How long should an Instagram Reel be?
Socialinsider's analysis of 6 million Reels found 45 to 60 seconds pulled roughly double the median views of clips under 30 seconds. The shorter-is-always-better rule does not survive the data.
7 minute read
The only length limit that changes anything is three minutes
Instagram has two ceilings and only one of them matters. The technical maximum has moved several times and still varies by account and rollout, with Buffer's guide putting in-app recording at up to 20 minutes and camera roll uploads at 15, while Instagram's own Reels help still describes recording up to three minutes. Ignore that number. The one with teeth is three minutes, because Instagram does not recommend Reels longer than that to people who do not already follow you. Cross it and the Reel stays inside your follower graph instead of going out through Explore and the Reels tab. Instagram raised that recommendation cutoff from 90 seconds to three minutes in January 2025, which is recent enough that plenty of advice online is still written against the old limit.
What the length data actually shows
Socialinsider analysed 6 million Reels from brand accounts active between January and June 2026 and found the peak at 45 to 60 seconds, on both measures it tracked. Engagement rate topped out at 0.35 percent in that band, against 0.30 percent at 30 to 45 seconds and 0.28 percent under 30 seconds. Median views tell a starker story: 10,374 for a 45 to 60 second Reel versus 4,700 for anything under 30 seconds, and 4,428 once a Reel passes three minutes. The interesting part is the middle of the curve, which barely sags at all. A two to three minute Reel still returned a 0.33 percent engagement rate and around 9,000 median views, comfortably ahead of the shortest bucket.
Why very short Reels underperform here
The sub-30-second bucket finishing last will surprise anyone who has been told to cut everything to the bone, and it is worth being careful about what it means. This is not evidence that padding a clip helps. It is much more likely that short Reels are where people put the low-effort posts: the trend reuse, the quick repost, the thing filmed because nothing else was ready. Length is a proxy for intent in a dataset this broad. The honest reading is narrower than the headline: at brand accounts, in the first half of 2026, Reels around a minute long did best, and there is no evidence that cutting a good idea from 50 seconds to 20 improves anything. Cut for tightness, not to hit a number.
Completion is the mechanism, not a metric you can see
Instagram does not publish a completion-rate figure in Insights, but it is the thing everything else runs on. When Adam Mosseri set out which numbers creators should watch for reach in January 2025, he named three: average watch time, likes per reach, and sends per reach. Watch time sits at the top, and it is the metric length affects most directly. A 90-second Reel that loses people at second 12 banks less watch time than a 40-second Reel people finish, even though the first one is longer on paper. That is the whole argument for restraint, and it is a stronger argument than any recommended duration, because it is about your specific clip rather than an average of six million.
Replays quietly change the maths on short Reels
Instagram folded replays into its headline number. Since the metrics overhaul that began in 2024 and completed in April 2025, Views counts every time a piece of content is played, including repeat plays, and Replays no longer exists as a separate line in Reels insights. That matters for length in a way most guides miss. A short, loopable Reel does not just get watched, it gets watched again, and each pass adds to the same total. It is one of the few genuine advantages the very short format still has, and it only applies when the clip loops cleanly, which means the last frame has to lead back into the first rather than trailing off into a call to action.
Use the retention chart, because Instagram finally gives you one
In August 2025 Instagram added retention insights to Reels: a chart showing where viewers drop off during playback, and a skip rate that replaces the old view rate and counts the share of people who bail in the first three seconds. Instagram's own framing is that a flatter line means a more engaged audience. This is the closest thing you get to a completion rate, and it settles the length question per Reel rather than in general. If the line falls off a cliff at 22 seconds, that is your edit point next time. If it stays flat to the end of a two-minute Reel, you have permission to keep making two-minute Reels, whatever any average says.
When a longer Reel is justified
Three cases hold up. First, when the audience is already warm: past three minutes you are posting to followers, which is fine if the Reel is for followers and a waste if you wanted reach. Second, evergreen and instructional content, which Buffer flags as the reasonable use for the longest tier, because a tutorial keeps earning views long after the discovery window closes. Third, when the retention chart says so. Buffer's practical tiering is a useful default: under 90 seconds for anything meant to find new people, 90 seconds to three minutes when the story genuinely needs the room, and past three minutes only when reach is not the point. Length is a consequence of that decision, not the decision itself.
Making the cut is the hard part
None of this is difficult to agree with and all of it is annoying to execute, because deciding a Reel should be 50 seconds means finding the 50 seconds worth keeping. If you already record long form, the raw material is there: a good Reel is usually a self-contained moment that was interesting before anyone thought about distribution. FrameOS pulls those moments out of a long recording and returns them captioned and vertical, with the boundaries already set, so trimming to the retention data is an edit rather than a project. The point is to be able to test two lengths of the same idea without it costing you an afternoon.
FAQ
How long should an Instagram Reel be in 2026?
Socialinsider's analysis of 6 million Reels found 45 to 60 seconds performed best on both engagement rate and median views. Under 90 seconds is the safer default if reach matters, since Buffer reports those have the strongest chance of landing in the Reels tab. Keep every Reel under three minutes unless you are deliberately posting to existing followers.
What happens if a Reel is longer than three minutes?
Instagram does not recommend Reels over three minutes to people who do not follow you, so it stays largely inside your follower graph rather than going out through Explore and the Reels tab. The Reel still works, it just stops being a discovery tool. Instagram raised this cutoff from 90 seconds to three minutes in January 2025.
Do replays count toward Instagram Reels views?
Yes. Instagram's metrics overhaul folded repeat plays into the Views number and removed Replays as a separate metric, a change announced in 2024 and completed across formats in April 2025. A short Reel that loops cleanly can accumulate views and watch time through repetition, which is a real advantage of the very short format.
Does completion rate affect Instagram Reels reach?
Indirectly but strongly. Instagram does not show a completion rate, but Adam Mosseri named average watch time as the first of three metrics creators should watch for reach, alongside likes per reach and sends per reach. Since August 2025 the Reels retention chart shows where viewers drop off, which is the practical way to check it.
Sources
- [What Data Says] What Is the Ideal Instagram Reels Length? · Socialinsider
- Instagram Reels Length: How Long Can Reels Be? · Buffer
- Finding success on reels in 2025 · Instagram
- Instagram Adds Retention Metrics for Reels · Social Media Today
- Instagram Shares Algorithm Insights To Inform Strategy · Social Media Today
Related pages
Keep reading
More Instagram Reels →The best time to post on Instagram, according to four studies
Buffer says Thursday 9am. Later says 5am. Sprout says early afternoon. They analysed billions of posts and still disagree — here is why, and what to do about it.
8 minute readThe best time to post on TikTok, according to five studies
Buffer analysed 7.1 million posts and rates Saturday the best day. Sprout Social analysed 2 billion engagements and says avoid the weekend entirely. Both are right, for different accounts.
8 minute readThe best time to post on LinkedIn, according to five studies
Buffer's 4.8 million posts point to Wednesday at 4pm and rank Tuesday among the worst days. Sprout Social's 2 billion engagements call Tuesday the best day of the week. Here is why.
8 minute readTurn one long video into a week of posts
300 credits for 3 days · no card.
Start free