How to Analyze Competitor Instagram Reels
Views, Ratios and the Creative Choices Behind the Reels That Travel
Short answer
Analyze competitor Instagram Reels by logging the last twenty from their Reels tab with views, likes, comments, length, hook and audio. Compare each Reel with that account's own median views, not with other brands. Then watch the top and bottom three side by side to find the creative choices that separate them.
1. What does a competitor Reel view count actually measure?
Starts and replays, not people. Instagram moved to views as its main metric across formats, and a Reel view is counted each time it starts or replays, so one person watching three times counts three times.
- Reel views
- The public count of times a Reel started playing or replayed on a screen, including repeat views from the same account, shown on the Reel itself.
- Also known as: Reel plays, Play count
- Full definition of Reel views
Instagram shifted to views as the primary metric across Reels, photos, carousels and stories, replacing plays and impressions in its reporting [s1]. For an outsider the practical meaning is unchanged: the number on a competitor Reel is a count of plays, which makes it a distribution signal but not reach.
That still makes it the most useful public number on Instagram. Likes and comments tell you how the audience responded. Views tell you how far Instagram carried the Reel in the first place, and Reels are the format most likely to be shown to people who do not follow the account.
2. How do I collect a competitor's Reels baseline?
Open the Reels tab on their profile, log the last twenty Reels with seven fields each, and compute the account's own median views. Twenty Reels is enough for a stable median on most brand accounts.
- Open the Reels tab, not the main grid. The Reels tab lists the account's Reels in order, each with its public view count. The main grid mixes formats and may show pinned posts first, which breaks the chronological order you need.
- Log seven fields per Reel. Record date, views, likes, comments, length in seconds, the hook (what happens in the first seconds) and audio type (original, trending track or voiceover). Add the topic in a few words.
- Flag collabs and paid partnerships. A collab Reel appears on several profiles with one pooled set of views and likes, and a paid partnership Reel has a creator audience behind it. Mark both, because they are not comparable with solo brand Reels.
- Compute the median views. Sort by views and take the middle value, not the average. One Reel that reached far beyond the followers will drag an average upward for months and make every normal Reel look like a failure.Express every Reel as a multiple of that median. A 3x Reel is an outperformer for this account whether the median is 800 or 80,000.
3. Which ratios tell me the most about a competitor's Reels?
Three: views against followers, likes per thousand views, and comments per thousand views. The first shows reach beyond the audience. The other two show how the people who watched responded.
| Ratio | Formula | High means | Watch out for |
|---|---|---|---|
| Views to followers | Reel views divided by follower count | Instagram carried the Reel well beyond the existing audience | Boosted Reels, which buy the same effect |
| Likes per 1,000 views | Likes divided by views, times 1,000 | Viewers liked what they saw, not just scrolled past | Hidden like counts, which make it impossible |
| Comments per 1,000 views | Comments divided by views, times 1,000 | The Reel prompted conversation or questions | Giveaways and "comment to get the link" calls to action |
The combination is where the insight sits. High views with low likes per thousand means the Reel was distributed widely but did not land: a hook that stopped the scroll and content that did not deliver. Moderate views with high comments per thousand usually means a Reel that spoke directly to the existing audience, which is a different and equally valid goal.
4. What creative choices should I compare between their top and bottom Reels?
Watch the three highest and three lowest multiples back to back and compare the hook, length, audio, on-screen text and who appears on camera. The differences between the groups are the finding.
Compare the two groups on these dimensions:
- Hook: a question, a bold claim, a visual surprise, or a slow brand intro. The first seconds decide whether the view count grows at all.
- Length: Instagram allows Reels up to three minutes [s3]. Note whether their winners cluster short or long, rather than assuming either.
- Audio: original voiceover, trending track or no audio at all. Trending audio can ride a wave; original voice builds recognition.
- On-screen text: whether the Reel still makes sense with the sound off.
- Face on camera: founder, employee, customer, creator or none. Brands often find one of these works far better, and switch visibly.
- Topic and format: tutorial, behind the scenes, product demo, meme or customer story.
Look for a dimension that separates the groups cleanly. If Veltrix's top three Reels all feature an employee talking to camera and the bottom three are all polished product montages, that is a lesson they have probably already learned and are acting on. Check their next few Reels to see whether the pattern is becoming the norm.
5. How do I keep boosted and collab Reels from distorting the analysis?
Check the Meta Ad Library for boosted Reels, flag collab and partnership Reels by their header labels, and keep both in separate lists. Their views measure budget or a borrowed audience, not the brand's own content.
A boosted Reel is also an ad, so while the boost is active it appears in the Meta Ad Library for the linked Facebook Page, alongside the brand's other Instagram placements. A Reel several times above the median with ordinary likes per thousand views is worth that check.
Collab Reels show more than one account name at the top and pool their numbers across all collaborators. A paid partnership label means a creator was paid to make or publish it. In both cases the view count reflects an audience the brand borrowed, which is useful to know about their partnership strategy but useless as a benchmark for their own content.
6. How often should I rerun the analysis, and what should come out of it?
Monthly is enough for most competitors. Each pass should produce one sentence per competitor about what their winning Reels have in common, and one change since the last pass.
Reels strategy shifts over weeks, not days. A monthly pass of twenty Reels per competitor keeps the median current and catches shifts in length, hooks or on-camera talent while they are still new.
The output worth keeping is short: "Sondera's top Reels this month were all under 20 seconds with a customer on camera, up from none last month." That sentence is something your team can test. A table of view counts is not.
Oppira tracks the posts and Reels of the competitors you follow, scanning active profiles daily, with engagement history per rival and head to head, so the median and the outliers are already in place when you sit down to watch the Reels that matter.
Key Takeaways
Views count plays, not people
Replays count, so a Reel view number is a distribution signal and never a reach figure.
Benchmark against the account's own median
Express each Reel as a multiple of that account median views, which removes audience size from the comparison.
Use three ratios together
Views to followers shows distribution. Likes and comments per thousand views show how viewers responded.
Compare top and bottom Reels directly
Hook, length, audio, text and who is on camera usually separate a competitor's winners from the rest.
Separate borrowed audiences
Boosted, collab and paid partnership Reels measure budget or someone else's audience, so list them apart.
You only see published survivors
Trial reels do not appear on the profile unless shared later, so heavy testers show fewer Reels than they make.
Frequently Asked Questions
Sources
- Instagram Updates Metrics to Focus Creators on Views Social Media Today, August 2024.Reports Instagram move to views as the primary metric across formats, with repeat views counted.
- Test Content With Non-Followers Using Trial Reels Meta, December 2024.Official announcement stating trial reels are not shown on the profile grid or Reels tab unless shared with everyone.
- Instagram Officially Expands Reels Length to 3 Minutes Social Media Today, January 2025.Reports the January 2025 change to three-minute Reels announced by the head of Instagram.
- Hide or unhide like counts on Instagram posts and reels Instagram Help Center, September 2026.Confirms account owners can hide like counts on their posts and Reels.
- Meta Ad Library Meta Platforms, September 2026.Where a boosted Reel appears as an ad while the boost is active.
Explore More
Related analyses, benchmarks, and industry insights
Related Guides
Glossary Terms
Turn reading into a reaction
Oppira watches your competitors, keeps your playbook current, and drafts the response. See your market clearly by tomorrow.