Learn / Method

How Instagram outlier scores work

An outlier score compares a post’s views with the median views of the same creator’s comparison posts. A score of 1 means the post matches that median. It describes observed reach; it does not explain why a post performed or predict your results.

This fixed example uses public posts from @bilawal.ai. Verified 14 Oct 2026, 11:45:40 Bangkok (UTC+7). The live pages may now show newer counts.

One post, 30 comparison posts

The example Instagram reel was published 2 Oct 2026, 11:58:07 Bangkok. Its recorded view count was 48,761 at 11 Oct 2026, 07:00:01 Bangkok.

48,761 ÷ 41,220.5 = 1.18×

Sort the 30 comparison counts from lowest to highest. The middle two are 37,353 and 45,088; their average is 41,220.5. The post therefore had about 18% more recorded views than this median. The product rounds the displayed median to 41,221, but calculates the ratio before that rounding.

See the live post analysis or analyze your Instagram account.

What enters the comparison?

The current method selects the creator’s 30 most recently published other live posts in our catalog, excluding the selected post. It takes each comparison post’s latest stored snapshot and retains positive view counts. At least five comparison counts are required for a score.

This is a 30-post selection, not a 30-day window. These comparison posts were published between 1 July and 1 October 2026. Their snapshots were captured on 7, 9, 11 and 13 October 2026 at 07:00:01 Bangkok (00:00:01 UTC). They were not measured simultaneously or at equal ages after publication.

The catalog may not contain every post. Counts change, and newer catalog posts can change the comparison set. The general method does not restrict comparisons to posts published before the selected post. A high score is a research lead, not evidence that a hook, format or tool caused success.

The exact comparison data

Every row links to the original public post. All displayed dates use Bangkok time (UTC+7); the machine-readable dates include UTC. No account analytics or private audience data are used.

Frozen snapshot set, verified 14 October 2026
PostViewsPublished (Bangkok)Captured (Bangkok)
Post 137,353
Post 2220,325
Post 3329,969
Post 4559,568
Post 545,088
Post 6611,583
Post 7255,580
Post 81,375,084
Post 918,974
Post 10227,074
Post 11118,691
Post 12442,755
Post 135,475,004
Post 1435,337
Post 15196,364
Post 1633,384
Post 1749,749
Post 1834,516
Post 1934,550
Post 2035,181
Post 2121,946
Post 2264,203
Post 233,139
Post 24760
Post 25170
Post 26158
Post 27484
Post 2824,460
Post 2913,940
Post 3076,952