How to find outlier reels on Instagram

An account with 2,514 followers posted a short list of Nintendo sounds. It got 3.5 million views, about 1,400 times its follower count.

An account that small doesn't have an audience big enough to carry a reel that far. Something in the idea did it. That's why reels like this are worth your time: you can borrow an idea that works without a famous face behind it.

People call these reels outliers. Instagram won't point them out for you, though. It shows a view count and nothing to measure it against. So this guide shows you how to find outliers by hand, why you should judge them against accounts your size, and which tools do the finding for you. Fair warning: we make one of those tools, Aubrium, so we show where every number comes from.

What is an outlier reel?

It's a reel that did far better than you'd expect from the account that posted it. The word comes from statistics, where an outlier is a point that sits well away from all the others.

To call something far better, you need a yardstick. There are two in common use, and they answer slightly different questions.

Most tools use the first. Aubrium and Virlo use the second. Both are fair, as long as there's a yardstick at all, because a view count on its own tells you very little. Take 600,000 views. It sounds like a hit. But among viral reels from accounts with 50,000 to 100,000 followers in our library, the middle one had 637,342 views. For an account that size, 600,000 is ordinary.

Why compare accounts your size?

Because the bigger the account, the less its views say about the idea. A big following carries a reel a long way on its own. A small one can't, so when a small account's reel travels, it's the reel doing it.

You can see this in our library of viral reels. The smaller the account, the further its hits travelled against its following.

Now sort the same reels by raw views and the order flips. Only 12.5% of the viral reels from accounts under 2,000 followers passed a million views. For accounts with 50,000 to 100,000, it was 38.5%. Sort any list by views and the bigger accounts float to the top, while the small accounts' hits, the ones that travelled furthest for their size, sink out of sight.

So the most useful outliers for you come from accounts close to your own size. Their reels had the same head start yours will have, which is not much.

How do you find outlier reels by hand?

You need a phone, a notes app and about a dozen accounts in your niche. Here's the method we'd use.

Here's how the sums work. Say an account's last 12 reels got between 1,500 and 30,000 views, and the middle one sits at 4,000. A reel with 12,000 views is three times its usual. The one with 30,000 is seven and a half.

There's no official line for what counts. Tools draw it in different places: CreatorHouse flags a reel at twice the account's median,⁶ Creadar at three times,¹¹ and Octupie's own example is a reel at ten times the creator's usual.³ Three is a sensible place to start. If you find dozens a week, raise it. If you find none, lower it.

The method works. It's just slow. Ten accounts means writing down 120 numbers, and you'll only ever find outliers from accounts you already know about. The reel that's new to you, from an account you've never heard of, won't turn up this way.

How do you find outliers from accounts you don't know?

You search viral reels by topic instead of by account. That's what viral reel finders are for. Some search a library of reels they've already collected, some watch accounts you give them, and some pick the accounts for you.

Which kind you choose decides what you'll see. A tool that only watches your list will never show you a stranger's breakout. A tool that picks accounts for you tends to pick big ones: when our team asked these tools to suggest accounts for small creators, every pick we saw had a big following, from 279,000 followers to 4.9 million in Sandcastles' starter list.⁹

In Aubrium, you pick your niche and get reels that already reached far more people than their creators have followers, all from accounts with 1,000 to 100,000 followers. You can filter by account size, country, format and hook type, and when your topic isn't in the library, you can search it. A search costs 10 Sparks, and one that finds no viral reels costs nothing.

Which tools find outlier reels for you?

These are the tools people name for this job, and what each one measures. Prices are each tool's lowest monthly plan, read on its own site.

A few other names come up for this, but they do a different job. Metricool is a scheduler that lists competitors you add, sorted by whatever metric you pick; it doesn't flag outliers.⁷ Social Blade tracks how accounts grow, with followers, engagement rate and average likes, but not reels.⁸ FastMoss is a TikTok Shop analytics tool and says nothing about Instagram.¹² So they sit outside the table.

Can you trust a tool's "viral" label?

Not without checking. Our team studied these apps hands-on in October 2026 and ran more than 1,300 searches across them, with the same keywords in each. Each tool decides "viral" its own way, so we checked each one against its own rule.

What we found when we checked the outliers

• No: Viral Outliers: filtered to Instagram accounts under 100,000, 45% of its top results did worse than their creator's usual
• No: Virlo: 87% of the Instagram reels its beauty research agent collected had fewer views than the account has followers
• No: Sandcastles: 7% of its outlier scores were broken, with the score equal to the view count
• No: CreatorHouse: the same reel scored 0.5 or 1.6, depending on the time window picked
• Yes: Aubrium: every reel reached far more people than its creator has followersaubriumaubrium

The fix is the same divide step from the method above. Before you build a reel on someone's "viral" post, check its views against the account's followers or its usual. Thirty seconds of sums saves you from building on a reel that was only average.

What should you do with an outlier once you find it?

Work out why it travelled, then make your own version. Look at the first line, what's on screen in the first second, the format and the sound. Those are the parts you can borrow.

Don't re-upload it. Instagram recommends the original over a copy and stops recommending accounts that mostly post other people's work.¹⁰ Keep the idea and the shape, and make everything else yours.

Questions people ask

How we did this

The method is the one we'd use ourselves, written out step by step. The tools' findings come from our team's hands-on study of these apps in October 2026: more than 1,300 searches with the same keywords in each, every result saved and checked against the tool's own rule. Aubrium's library numbers come from its own database, read on 4 November 2026: the middle viral reel by account size, and the share of each size band past a million views. Every reel in the library already went viral, so none of these numbers is the chance that a new reel will.

What each tool says about itself, and its price, comes from its own site, read on 4 November 2026. We didn't test Metricool, Social Blade or FastMoss for this guide, because none of them finds outliers from accounts you don't already know. Apps change; if something here is out of date, tell us and we'll fix it and move the date. And yes, Aubrium makes this page.

Sources

1. 1
   Aubrium pricingaubrium.com · Plans and what a search costs. Read 4 November 2026.
2. 2
   ViralSpotviralspot.ai · Price; outliers scored against each account's normal. Read 4 November 2026.
3. 3
   Octupieoctupie.com · How it picks accounts; "If a creator usually gets 20k views and one hits 200k, that post, at 10 times their usual, is the outlier." And its pricing: a free plan with no end date, Creator $29 a month. Read 4 November 2026.
4. 4
   Sandcastles video filter settingsSandcastles help centre · The outlier score: a video's views divided by the median views of the channel's videos in the 90 days before it. Read 4 November 2026.
5. 5
   Viral Outliersviraloutliers.com · Plans; scores a post against its account's own average. Read 4 November 2026.
6. 6
   ViralFindr alternativesCreatorHouse blog · Every reel scored against the account's median views, "anything at 2x or more flagged as an outlier". Read 4 November 2026.
7. 7
   Competitor analysisMetricool help · Competitors you add, their posts sorted by the metric you choose. Read 4 November 2026.
8. 8
   Social Blade, archived copy of 3 November 2026Social Blade · "tracks daily statistics and growth"; an Instagram profile shows followers, engagement rate and average likes and comments. The live site blocks fetchers, so we read archived copies. Read 4 November 2026.
9. 9
   We ran 1,300+ searches in 8 viral reel findersAubrium · Our team's hands-on study, October 2026: the starter watchlist, the outlier checks. Read 4 November 2026.
10. 10
    Original content guidelines, archived copy of 30 September 2026Instagram for Creators · An account that "primarily posts unoriginal reels, photos, or carousels" may not be seen in recommendations to new audiences. The live page answers fetchers with an error, so we read the archived copy. Read 4 November 2026.
11. 11
    Best Instagram competitor analysis tools in 2026Creadar · Its own tracker flags "Reels at 3x the account's median"; price. Read 4 November 2026.
12. 12
    FastMossfastmoss.com · "a TikTok Shop analytics platform"; no Instagram feature on its home or pricing page. Read 4 November 2026.

Changelog

• 2026-11-04First version.