Reels report Updated July 20, 2026

Instagram Reels Report 2026: What Actually Works

Benchmarks from 315 reels across 96 Instagram accounts, analyzed frame by frame.

341
Median views per reel
4.1%
Median engagement per view
3.3%
Median like-to-view ratio
31.7%
Median watch ratio

How many views does the average reel get?

The median Instagram reel in this dataset gets 341 views, and the middle half of reels land between 140 and 1,542 views.

Most published reel benchmarks lean on large brand and celebrity accounts, which drags the average way above what a normal account ever sees. That's the wrong yardstick for most creators. Our sample skews small and mid-size, so the middle of the range tells you more than any single average.

Views per reel
The top quarter of reels stretches far beyond this range, so the scale stops early to keep the middle readable.
See the numbers as a table
Views per reel
Bottom quarter Median Top quarter
140 341 1,542

What is a good like-to-view ratio?

The median like-to-view ratio on Instagram Reels is 3.3%, about 1 like for every 30 views, across the 268 reels with view data.

Likes are the interaction most people can see, so the like-to-view ratio is the easiest benchmark to sanity-check your own reels against. A quarter of reels sit below 1.5%, and a quarter clear 5.2%. If your reels land inside that band, you're in normal territory for this sample.

Like-to-view ratio across 268 reels
See the numbers as a table
Like-to-view ratio across 268 reels
Bottom quarter Median Top quarter
1.5% 3.3% 5.2%

Average watch time and retention

The median Instagram reel holds 31.7% of its length in watch time, and shorter reels hold far more: reels under 15 seconds keep 66.9% of their length versus 15.9% for reels over 60 seconds.

Watch ratio here is average watch time divided by reel length, so it reads as the share of a reel a typical viewer sits through. Length is the biggest lever: shorter reels are far more likely to be watched in full, while long reels lose most of their audience before the end.

Avg watch ratio by length
Average share of each reel watched, grouped by length. Bucket sizes run 114, 84, 54 and 56 reels.
Avg engagement per view by length
Average interactions per view, grouped by length.
See the numbers as a table
Watch ratio and engagement by reel length
Reel length Reels Avg watch ratio Avg engagement per view
0 - 15 seconds 114 66.9% 7.5%
15 - 30 seconds 84 31.6% 9.4%
30 - 60 seconds 54 20.3% 4.5%
60+ seconds 56 15.9% 7.8%
66.9%

Reels under 15 seconds hold 66.9% of their length on average, versus 15.9% for reels over 60 seconds. If retention is your problem, cutting length is the fastest fix.

Which hooks perform best

Face-to-camera and cinematic B-roll openings earn the highest engagement per view in this sample, while silent, music-only reels hold attention longest.

Avg engagement per view by opening style
See the numbers as a table
Opening style, reels with 20 or more examples
Hook type Reels Avg engagement per view Avg watch ratio
B-Roll / Cinematic 31 9.1% 27.7%
Face-to-Camera Direct 53 9.0% 38.2%
Text-Overlay Promo 70 5.3% 37.8%
Product Showcase 28 5.1% 28.0%
Silent / Music-Only 36 3.3% 48.4%

Smaller-sample hook types (list hooks, story hooks, question hooks and others) exist in the data but are excluded here because each has fewer than 20 reels, too few to report a trustworthy average.

The first 3 seconds

What's on screen in the opening frames pulls engagement and retention in different directions, so there's no single winning combination. Here are the honest splits.

Engagement per view, with vs without
Watch ratio, with vs without
See the numbers as a table
Opening signals, with vs without
Signal Reels Avg engagement per view Avg watch ratio
Text overlay, with 229 7.7% 37.4%
Text overlay, without 76 5.6% 46.8%
Face on screen, with 157 6.7% 39.7%
Face on screen, without 148 7.9% 38.8%
Eye contact, with 118 7.6% 39.9%
Eye contact, without 187 7.0% 38.9%

Text overlay lifts engagement (7.7% versus 5.6%) but lowers watch ratio (37.4% versus 46.8%), so captions buy taps at the cost of hold. Reels without a face on screen slightly out-engage reels with one (7.9% versus 6.7%), yet when a face is present, direct eye contact edges ahead (7.6% versus 7.0%).

What the top 10% do differently

The top 10% of reels by engagement (22 reels) averages 30.0% engagement per view, against 4.1% for the mid-pack, and holds a 49.2% watch ratio versus 30.2%.

Top 10% vs mid-pack, headline measures
See the numbers as a table
Top 10% vs mid-pack by engagement
Group Reels Avg engagement per view Avg watch ratio
Top 10% 22 30.0% 49.2%
Mid-pack 21 4.1% 30.2%

The split isn't about polish. The top 10% actually score higher on visual clarity (78.4 versus 76.7 on a 0 to 100 scale) but lower on human presence (42.7 versus 58.8). In this sample, a clean, readable frame travelled further than simply putting a person on camera.

Component scores, top 10% vs mid-pack (0 to 100)
See the numbers as a table
Component scores on a 0 to 100 scale
Component Top 10% Mid pack
Pattern clarity 66.8 72.6
Human presence 42.7 58.8
Emotion energy 51.8 56.7
Motion pacing 55.2 64.8
Text context 54.8 64.0
Visual clarity 78.4 76.7

Common strengths and weaknesses

Across every reel we read, the same handful of strengths and weak spots kept coming up. The number beside each is how many reels showed that pattern.

Most common strengths
  • High motion intensity with multiple scene changes enhances visual engagement. 65
  • Text readability is good, enhancing message clarity. 48
  • Direct eye contact enhances viewer engagement. 37
  • Good brightness and visual clarity improve viewer experience. 34
  • Visual clarity is high, making the content visually appealing. 27
Most common weaknesses
  • Lack of human presence reduces potential emotional connection; consider adding a voiceover. 55
  • Lack of spoken content reduces potential engagement; consider adding a verbal hook. 25
  • Limited motion and scene changes; increasing dynamism could improve engagement. 20
  • Dark brightness could be improved for better visual appeal. 11
  • Limited emotional energy; incorporating dynamic visuals or music could enhance engagement. 9

Methodology

These benchmarks are aggregated and anonymized from 315 reels across 96 Instagram accounts analyzed by Highviz. Every reel is read frame by frame, and the numbers here are pooled across accounts with no account, handle or individual reel identified.

Engagement per view is total interactions divided by views, per reel. Watch ratio is average watch time divided by reel length. Like-to-view ratio is likes divided by views, for the 268 reels that had view data. We lead with medians rather than averages because small datasets are outlier-prone, and a few viral reels can pull an average far away from what a typical reel sees. Internal and deleted accounts are excluded. Last updated July 20, 2026.

We deliberately leave out best-time-to-post tables. With 315 reels split across 24 hours and 7 days, the per-slot samples are too small to trust, and publishing them would imply a precision the data doesn't support.

See how your reels compare

Free analysis of your 10 latest posts and 3 reels, no credit card.