Meta Publishes a Reels Skip Rate the Instagram App Never Shows You

What is reels skip rate on Instagram?

Instagram's reels skip rate is the share of first-time plays where the viewer moved on inside the opening 3 seconds, published as the field reels_skip_rate on the media insights endpoint. Meta's Instagram Media Insights reference, last updated 18 June 2026, defines it in full: "The percentage of views from people who skipped during the first 3 seconds of the reel. This is calculcated as the number of views that skipped the reel during the first 3 seconds divided by the number of intial views. An intial view is when the reel starts to play for the first time in a reel session."

The typos are Meta's. That reference is the only place the 3 second cutoff appears in writing, and Meta publishes no benchmark for a good skip rate, so the only comparison available to you is your own back catalogue.

How does Meta calculate reels skip rate?

Numerator over denominator, both halves named in the definition:

Reels skip rate = views that skipped during the first 3 seconds / initial views

Work it. A reel returns 12,400 initial views, and 5,208 of those viewers left before the 3 second mark.

5,208 / 12,400 = 0.42, so reels_skip_rate reports 42 percent.

The denominator is the part people get wrong. An initial view is "when the reel starts to play for the first time in a reel session", so replays are excluded. The Views number on the same reel counts something wider: Instagram's Help Center defines reels Views as "The number of times your reel starts to play or replay". If that reel shows 15,900 Views in the app, dividing 5,208 by 15,900 gives 32.8 percent, which is not the skip rate and never was.

Where can you see reels skip rate?

In the API, and nowhere else on Instagram. The Help Center article "View insights on your Instagram reels" lists what the app surfaces: Views, Watch time, Viewers, Average watch time and Follows. Skip rate is absent.

Average watch time and skip rate share the same denominator, which makes them the honest pair to read together. The Help Center spells that out for the app-side metric: average watch time "is calculated by the watch time divided by the number of initial views", using the identical initial view definition.

That shared base is why a tool can show you both without a caveat, while any ratio you assemble from the app's Views number needs one. It is the same trap that separates reels from feed posts and stories in reporting.

Why does the Instagram app hide it?

Meta has not published a reason. What Meta has published is a warning label: reels_skip_rate carries "Metric is estimated and in development", the strongest pair of disclaimers on the media insights table.

The estimated half means sampling. Meta's metrics labeling page states that "Estimated metrics are derived through statistical sampling or modeling, not a straight count", and explains why account-counting metrics qualify: "Our metrics that count Meta Accounts are sampled because it takes a large amount of data to calculate them."

The in development half is the sharper one. Meta writes that in development metrics "are measurements we're still testing. We're working out the best way to measure them and may make adjustments until they're accurate", then states plainly that businesses should not use them for "Historical comparison of advertising performance" or "Strategic planning or important business decisions". Meta is telling you not to put this number in a year-over-year chart.

What does a high reels skip rate tell you?

That the opening 3 seconds lost people, and nothing about the rest of the reel. The metric's window closes at 3 seconds by definition, so a reel with a 60 percent skip rate and strong average watch time is holding the 40 percent who stayed. Those are two separate problems with two separate fixes.

Read skip rate against ig_reels_avg_watch_time before you change anything. High skip with high average watch time points at the first frame or the cover image. Low skip with low average watch time points at the middle of the reel. The Instagram reels algorithm responds to what people do after the opening, so fixing the wrong end of the video moves nothing.

Which media types return reels skip rate?

Reels only. The media insights reference lists reels_skip_rate against REELS and no other media product type, so feed posts and stories return nothing. That is narrower than most of the table, where reach, views, shares and total_interactions all span FEED, REELS and STORY.

The request itself is one call:

GET https://graph.instagram.com/v26.0/<INSTAGRAM_MEDIA_ID>/insights
  ?metric=reels_skip_rate,ig_reels_avg_watch_time,views
  &access_token=<TOKEN>

Period is fixed. Meta states that "Period is automatically set to lifetime in the request and cannot be changed", so the percentage covers the reel's whole life to date and shifts as the reel keeps circulating. Two pulls a month apart on the same reel give two different numbers, both correct.

Do not pass a breakdown parameter with it. Meta warns that requesting a breakdown on a metric that does not support one returns "An unknown error has occurred.", and at media level only profile_activity and navigation accept one.

What limits apply to reels skip rate data?

Four, all from the same reference and the same Help Center article.

Data used to calculate media metrics "can be delayed up to 48 hours", so a skip rate read the hour after publishing is provisional. Metrics data "is stored for up to 2 years", and the Help Center confirms reels insights "are available to view for up to two years after the reel's original publish date". Reels insights exist only for reels "posted on and after May 14, 2021". And when a value is unavailable, Meta "returns an empty data set instead of 0 for individual metrics", so a reel with no skip rate entry has an unknown skip rate, not a zero one.

That last one breaks averages quietly, because code that reads a missing value as 0 reports a healthier hook than you have.

How should you act on reels skip rate?

Test openings against it, one variable at a time, and judge the result on a batch of reels instead of one. A single reel's skip rate carries the sampling noise Meta warns about plus whatever the recommendation system did that week, so a 4 point move on one video is not a finding.

Publishing enough reels on a schedule to make that comparison possible is the practical bottleneck, and it is what the Instagram scheduling tool and social media analytics in adaptlypost handle together. Pull the field with every other number in the Instagram analytics tool view, the same way you would track impressions against views, and let the trend across the batch tell you whether the hook improved.

Frequently asked questions

What counts as a skip on Instagram reels?

Meta defines it as a view where the person left during the first 3 seconds of the reel. That cutoff appears in the reels_skip_rate description in the Instagram Media Insights reference and nowhere else in Meta's documentation. There is no separate skip definition for stories, feed video or any other surface.

Why does my skip rate not match my Views number?

Because they use different bases. Skip rate divides by initial views, which Meta defines as the reel starting to play for the first time in a reel session, while the app's Views metric counts "the number of times your reel starts to play or replay". Replays inflate Views and never enter the skip rate denominator.

Can I see reels skip rate in the Instagram app?

No. The Help Center article on reels insights lists Views, Watch time, Viewers, Average watch time and Follows, with no skip rate entry. The only way to read the number is through reels_skip_rate on the media insights endpoint, which is why it shows up in third-party dashboards and not on your phone.

Is reels skip rate reliable enough to report?

Meta labels it estimated and in development, and its own guidance says in development metrics should not be used for historical comparison or strategic planning. Use it as directional guidance on hooks across a batch of reels. Keep it out of quarterly comparisons and out of any chart a budget decision rests on.

Does reels skip rate include boosted reels?

Meta does not state it for this field. The media insights limitations name comments, likes, views and total_interactions as organic-only examples, and reels_skip_rate appears on neither that list nor any paid-inclusive one. Treat the paid component as unstated instead of assuming either way.

How far back can I pull reels skip rate?

Two years from the reel's publish date, and no earlier than 14 May 2021. Meta stores media metrics "for up to 2 years", and the Help Center confirms reels insights start with reels posted on or after that date. Anything older than the two-year window has to come from data you pulled and stored yourself.

How fresh is reels skip rate data right after I publish?

Data used to calculate media metrics can be delayed up to 48 hours, so a skip rate pulled right after publishing is provisional and shifts once Meta finishes processing it. Wait a day or two before treating an early number as final.

What does it mean when reels skip rate is missing for a reel?

Meta returns an empty data set instead of a zero when a metric has no value, so a reel with no skip rate entry has an unknown skip rate, not a perfect hook. Code that reads a missing value as 0 will report a healthier skip rate than the reel actually had.

Can I request a breakdown parameter with reels skip rate?

Requesting a breakdown on reels_skip_rate breaks the call: at media level only profile_activity and navigation accept a breakdown parameter, and Meta warns any other combination returns an unknown error. Leave the metric on its own in the request.

Why should I read reels skip rate together with average watch time?

Average watch time and skip rate divide by the same initial-views denominator, which makes them comparable without the caveat any math built on the app's Views number needs. Reading them side by side shows whether a low average watch time comes from an opening that lost people or a middle that did.