Average Engagement Rate on Instagram (12,505 Reels, 2026)

Accounts under 10,000 followers earn a median engagement rate of 5.2% on their Reels. Accounts over a million followers earn 0.21%. Both numbers come from the same set of 12,505 public Reels, and the 24x gap between them has almost nothing to do with whose content is better. Measured per view, those Reels got liked at nearly the same rate: 2.1% for the small accounts, 2.4% for the giants.

That's the catch with asking for the average engagement rate on Instagram. The formula divides likes and comments by followers, but a Reel's audience isn't limited to your followers, so the result mostly measures how far the Reel travelled. Below are the real medians by account size, where "good" starts at each size, and the per-viewer numbers that tell you whether a Reel landed.

One of those per-viewer numbers comes out almost three times too high if you divide by the wrong view count, and it's an easy mistake to make. I'll show you which count to use.

TL;DR#

The average engagement rate on Instagram depends on account size: across 12,505 public Reels posted in 2026, the median account earned 5.2% under 10K followers, 0.91% at 10K–50K, 0.82% at 50K–250K and 0.21–0.27% above 250K (likes plus comments, divided by followers). A good engagement rate beats your tier's median, and the top quarter starts at about 18.5% under 10K, 1.8% from 10K to 250K and 0.73–0.78% above that. Per viewer, size barely matters: the median Reel's like-to-view ratio is 2.35%, the top quarter starts at 3.84%, and every tier's median sits between 2.1% and 2.6%. What separates the tiers is reach, which is why how far your Reels travel past your follower count says more about your account than the rate does.

What's the average engagement rate on Instagram by follower count?#

Engagement rate by followers is likes plus comments, divided by followers, times 100. It's the formula most engagement calculators start with, so it's the first thing I measured. Each account below gets one number, the median across its own Reels, and only accounts with at least five qualifying Reels count. That leaves 264 accounts, and it means the busiest account (269 Reels in the set) counts exactly as much as one with five.

Followers	Accounts	Median engagement rate	Top quarter starts at
Under 10K	54	5.16%	18.5%
10K–50K	44	0.91%	1.75%
50K–250K	78	0.82%	1.79%
250K–1M	47	0.27%	0.73%
1M+	41	0.21%	0.78%
All accounts	264	0.84%	2.34%

The rate is roughly flat from 10K to 250K followers, then drops by about two thirds past 250K. The big accounts also land right next to the published industry figure: Rival IQ's 2026 benchmark report puts the median Instagram post at 0.30% by followers, down from 0.36% a year earlier.

Then there's the first row. A median of 5.2%, with the top quarter starting at 18.5%, looks like a different platform. It's the same platform, measured with a denominator that stops making sense once Reels reach people who don't follow you.

The reach illusion: why small accounts win on engagement rate#

The reach illusion is when an engagement rate by followers rises or falls because of how far a Reel travelled, not because of how much its viewers liked it. You can see it by splitting the rate into two parts: views per follower (how far the Reel went) and engagement per view (how much the people who saw it cared).

Followers	Median views, as % of followers	Median engagement per view	Median like-to-view
Under 10K	227%	2.40%	2.13%
10K–50K	34%	2.69%	2.51%
50K–250K	30%	3.00%	2.64%
250K–1M	12%	2.75%	2.29%
1M+	8%	2.50%	2.43%

Reels from accounts under 10K pulled a median of 2.3 times their follower count in views. Reels from million-follower accounts pulled 8%. Engagement per view, meanwhile, stayed between 2.4% and 3.0% in every tier. Multiply the two columns and you land within a few tenths of a point of each tier's engagement rate: 2.69 interactions per 100 views, on views worth 33.6% of your followers, gives 0.90%, against a real median of 0.91% for the 10K–50K tier.

The extreme end makes it obvious: 100 of the 1,573 Reels from accounts under 10K (6.4%) collected more likes and comments than the account had followers, which is an engagement rate above 100%. Of the 2,210 Reels from million-follower accounts, exactly one did.

So stop comparing your engagement rate with anyone else's. A 3,000-follower account at 9% and a 400,000-follower account at 0.4% can be making Reels that viewers like equally, and the rate can't tell them apart. It also means your rate should drop as you grow, even while your Reels get better.

That isn't a penalty. It's arithmetic.

What counts as a good engagement rate on Instagram?#

A good engagement rate on Instagram is one that beats the median for your follower tier in the first table, and top-quarter territory is great. I wouldn't judge a single Reel that way, though. A tenth of the Reels from sub-10K accounts in this set went past 54% on their own, so one Reel's rate is mostly a reading of how far it spread.

Instead, run what I call the per-viewer check: divide likes and comments by the Reel's views instead of your followers. It answers the question the follower rate can't, which is whether the people who saw the Reel cared. These are the per-viewer benchmarks across all 12,505 Reels:

Per-viewer number	How to work it out	Median Reel	Top quarter starts at	Top 10% starts at
Engagement by views	(likes + comments) ÷ views × 100	2.69%	4.42%	6.42%
Like-to-view ratio	likes ÷ views × 100	2.35%	3.84%	5.84%
Comments per 1,000 views	comments ÷ views × 1,000	0.77	2.77	10.6

To run the check on your own account:

1. Take your last 12 Reels that are at least a week old. Younger ones are still collecting likes.
2. For each one, note the views shown on the Reel, the likes and the comments, plus your current follower count.
3. Put each Reel through the free engagement rate calculator, which returns the by-followers rate, the by-views rate and the like-to-view ratio in one pass.
4. Take the median of each number across the 12, not the average.
5. Compare your by-followers median with your tier in the first table, and your per-view medians with the table above.

If your per-view numbers beat the median but your follower rate trails your tier, the content is fine and the reach isn't. That's an opener problem: the algorithm shows each new Reel to a test audience of 100 to 500 viewers and caps reach when roughly 60% of them swipe away within 1.5 seconds, so the opening patterns that pull strangers into a Reel are where I'd start.

Bottom-quarter per-view numbers are a different story. Reach isn't your problem yet: the Reel is finding people and they aren't reacting, and retention benchmarks for hook and hold rate will show you where they bail.

Of the three per-viewer numbers, the like-to-view ratio gets searched the most. It's also the one that goes wrong if you grab the wrong view count.

What's a good like-to-view ratio on Instagram Reels?#

Like-to-view ratio is likes divided by views, times 100: 24 likes on 1,000 views is 2.4%. The average like-to-view ratio on Instagram Reels in this set (the median, to be exact) is 2.35%, and it moves a little with how far a Reel went:

Views on the Reel	Reels	Bottom quarter	Median	Top quarter	Top 10%
1K–10K	3,505	1.31%	2.21%	3.63%	5.24%
10K–100K	5,339	1.32%	2.26%	3.54%	5.31%
100K–1M	2,874	1.30%	2.56%	4.41%	6.94%
1M+	787	2.13%	3.41%	5.47%	7.34%
All Reels	12,505	1.35%	2.35%	3.84%	5.84%

So a good like-to-view ratio on Instagram Reels is anything above about 2.4%, and 3.8% or more puts a Reel in the top quarter. Below about 1.3% is the bottom quarter at every size up to a million views, which usually means the hook pulled a crowd that wasn't the Reel's real audience. The 1M+ row runs higher, with a 3.41% median. Reels that got liked more may have kept getting shown, or the ones shown to more people may have found their fans; this data can't separate the two, so don't treat likes as the cause of reach.

Now the view count I promised. Instagram's data carries two view counts per Reel: the one shown on the Reel, and an older counter that some scrapers and third-party tools still pull. On these 12,505 Reels, the older counter is a median 37% of the visible one. Divide the same likes by it and the median like-to-view ratio jumps from 2.35% to 6.82%, nearly three times higher, for Reels that didn't change at all.

If a like-to-view benchmark you've seen puts the typical Reel at 6% or 7%, that's a likely reason. Use the number shown on the Reel. It's the one your viewers see, and it's the one every figure here uses.

Comments are the easiest number to inflate#

Comments per 1,000 views has a median of 0.77, and it shrinks as Reels get bigger: 1.14 for Reels with 1K–10K views, 0.33 past a million. Big audiences like and keep scrolling. Very few stop to type.

The top of the range is where it gets strange. The top 10% of Reels pulled 10.6 or more comments per 1,000 views, and when I checked their captions, 39% asked for a keyword comment ("Comment GUIDE and I'll send it to you"). Across all 12,505 Reels that share was 11.2%, and in the bottom tenth it was 1.8%.

Those prompted Reels didn't earn more likes per view, either. The top tenth by comments had a median like-to-view ratio of 2.30%, next to 2.35% for everything. So a comment-heavy Reel tells you the prompt worked, not that the content did. Count prompted comments as a DM-list number, and judge the Reel itself on likes per view.

How I measured this (and what it can't tell you)#

Everything here comes from public Instagram Reels on accounts that CreatorHouse users track, pulled from our database on September 29, 2026.

• 12,505 Reels from 321 accounts, posted between January 1 and September 16, 2026, and last scraped between February 1 and September 29.
• One row per Reel, from its most recent scrape, kept only if the Reel was at least seven days old at that scrape, had 1,000+ views (the count shown on the Reel) and showed its likes. About 3.1% of Reels hide their like count, and those are out.
• Follower counts come from our profile lookups, and a Reel only counts if that lookup happened within 60 days of the Reel's numbers.
• Every figure is a median or a percentile. Averages would let a handful of viral Reels set the benchmark.
• Engagement means likes plus comments. Saves and shares aren't public, so if you add yours from Insights, expect a higher number than these.

The caveat that matters most: people track accounts they want to learn from, so this set leans toward accounts that are already working. Read the medians as "healthy", not as a floor. And none of it shows cause and effect: a big view count and a high like-to-view ratio travel together, but the data can't say which pulled the other.

Frequently asked questions#

What is the average engagement rate on Instagram in 2026?#

By followers, the median account in our 2026 Reels data earned 0.84%, while Rival IQ's 2026 industry report puts the median brand post at 0.30%. Account size moves it more than anything: about 5.2% under 10K followers, 0.8% to 0.9% from 10K to 250K, and 0.21% to 0.27% above that.

Is 3% a good engagement rate on Instagram?#

Above 10,000 followers, yes: 3% by followers clears the top quarter of every tier from 10K up. Under 10K it's below the 5.2% median, because small accounts' Reels travel well past their follower count. Measured per view, 3% beats the 2.69% median Reel.

What is a good like-to-view ratio on Instagram Reels?#

Anything above about 2.4% beats the median Reel: across 12,505 Reels posted in 2026, the median was 2.35% and the top quarter started at 3.84%. Under 1.35% is the bottom quarter. Divide by the view count shown on the Reel, since an older counter some tools pull inflates the ratio almost threefold.

How do you calculate engagement rate on Instagram?#

Add likes and comments (plus saves and shares if you're working from your own Insights), divide by followers and multiply by 100. For Reels, also divide the same total by views, because a Reel's audience isn't limited to your followers. Use the median of your last 12 Reels rather than any single post.

Why does engagement rate drop as you gain followers?#

Because views don't keep pace with followers. In our data, median views fell from 227% of the follower count under 10K to 8% above a million, while the median like-to-view ratio stayed between 2.1% and 2.6% in every tier. A falling rate while you grow is expected.

The comparison worth running every week#

A cross-section of 321 accounts tells you what normal looks like. It can't tell you what to post next. The comparison that can is narrower: your last 12 Reels against the last 12 from five accounts in your niche, measured per viewer, every week. By hand that's 72 Reels and three numbers each, typed into a spreadsheet every Monday and stale by Thursday.

CreatorHouse tracks those accounts for you and lists every Reel's views, like rate and engagement in one sortable table (add your own handle and your Reels sit in the same list), so the Reels that broke out are obvious at a glance. Then it transcribes them, so you can see exactly what the opener said.