3 days free on every plan · Try any plan before you pay.

Start free trialStart

BlogShort-video data

Why your Reels view counts differ from one dashboard to the next

How Meta counts Reels views, why the app, Insights, the API, ad tools and outside trackers show different numbers, and what to put in a report.

By Abondy ResearchPublished 8 min read

In short: Meta counts a Reel's views as the number of times it was played, and says the count works the same way on Instagram and Facebook. The numbers still differ between dashboards because each one counts a different slice. The Instagram app adds plays on Facebook when a Reel is crossposted. The API's basic views metric counts organic plays on Instagram only, and leaves out ads. Reach counts people, not plays. Meta's data can lag by up to 48 hours. And a tool that reads a Reel's public page can see far less: in our 16 November 2026 check of 37 Reels, the public page showed about a third of the app's count.

So when a creator's screenshot, your social tool and a partner's report all show different views for the same Reel, usually nobody is wrong. They're reading different counters. This post goes through each one, with Meta's own wording, read November 21, 2026. Then it covers what we see when we count Reels ourselves, and what to put in a report.

How does Instagram count a Reel view?

Meta changed the name and kept the maths. In a creator post on November 14, 2024, it said creators "will now have a single distribution metric for all content types that's calculated the same way across Facebook and Instagram." It went on: "The new metric – called Views – tells you the number of times a reel or video was played or the number of times photo or text posts were on screen." For video it added: "On reels and videos, Plays will now be called Views. How Views are calculated will not change" (Meta for Creators).

Three more points from the same post:

  • Rewatches count. "Views also capture when people look at your content multiple times." Meta also said it was "removing the separate replays metric".
  • Some older counters stayed. "Other metrics like Reach, 3-second views, 1-minute views, Reactions, Comments and Shares will remain unchanged." So a dashboard that shows 3-second views is showing a different, stricter count.
  • It covers both apps. The same Views metric is used on Facebook.

Meta's API reference defines the Instagram metric views as "Total number of times IG Media has been played on Instagram" and labels it "Metric in development" (Meta for Developers). Meta's help page on labels explains 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" (Meta Business Help Center).

Why do Reels views differ between dashboards?

Because each place reads a different counter. Here they are, from Meta's API reference and changelog unless marked:

Where you lookWhat it countsWhy it can differ
The Instagram app and its InsightsViews across surfaces. Meta's changelog says the combined Instagram and Facebook views of a crossposted Reel "shows up on the native Instagram app"It can include Facebook plays and, for boosted Reels, paid views
API metric viewsPlays on Instagram. Meta says views and similar metrics "report organic interaction metrics only; interactions on ads containing a media object are not counted"Leaves out Facebook and ads, so it can sit below the app
API metric total_views"Total number of times the media has been seen across all surfaces, including views from promoted/boosted/ad media and Facebook"Meta says these totals "match what users see in the Instagram Insights Dashboard", but only for apps using Facebook Login
API metrics facebook_views and crossposted_viewsPlays on Facebook, and the total across both apps, for Reels shared to FacebookOnly exist for Reels that reached Facebook
Reach"Number of unique Instagram users that have seen the reel at least once"Counts people, not plays, and is "estimated"
Old tool fields plays, clips_replays_countCounters Meta retiredDeprecated in API v22.0 (changelog of January 21, 2025), and in all versions from April 21, 2025
A Reel's public page, read logged out (our check)Whatever count the page carries for visitorsIn our check, about a third of the app's count

Two notes on the table. Meta's API page says "Data used to calculate metrics can be delayed up to 48 hours", so a dashboard read in the first two days can trail the app. And Meta's label page says estimated metrics "are derived through statistical sampling or modeling, not a straight count". So reach is an estimate, not a tally.

Lookups of someone else's public Reel changed in 2026. Meta's Instagram API changelog entry for April 22, 2026 says the view_count field "now returns combined Instagram and Facebook views for crossposted video content" when looking up another account's public media. Its December 3, 2025 entry had said that count reflected only Instagram views. A tool's numbers can jump when it picks up a change like that.

Do replays and your own views count on Reels?

Replays do. Meta says Views "capture when people look at your content multiple times". Whether a creator's own plays count, the Meta pages we read don't say, so we won't guess. Either way, a high view count can mean a few people watching many times. To see how many people watched, use reach.

Why does our count of a Reel read lower than the app?

Some of our Instagram counts come from each Reel's public page, read logged out, because that page doesn't need anyone's login. In a 16 November 2026 check of 37 Reels, the app's count was a median 2.9 times what the public page showed, so the page showed about a third. An earlier, smaller check of 31 Reels found the page at about half the app's count. The gap moves, so we don't correct for it. We print Reels numbers as a floor instead.

Two things follow for brands on Abondy:

  • Our Reels counts only go up. A Reel keeps the highest view count we have read for it, so a later, lower reading never takes views away.
  • The bar uses our count. You're charged only for posts that reach 100 views within 24 hours of the link being handed in, judged on the views we read. (Our terms define views for the Ultra plan's guarantee the same way, "as counted in the Service".) So a Reel we under-read is more likely to fall under the bar, and a post under the bar isn't charged.

In our own data, the median Reel we read got 56 views, from 10,051 Reels posted from August 29 to November 12, 2026. For another 4.1% of counted Reels our reader never got a count back, so they're left out. Read that as a floor. For how Reels compare with TikTok and Shorts, see TikTok vs YouTube Shorts vs Instagram Reels. Reels also gather views more slowly in our data than the other apps; see when views stop coming in.

Are Facebook Reels views counted the same way?

Meta says Views is calculated the same way on Facebook and Instagram. But one Reel can earn views in both places. If an Instagram Reel is crossposted or recommended on Facebook, Meta's facebook_views metric counts "either plays from crossposted or cross recommended from Instagram to Facebook". So the Instagram app's number can be bigger than the Instagram-only count a tool reads. What Facebook Reels got in our data is in are Facebook Reels worth it.

What should a brand put in a Reels report?

A number nobody can check is worth little. A few habits keep Reels reports honest:

  1. Name the counter. Write "Instagram app views", "API views, organic" or "public count". Never just "views".
  2. Use one source per app for the whole report. Mixing a creator's screenshots with API reads makes some Reels look better than others for no reason.
  3. Split organic and paid. If a Reel was boosted, views leaves the paid plays out and total_views puts them in. Say which one you used.
  4. Wait two days before final numbers. Meta says its data can lag up to 48 hours.
  5. Report people as well as plays. Reach shows how many accounts saw it; views can count the same person many times.
  6. Use medians across many Reels. One big Reel can carry a whole campaign's total, so the middle Reel tells you more. For checking whether views are real at all, see is a clipping agency legit.

Where Abondy fits

On Abondy, every post and its views show on your dashboard, for TikTok, YouTube Shorts, Instagram Reels and Facebook, read the same way across every campaign. Reels counts are marked as a floor in our monthly views report for the reasons above. A post costs 12 credits, about 16¢ to 20¢ a post on a plan, and you're charged only for posts that reach 100 views within 24 hours. Plans start at $20 a month. We make Abondy, so weigh our part with that in mind.

How we counted

  • Our check: on 16 November 2026 we compared 37 Reels' counts in the Instagram app with their public pages. The median gap was 2.9 times. An earlier check of 31 Reels found about 2 times.
  • Our Reels figure: 10,482 counted Instagram posts made from August 29 to November 12, 2026 (New York time), pulled on November 21, 2026; the median is over the 10,051 we managed to read. Mostly gaming clips, largely Roblox, on small accounts. A few campaigns dominate: one brand's campaign accounts for about half the Reels we read, and Abondy's own campaigns for about a quarter. Views are each post's count when we last read it; a few were last read before the post was 7 days old. Flagged, rejected or deleted posts, Abondy's test accounts and posts with views confirmed as bought are left out.
  • Meta's rules: quoted from the pages linked above, read November 21, 2026. Meta changes these often, and its changelog said different things about public view counts in December 2025 and April 2026, so check the page before you rely on a detail.

Questions people ask

How are Reels views counted?

Meta counts the number of times a Reel was played, including when people watch it more than once. It says the count works the same way on Instagram and Facebook.

Why do my Reels show more views in the app than in my analytics tool?

The app can include views from Facebook and from boosts. A tool that reads the API's views metric gets organic plays on Instagram only. Meta's data can also lag up to 48 hours.

Do rewatches count as views on Instagram?

Yes. Meta says Views "capture when people look at your content multiple times". Reach is the metric that counts each account once.

Why do some trackers show far fewer Reels views?

One likely reason is that the tracker reads a Reel's public page instead of Meta's API or the app. Ours does for some Reels, and in our 16 November 2026 check of 37 Reels, the public page showed about a third of the app's count.

Read next