low rate is not evidence of purchased followers, and a high one is not proof of a clean audience.

InstaSeer is independent and not affiliated with TikTok. A rate reflects the public videos visible in the loaded window at the time of the pull, not a fixed archive guarantee, and checking a public profile is not an interaction: TikTok does not notify an account when someone views its public profile, and InstaSeer never follows, likes, comments, or messages the account it reads. The methodology page documents how each number is read, and the TikTok follower count checker lists the public profile fields in full.