Instagram Reels Analytics: The Metrics That Say a Reel Worked

Instagram Reels analytics sit behind View insights on every reel, and only three of the numbers there tell you whether a reel worked: skip rate compared with your account's typical skip rate, how many viewers are still watching after 5 to 10 seconds, and shares per viewer. Views, likes and total watch time mostly measure how big your account already is.

One thing the big guides leave out: the retention chart, the only place that shows where people left, disappears 90 days after you post, and no analytics tool built on Meta's API can save it for you.

Key Takeaways

• Read skip rate against your own typical skip rate, which Instagram shows right next to it. There is no universal good number.
• Retention past the first 5 to 10 seconds and shares per viewer separate a reel that travels from one that stalls.
• Views count every start and every replay. On their own they say more about your follower count than about the reel.
• Most Reels insights last two years. The retention chart lasts 90 days and is not in Meta's API, so a screenshot is the only way to keep it.

Where do you find Instagram Reels analytics?

Open the reel in the Instagram app and tap View insights under it. Instagram's Help Center says you need a public account, personal or professional, so if the button is missing, check that the account is not private first.

To compare reels instead of opening them one by one, open Insights, tap under reels to reach the content library and filter by reach or interactions (Instagram Help Center, checked November 2026).

Two limits from Instagram's Help Center, checked in November 2026: insights exist only for reels posted on or after May 14, 2021, and each reel's insights stay available for up to two years after it was published (Instagram Help Center). One chart inside them has a much shorter life, and we will get to it.

What does each Reels metric actually measure?

Instagram's definitions are narrower than the labels suggest, and two of them catch people out.

Metric	What Instagram counts	What it tells you
Views	Every time the reel starts to play or replay	Distribution, inflated by replays
Viewers (reach)	Unique accounts that saw it on screen, estimated	How far it actually got
Followers and non-followers	Accounts engaged, split by whether they follow you	Whether it left your own audience
Watch time	Total time played, replays included	Mostly a function of length and views
Average watch time	Watch time divided by initial views	How long a typical viewer stayed
Skip rate	Share of views where people skipped in the first 3 seconds	Whether the opening earned attention
Retention	Share of views still watching at each second	Exactly where people left
Interactions	Likes, comments, saves and shares	What people did after watching
Follows	Accounts that followed you because of this reel	Whether it converted

Definitions from the Instagram Help Center pages on reels insights and insights, Meta's API reference for interactions, and Instagram's August 2025 announcement of skip rate and retention, checked November 2026.

The first trap is views. An "initial view" is the first time a reel starts to play in a session, and average watch time is divided by those, but the views headline counts replays too. A 7 second loop that people watch three times looks three times bigger than it is.

The second trap is history. Meta introduced views in its API on January 21, 2025 and switched off the old plays metrics for every API version on April 21, 2025 (Meta for Developers, 2025). If you line up a 2024 report built on plays against a 2026 report built on views, you are comparing two different counts. Our post on reach vs impressions on Instagram covers the matching change from impressions to views.

Which three numbers tell you a Reel worked?

1. Skip rate, compared with your typical

Skip rate is "the percentage of views from people who decided to skip your reel during those first 3 seconds," in Instagram's words when it announced the metric in August 2025 (Social Media Today, 2025). Meta added it to the API as reels_skip_rate on December 3, 2025 (Meta for Developers, 2025), and its reference page flags it as an estimate.

Instagram shows each reel's skip rate next to your account's typical skip rate. That comparison is the reading. The raw number is not.

The common advice that a good skip rate is under 30% does not survive real data. heyDominik, working from data on hundreds of accounts, looked at the reels that got at least double their usual views: their median skip rate was 45% or lower, and only the best 10% got to 31% or lower (heyDominik, October 2026). His other finding matters more. On its own, skip rate barely predicted how a reel did.

2. How many people are still there at 5 to 10 seconds

The retention chart plots, second by second, the percent of views from people still watching. Instagram's own note on reading it: "you'll always see a downward slope, but the flatter the line, the more engaged your audience is" (Social Media Today, 2025).

Look hardest at the 5 to 10 second mark. In heyDominik's data, holding viewers past that window was the strongest single sign of a breakout reel, whatever the reel's length. A reel with a low skip rate and a cliff at second 6 has a good hook and nothing behind it, and it will not travel.

Average watch time is the same idea squeezed into one number, and it only means something against the reel's length. These are the averages his breakout reels reached:

Reel length	Average watch time of breakout reels
10 seconds or less	Close to the full reel
About 30 seconds	8 to 11 seconds
30 to 60 seconds	13 to 17 seconds
60 to 90 seconds	About 18 seconds
Over 90 seconds	About 28 seconds

These come from one creator's own data, not a published study with a method, so treat them as a rough yardstick. The principle holds either way: a third of a 60 second reel is a good result, and a third of a 9 second reel is a miss.

He walks through all three thresholds, skip rate, watch time by length and share rate, in this video:

3. Shares per viewer

Adam Mosseri, head of Instagram, named the three signals that matter most for ranking in January 2025: "watch time, likes and sends." He told creators to "pay close attention to average watch time, likes per reach, and sends per reach," and added that likes weigh slightly more with followers while sends weigh slightly more with people who don't follow you (Social Media Today, 2025).

Since April 2026, Instagram's insights include a share rate percentage, the closest thing to sends per reach (Social Media Today, 2026). If yours does not, divide shares by viewers. heyDominik's share rate bands from the same data: 0.5% to 1% changed nothing, the first clear lift came around 2%, above 3% reels started to take off, and at 5% the other numbers stopped mattering.

If you keep only one ratio, keep this one. It is the signal that carries a reel to people who don't follow you yet, which is the whole point of posting Reels. Our post on Instagram saves and shares covers what makes content people send.

Which Reels metrics can you ignore?

Views on their own. A view count mostly reflects your follower count, your posting history and how often people replayed. Report it to a brand if they ask. Don't learn from it.

Likes on their own. Mosseri says likes weigh more with followers, so a reel full of likes can be a reel your existing audience enjoyed and nobody else saw. Check the follower and non-follower split before you celebrate.

Total watch time. It grows with length and with views, so the longest reel almost always wins. Average watch time against length is the useful version.

One reel at a time. A single reel is a sample of one. A topic can carry a weak edit, and a strong edit can die on a dull topic. Line up your last 10 or 20 reels and the pattern becomes real.

How do you judge a Reel against your own account?

Instagram's choice to show skip rate next to your typical skip rate is the right instinct, and OwlStat's own data says the same about engagement.

In our study of 22,646 public business and creator accounts, measured between August and October 2026, the top quarter of every tier from 1,000 followers up had 7 to 12 times the engagement rate of the bottom quarter (OwlStat engagement benchmarks, 2026). Two accounts the same size can sit an order of magnitude apart, which makes a generic "good reel" number close to useless. Your last 20 reels are the benchmark that matters.

Where the study does help is with public numbers, the likes and comments anyone can see, including on a competitor's reels. It measures likes plus comments on each account's last 12 posts, divided by followers:

Followers	Median engagement	Top quarter starts at	Comments per 100 likes
100 to 1K	4.69%	12.37%	8.4
1K to 10K	1.38%	3.72%	7.8
10K to 50K	0.68%	2.22%	5.6
50K to 100K	0.57%	1.71%	4.3
100K to 500K	0.59%	1.72%	2.8
500K to 1M	0.52%	1.59%	2.0
1M+	0.62%	1.98%	1.3

Two ways to use it. A reel whose likes plus comments, divided by your followers, pass your tier's top quarter line did better than what three quarters of accounts your size get on a typical post. And comments per 100 likes tells you whether a reel started a conversation: at 1K to 10K followers the median account gets 7.8 comments for every 100 likes, over 1M it gets 1.3. A reel well above your tier's ratio got people talking, which a like count never shows.

The honest limit: the study covers all post formats, not Reels alone, and it cannot see saves, shares or watch time, because those are not public. Use it to place public numbers. Use your own insights for everything else.

What happens to your Reels data after 90 days?

Most of it survives. Instagram keeps each reel's insights for two years, and Meta's API says metrics data is "stored for up to 2 years" (Meta for Developers, checked November 2026).

The retention chart does not. Instagram's info text under the chart reads: "Retention rate data is only available for 90 days after reels are shared," as shown on screen in a January 2026 walkthrough of the feature (Tech Guru Support, 2026). Tap the info icon on your own chart to check the current wording.

It is also missing from Meta's API. The Reels metrics a tool can request include views, reach, likes, comments, saves, shares, reposts, total and average watch time, skip rate and a few Facebook cross-posting counts, but no retention curve (Meta for Developers, checked November 2026). That means no analytics tool built on Meta's API, ours included, can store that chart for you.

So when a reel beats your usual views, screenshot its retention chart before day 90, with skip rate and average watch time in the same shot. Three months later, when you are trying to work out why that reel worked and the next ten did not, the screenshot is the evidence.

For everything the API does expose, keeping history is easier. An Instagram analytics tool like OwlStat pulls those numbers from the official Meta Graph API, keeps them from the day you connect the account, and surfaces your top posts and Reels by reach, engagement or saves, so this quarter's reels can sit next to last year's without opening each one.

What should you check after every Reel?

A routine that takes about ten minutes a week:

1. Wait 48 hours. Meta's API documentation says the data behind its metrics can be delayed by up to 48 hours, so earlier numbers may be partial.
2. Write down five numbers: skip rate and whether it beat your typical, the percent still watching at 5 to 10 seconds, average watch time against length, shares divided by viewers, and follows.
3. Compare with your last ten reels, not with someone else's benchmark.
4. Read the combination. Skip rate worse than typical: the first frame or first line lost them. Skip rate fine but a cliff at second 5 or 6: fix what comes right after the hook. Both fine and shares low: the reel was watchable but not worth sending, which is a topic problem, not an editing one.
5. Screenshot the retention chart of anything that beat your usual views.

If you post one reel a month, skip the full routine and compare each reel with the previous three. Patterns need volume. Once you have 20 reels logged, you will know your own typical better than any benchmark can tell you.

Where to go next

• The Instagram analytics guide shows where Reels metrics sit next to everything else in Insights.
• Instagram saves and shares goes deeper on the interaction that carries reels to non-followers.
• The full Instagram engagement rate benchmarks study has every tier, the method and the CSV.

Sources

• Instagram Help Center, View insights on your Instagram reels, checked November 2026.
• Instagram Help Center, About Instagram insights, checked November 2026.
• Instagram creators account, skip rate and retention announcement, reported by Social Media Today, August 24, 2025.
• Instagram, Insights update with share rate and skip rate, reported by Social Media Today, April 26, 2026.
• Meta for Developers, Instagram Media Insights reference, checked November 2026.
• Meta for Developers, Instagram Platform changelog, entries of January 21, 2025 and December 3, 2025.
• Adam Mosseri, Instagram post on ranking signals, January 2025, as reported by Social Media Today, January 22, 2025.
• OwlStat, Instagram engagement rate benchmarks 2026, 22,646 accounts, August to October 2026.
• heyDominik, How to Force the Instagram Algorithm to Push Your Reels (based on data), YouTube, October 26, 2026.
• Tech Guru Support, What is retention in Instagram Reels, YouTube, January 15, 2026.