Engagement Rate Grader: A Verdict With Its Source Attached

What a verdict can honestly say

Ask most engagement tools whether 2.4% is good and you get a one word answer. Good, average or poor. Those answers require thresholds, and thresholds have to come from somewhere. When you look for the source of the thresholds in common use, there usually is not one.

This tool is built the other way round. It starts from what has actually been published, which is a small set of platform-level figures, and reports your rate's position against them. That is less satisfying than a grade and it is defensible, which a grade is not.

Position is a real verdict

Saying that a rate is above or below a specific published figure, and by what multiple, is a genuine finding. If you tell me an Instagram account runs at 1.2% on followers, I can tell you that is four times Quid's all-industry median of 0.30% from its report published 5 March 2026 and roughly two and a half times Socialinsider's 2025 average of 0.48% from its report published 16 January 2026. Both of those statements are checkable.

What I cannot tell you is whether 1.2% is good, because good depends on the niche, the audience, the content format and what you want from the partnership, and because no published source defines a cutoff where acceptable becomes strong. The grader gives you the part that is knowable and stops.

Why the multiple matters more than the direction

Above the median sounds impressive until you notice that published medians are low. Quid's Instagram median is 0.30%, so more than half the accounts in its sample sit below that. A rate of 0.31% is above the median and nearly identical to it. A rate of 3% is ten times the median and a different proposition entirely.

That is why every comparison here carries the multiple. Direction alone flattens the difference between barely ahead and far ahead, and that difference is usually what you are trying to establish.

Source: Quid 2026 Social Media Industry Benchmark Report, published 5 March 2026 · Socialinsider Social Media Benchmarks, published 16 January 2026

What this tool will and will not grade

This is the complete coverage map. Six platforms times four denominators gives 24 combinations. Six of them have a published figure. The other 18 return no verdict, and the table says which is which before you enter anything.

Reading the refusals

The pattern in that table is not random. Every graded row divides by followers, except LinkedIn, which divides by impressions. That is because follower counts are public and reach, impressions and views are not, so the publishers who build benchmarks at scale build them on the number they can see from outside. There is no reach-denominated benchmark for any platform in this set.

LinkedIn is the mirror image. Socialinsider publishes an impressions-based LinkedIn average and no follower-based one, so a LinkedIn rate divided by followers cannot be graded here while a LinkedIn rate divided by impressions can. If you have a follower-based LinkedIn figure, the honest options are to recompute it against impressions or to treat it as a number with no outside reference.

YouTube deserves a specific warning. The YouTube row that grades is the follower row, and Quid's YouTube median of 0.21% divides by subscribers. Most YouTube engagement rates in circulation divide by views instead, which produces a much larger number. Grading a view-based YouTube rate against a subscriber-based median would make almost any channel look exceptional, so this tool does not do it.

Source: Quid 2026 Social Media Industry Benchmark Report, published 5 March 2026 · Socialinsider Social Media Benchmarks, published 16 January 2026 · Socialinsider LinkedIn Benchmarks, published 16 March 2026

Why there is no grade by follower tier

The most common request for a tool like this is a verdict adjusted for account size. A nano creator and a mega creator should not be held to the same number, and everybody who works with creators knows it. Engagement falls as follower count rises, consistently, across every source that looks at it.

The problem is that knowing the direction is not the same as having the numbers. To grade by tier you need published bands with a stated sample size, a stated collection window and one consistent denominator. No source read for this page provides all of that.

The only named primary with tier bands

HypeAuditor publishes bands by tier on a page created 30 April 2025 and last updated 7 August 2026. They are reproduced here because they are the only tier figures from a named publisher that this page found, and they come with three problems that rule them out as a grading basis.

The first problem is that the publisher calls them rough benchmarks itself. The second is that no sample size and no collection window are stated, so there is no way to judge how much data sits behind them. The third is the serious one. HypeAuditor's stated primary formula divides likes and comments by views, with a fallback to followers for accounts without enough Reels, which means a single band can contain accounts measured two different ways. A band that mixes denominators cannot be used to grade a rate whose denominator you have just carefully established.

There is a further tell. The same page gives a worked example of 50,000 followers, 1,500 likes and 100 comments producing 3.2%, which is 1,600 divided by 50,000. That example is follower-based, while the page names views as the primary denominator. When a publisher's own example does not follow its own formula, its bands are not a foundation to build a verdict on.

Also note the tier boundaries. HypeAuditor's nano band runs from 1,000 to 5,000 followers and its micro band from 5,000 to 50,000, which does not match the 1,000 to 10,000 and 10,000 to 100,000 convention used in most rate cards. Macro and mega are combined into one band, so there is no separate mega figure at all. Even if the denominator were consistent, the tiers would not line up with the tiers you are probably using.

Source: HypeAuditor Instagram engagement rate benchmarks, created 30 April 2025, last updated 7 August 2026

Median against average, and why your verdict depends on it

Two of the figures this tool grades against measure Instagram in 2026. Quid gives 0.30% and Socialinsider gives 0.48%. Both divide by followers. Neither is wrong. One is a median and the other is an average, and on engagement data that distinction is large.

What skew does to the two statistics

Engagement is distributed unevenly. Most posts take a modest number of interactions and a small number of posts take a great many. When a distribution has a long tail like that, the average is pulled upward by the tail while the median stays near the middle of the pack. An average above a median is the signature of exactly that shape.

For grading, the practical effect is that the median is the harder test of being typical and the average is the harder test of being exceptional. A rate of 0.40% on Instagram is above Quid's median and below Socialinsider's average, which is why the tool reports a split verdict for rates in that window rather than picking a side.

Why both are shown

It would be simpler to average 0.30% and 0.48% into 0.39% and grade against that. It would also be wrong. The result would be a figure neither publisher reported, built from a median and an average of different samples over different windows, and you would not be able to check it anywhere.

The same reasoning applies to TikTok, where Quid's median is 2.01% and Socialinsider's 2025 average is 3.70%. That is a wide gap and it is reported as a gap. A rate of 2.5% on TikTok is above one published figure and below the other, and the useful thing to know is precisely that.

One discrepancy inside a single source

Socialinsider's own page gives its TikTok 2025 figure as 3.70% in a table and 3.73% in the body text. The table value is used here. The difference is too small to change any verdict, and it is reported because a source that disagrees with itself by a small amount in public may disagree with itself by a larger amount elsewhere.

Source: Quid 2026 Social Media Industry Benchmark Report, published 5 March 2026 · Socialinsider Social Media Benchmarks, published 16 January 2026

What to do with the verdict you get

A position against a published figure is a starting point for a decision rather than the decision. Here is what each outcome actually supports.

Far above every published figure

A rate many times the published median on a follower denominator is worth investigating in both directions. It can mean a small, genuinely engaged audience, which is the best thing you can find. It can also mean engagement that was bought, which inflates the numerator while the follower count stays put, or a follower count that collapsed while old interactions remained.

The check is whether the interaction pattern is plausible. A high rate with comments that say nothing, arrive in bursts or come from accounts with no history is a different signal from a high rate with a conversation under the post. Rate alone cannot separate those two, which is the main limit of any tool on this page.

Slightly above or slightly below

Most accounts land here, and the correct reading is that the rate is unremarkable. Published medians are low, so being near one is the normal case rather than a warning. Do not build a case for or against a creator on a rate within a factor of two of the published middle. Look at the content, the audience composition and whether previous brand posts performed like the organic ones.

Far below every published figure

A rate well under the published median on a large follower count is the classic padded-audience signal, and it is the one case where a rate on its own justifies real caution. The denominator is doing the work. If a large share of the followers are inactive or bought, they sit in the denominator and never interact, which drags the rate down.

It is not conclusive. Accounts with long histories accumulate followers who have drifted away without anything untoward happening, and accounts whose content shifted into a new niche often carry an audience that no longer responds. Both produce a low rate honestly.

Not graded

If the tool returns no verdict, the rate is not bad and the tool is not broken. It means no publisher in this set has measured that platform on that denominator. You have two sensible options. Recompute the rate on followers, where benchmarks exist, or keep the figure and compare it against your own history for the same account and the same denominator, which is often the more informative comparison anyway.

Source: Quid 2026 Social Media Industry Benchmark Report, published 5 March 2026 · Socialinsider Social Media Benchmarks, published 16 January 2026

What this grade cannot tell you

Being clear about the limits is part of making the verdict usable. These are the things a position against a published platform figure does not establish.

• Whether the engagement is real. A rate counts interactions. It cannot tell whether they came from people who might buy something. That needs audience data rather than arithmetic.
• Whether the rate suits your niche. No source in this set publishes engagement rate by niche or category as numbers, so the comparison is always against an all-industry figure.
• Whether the account size is being accounted for. It is not. No usable tier benchmark exists, so a nano creator and a mega creator are graded against the same platform figure. Given that engagement reliably falls with size, this understates small accounts and flatters large ones.
• Whether the creator will deliver for you. Published benchmarks are built mostly from brand and company accounts rather than creator accounts, so a creator graded against them is being compared with a different population.
• Whether the figure you entered is accurate. The grade is only as good as the rate. If the number came from a media kit with no stated formula, grade it and treat the result as provisional.

The population problem, stated plainly

This is the limit most worth understanding. Quid's sample is 150 companies drawn at random from each of 18 industries, from a database of over 200,000 companies, covering organic and paid posts. Socialinsider's is 70 million posts from a representative sample of international brands between January 2024 and December 2025. Both are built on brands.

Individual creators are not brands. They post more often, their audiences are usually there for them personally rather than for a product and they typically see higher engagement than a company page of the same size. Grading a creator against a brand benchmark is the best available comparison, and it is not a like-for-like one. Treat a creator sitting above these figures as expected rather than exceptional.

Source: Quid 2026 Social Media Industry Benchmark Report, published 5 March 2026 · Socialinsider Social Media Benchmarks, published 16 January 2026 · HypeAuditor Instagram engagement rate benchmarks, created 30 April 2025, last updated 7 August 2026

Check the rate before you grade it

A verdict inherits every flaw in the number you feed it. Four questions settle whether a rate is worth grading at all, and they are worth asking of any figure that arrives in a media kit.

• What was it divided by? If the answer is not available, the rate cannot be graded against anything. This is the question that gets skipped and it is the one that matters most.
• Which interactions were counted? Likes and comments produce a lower rate than likes, comments, shares and saves on the same post. Socialinsider counts the first pair on Instagram and all four on TikTok, so even one publisher is not internally consistent across platforms.
• How many posts? A rate from a single post says little. A rate averaged across recent posts is more stable, and a rate built from a best post is a sales figure rather than a measurement.
• When? Every source in this set reports engagement declining on at least one platform year over year. Quid records Instagram sliding by roughly a sixth in a single year, and Socialinsider records a smaller slip on the same platform over a comparable period. A rate measured two years ago is not a current rate, and nothing on a media kit says when it was taken.

A worked check

Suppose a media kit claims 7% engagement on Instagram with 40,000 followers. On a follower denominator that implies about 2,800 interactions per post, which is a large number for an account that size and would put it more than twenty times above Quid's published median. It is possible. It is also what a rate computed on reach rather than followers looks like.

Ask which denominator was used. If it was reach, recompute on followers and the figure will fall sharply, probably into the range where it can be graded meaningfully. Nobody is necessarily being dishonest here. Most analytics screens show a reach-based rate without labelling it, and the number gets copied into the kit as the engagement rate because that is what the screen called it.

Source: Socialinsider Social Media Benchmarks, published 16 January 2026

Related free tools

• Engagement Rate Calculator
• Engagement Rate Benchmarks
• Instagram Engagement Rate Calculator
• TikTok Engagement Rate Calculator
• Instagram Fake Follower Checker
• Instagram Quality Score Calculator
• All free tools