Instagram Outliers: How to Find Posts That Beat a Creator’s Usual

If you sort a creator’s posts by likes, you mostly learn how big their audience was that week. If you sort by how far each post beat that creator’s usual, you learn which ideas actually worked. That second number is an outlier score, and it’s the most useful way I know to find carousel ideas worth studying.

What an outlier post is

An outlier is a post that performed far above what that account normally gets. A carousel with 900 likes is ordinary for an account that averages 1,000. The same 900 likes on an account that averages 150 is an outlier: six times its usual.

The idea comes from YouTube research tools, where people compare a video’s views with the channel’s average. On Instagram, likes and comments are the clearest public signal for carousels, so that’s what we compare.

How the outlier score works

At Top Carousels, each carousel gets a score like 3.2×. Here’s exactly how it’s calculated:

1. Take the carousel’s likes + comments.
2. Find up to 20 other carousels from the same creator, the ones posted closest in time, before or after.
3. Take the median likes + comments of those posts. That’s the creator’s usual.
4. Divide. A carousel with 3,200 likes + comments from a creator whose usual is 1,000 scores 3.2×.

A few rules keep the number honest:

• At least 7 comparison posts. With fewer, one strange post can swing the result. Below 10 we mark the score as a small sample.
• The median, not the average. One viral post shouldn’t inflate the baseline for everything else.
• Fair comparisons only. Paid partnerships, giveaways and posts with comments turned off are left out of the comparison, and so are posts younger than a week, which haven’t finished collecting engagement.
• Collabs count. A collaboration post is still a post the audience chose to engage with.

Why it beats sorting by likes

Big accounts dominate any list sorted by raw numbers. That’s fine if you want to know who is famous, and useless if you want ideas for your own account.

An outlier score cancels out audience size. A small creator who tripled their usual sits next to a large one who tripled theirs, and both are more useful than a celebrity’s average Tuesday post. It also catches the posts that broke through on small accounts, which are often the most transferable: they worked because of the idea and the execution, not the follower count.

What counts as a real outlier

By definition, about half of any creator’s posts land at 1× or above. That’s what a median means. So 1× is the floor, not the finish line. Across the 7,394 carousels we’ve scored, here’s how rare the higher tiers are:

• 1× or more: 49% of carousels
• 2× or more: 24%
• 3× or more: 16%
• 5× or more: 9%
• 10× or more: 4%

A 3× carousel is in the top sixth of everything its creator posts. That’s where I’d start looking for patterns.

How to find outliers by hand

You can do this for any account in about ten minutes:

1. Open the creator’s profile and scroll back about 25 carousels. Skip Reels and single photos.
2. For each carousel, add up likes and comments. If likes are hidden, skip the account: you can’t score it fairly.
3. Write the totals down and find the middle value. That’s their usual.
4. Divide each carousel’s total by that number. Anything at 3× or more is worth saving.

It works, but it’s slow, and you have to do it again for every creator and every week. That’s the job Top Carousels does automatically: it collects carousels from creators across 13 niches, re-checks them as they mature and keeps only the ones that beat their creator’s usual.

What to do with an outlier once you find it

Don’t copy it. Study it. For every outlier, I ask three questions:

• What did the first slide promise? Most outliers make a specific promise or claim on slide one. See the hook patterns behind outlier carousels.
• What’s the structure? A list, a how-to, a story, an opinion? Each format has a shape you can reuse. Browse carousel examples by format.
• Could it work in my niche? The best references often come from a different niche, where the format is proven but the idea is new to your audience.

Then make your own version. I wrote up a full example of remixing an outlier, from the reference to a post that got 3K+ likes in 3 days.

Limits worth knowing

An outlier score is a reference, not a prediction. It measures likes and comments at the time we collected them, not reach, saves or shares, which Instagram doesn’t make public. A post can also outperform for reasons you can’t see, like a share from a bigger account. That’s why I look for patterns across many outliers rather than trusting any single one.

Frequently asked questions

What is an outlier post on Instagram?

A post that performed far above the account’s normal results. Instead of comparing raw likes across accounts, you compare a post with the same creator’s usual engagement. A carousel with 900 likes from an account that usually gets 150 is an outlier: six times its usual.

How do you calculate an outlier score?

Add up the post’s likes and comments, find the median likes + comments of about 20 of the creator’s other carousels posted around the same time, and divide. 3,200 against a usual of 1,000 is 3.2×.

What is a good outlier score?

2× or more is a real signal. 3× puts a carousel in roughly the top sixth of everything its creator posts, and only about 4% of carousels reach 10×.

Is there a tool to find Instagram outliers?

Yes. Top Carousels collects carousels from creators across 13 niches, scores each one against its creator’s usual and keeps the ones that beat it, so you can browse outliers by niche, format and score instead of calculating by hand.