# How Many Views Is Viral? Only 18% Hit 2x (1,008 Videos) (https://www.socialcrawl.dev/blog/how-many-views-is-viral)
> Viral is relative to the creator. 18.2% of 1,008 videos from 62 creators hit 2x their own median views on 1 October 2026. Try the free viral video finder.
How many views is viral depends on whose video it is. A video that gets 200,000 views is a breakout for one account and a bad week for another, so the useful test is a multiple of that creator's own median views. We measured 1,008 scored videos from 62 creators on TikTok, YouTube and Instagram Reels on 1 October 2026, and 18.2% reached 2x their creator's median while 4.6% reached 5x.
The formula is the one the API returns on every video: `computed.vs_creator.ratio`, the video's views divided by the median views of the creator's other comparable posts on the same page. A ratio of 2.0 means double the usual, 0.43 means 43% of it.
The posts that rank for this question give per-platform rules of thumb without original data. Outlier tools such as vidIQ score videos against a channel average but do not publish how often an outlier happens. This post publishes the base rate and the method, so you can check it.
## How many views is considered viral?
A video counts as viral relative to its creator when it reaches at least twice that creator's median views. In this measurement of 1,008 recent videos from 62 creators on TikTok, YouTube and Instagram Reels (1 October 2026), 18.2% did.
| Threshold vs. the creator's median | Videos | Share of 1,008 |
|---|---|---|
| 2x or more | 183 | 18.2% |
| 5x or more | 46 | 4.6% |
| 0.5x or less | 166 | 16.5% |
54 of the 62 creators (87%) had at least one 2x video on the page the API returned, and 27 (44%) had at least one 5x video. The median ratio across all scored videos is 0.99, which is what a ratio against a median should give. The average video is average, and the tails are where a finder earns its keep.
Almost as many videos fell to half their creator's usual (16.5%) as reached double it (18.2%). Outliers go down about as often as they go up. A 2x video is common, and a 5x video turns up in roughly one of every twenty-two. So how many views to go viral changes with every account. This post uses twice that account's own median as the test.
Before you quote any of this, the sample is hand-picked. It skews toward large, active, English-language brand accounts, and it is not "all creators". The shares describe these 62 accounts, not the platforms.
What is considered viral is a multiple of a creator's own median. In this
sample of 1,008 videos, 18.2% reached 2x and 4.6% reached 5x. Treat those as
base rates for well-known accounts, not as a threshold for yours.
## How many views is viral on TikTok, YouTube and Instagram Reels?
On TikTok, 23.8% of 189 recent videos from 19 creators reached 2x their creator's median views (SocialCrawl, 1 October 2026). YouTube and Instagram Reels came in lower.
| Platform | Creators | Videos | 2x or more | 5x or more | 0.5x or less | Creators with a 2x video |
|---|---|---|---|---|---|---|
| TikTok | 19 | 189 | 45 (23.8%) | 16 (8.5%) | 37 (19.6%) | 19 of 19 |
| YouTube | 20 | 580 | 101 (17.4%) | 24 (4.1%) | 96 (16.6%) | 19 of 20 |
| Instagram Reels | 23 | 239 | 37 (15.5%) | 6 (2.5%) | 33 (13.8%) | 16 of 23 |
| All | 62 | 1,008 | 183 (18.2%) | 46 (4.6%) | 166 (16.5%) | 54 of 62 |
The TikTok row leans on a few creators. When we keep only creators whose first page was enough, TikTok shrinks to 8 creators and 79 videos, with 24.1% at 2x. If you are deciding [how often to post on TikTok](/blog/how-often-should-you-post-on-tiktok), treat that row as a hint.
On YouTube, 19 of 20 channels had a 2x video. A first page of 30 YouTube videos reaches back a median of about 285 days. If you are comparing formats on the platform, [how Shorts and long-form views compare](/blog/youtube-shorts-vs-long-form) covers that. For anyone looking for YouTube outliers, the window length is the first thing to check.
On Instagram Reels, 16 of 23 creators had a 2x video. Reels show a public play count, [photos and carousels do not](/blog/how-instagram-counts-views), so this study reads Reels only.
Do not rank the platforms against each other with this table. The API returns 10 TikTok videos, 30 YouTube videos and 12 Reels per page. A first page covers a median of about 38 days on TikTok, 285 on YouTube and 99 on Reels, and the TikTok second pages (11 of the 19 scored TikTok pages) covered a median of about 11 days. A 10-video TikTok page and a 30-video YouTube page are different questions, so the ordering could be a window artifact. "Views" also means plays on TikTok and Reels and views on YouTube.
As a robustness check, dropping the 13 creators who needed a second page leaves 49 creators and 875 videos, with 17.8% at 2x, 4.1% at 5x and 16.3% at 0.5x or below.
## Why is a 125x outlier not a viral hit?
A ratio tells you how far a video sits above its own account's usual, and a low usual inflates it. The largest TikTok ratio in the sample was 125.7x, on a Tastemade video with 161,422 views. The account's page median was about 1.3k views, so that is 161k views on an account that usually gets about 1.3k.
| Platform | Creator | Ratio | Views | Posted |
|---|---|---|---|---|
| TikTok | [@tastemade](https://www.tiktok.com/@tastemade/video/7681370524655619342) | 125.7x | 161,422 | 3 September 2026 |
| YouTube | [BLACKPINK](https://www.youtube.com/watch?v=CgCVZdcKcqY) | 136.8x | 435,360,201 | 11 July 2025 |
| Instagram Reels | [@khaby00](https://www.instagram.com/reel/DceEuVnii0u/) | 17.5x | 286,037,925 | 25 August 2026 |
The other two rows come with large absolute numbers. BLACKPINK's YouTube page also holds a 30.3x video at 96M views and an 18.3x one, and the Instagram row is 17.5x at 286 million plays. At the other end, the TikTok page for `blackpinkofficial` came back with a median of 277 views, and its ratios are computed against that tiny base, reported as returned.
The same 2x is 20k extra views on one account and 20M on another. Ratio and absolute views answer different questions, so a finder needs both. The ratio is the same idea we used to compare hashtag use against [an account's own median](/blog/do-hashtags-work-on-tiktok), and it has the same weakness when the median is small.
## What is an outlier video, and how is it scored?
An outlier video is one whose views are a multiple of the median of the creator's other recent comparable posts; SocialCrawl reports it as `computed.vs_creator.ratio`.
The baseline is the median of the other comparable posts of the same kind on the same page. Pinned posts and posts under 48 hours old are left out. With fewer than 8 comparable posts, the page baseline is null with the reason `fewer_than_eight`. Each video carries `computed.vs_creator = { ratio, metric, reason }`, where `reason` is null when the video is scored and otherwise `too_young`, `pinned`, `no_peers` or `no_metric`. The page carries `data.creator_baseline` with `baseline.median_views`, `rows_used` and `cadence`.
Here is a real call, trimmed to the first two of ten videos.
```
GET https://www.socialcrawl.dev/v1/tiktok/profile/videos?handle=khaby.lame
x-api-key:
```
```json
{
"data": {
"items": [
{ "post": { "url": "https://www.tiktok.com/@khaby.lame/video/7689510970292489502",
"published_at": "2026-09-25T16:30:18.000Z",
"engagement": { "views": 5980490, "likes": 719826, "comments": 16146, "shares": 41050, "saves": 35242 },
"flags": { "pinned": false } },
"computed": { "vs_creator": { "ratio": 0.43, "metric": "views", "reason": null } } },
{ "post": { "url": "https://www.tiktok.com/@khaby.lame/video/7688786985896381727",
"engagement": { "views": 2262248, "likes": 281865, "comments": 5595 } },
"computed": { "vs_creator": { "ratio": 0.16, "metric": "views", "reason": null } } }
],
"creator_baseline": {
"baseline": { "rows_used": 10, "median_views": 10948982.5, "median_likes": 1063317.5,
"median_comments": 16065.5, "median_gap_hours": 143.93 },
"reason": null, "rows_used": 10, "cadence": "measured" }
},
"credits_used": 1
}
```
Dividing 5,980,490 by the page median of 10,948,982.5 gives 0.55, yet the API reports 0.43. The ratio excludes the post itself, and dividing by the median of the other nine videos reproduces 0.43 and 0.16 exactly. Read the ratio from the field instead of recomputing it from the page median.
Other tools use the same idea with different windows and cut-offs. [vidIQ's outlier score](https://support.vidiq.com/en/articles/9660010-outliers) colours videos 2-5x blue, 5-10x purple and above 10x red, measured against the channel's average views over a similar timeframe. 1of10 describes outliers as 10x the average and does not say which window it uses. We use the median because one spike drags a mean up, which would hide the next spike behind the last one. We did not run a mean-versus-median comparison, so that is reasoning, not a result.
## How do you find a creator's viral videos?
The free viral video finder at [/tools/outlier-video-finder](/tools/outlier-video-finder) applies the same rule as this study. It flags videos at 2x the median or more, leaves out posts under 48 hours old and pinned posts, needs 8 comparable posts, and reads a second page when the first is too thin. On Instagram it also scores photos and carousels by likes, while this study read Reels by views.
For developers, the data comes from three endpoints at 1 credit per page: `tiktok/profile/videos`, `youtube/channel/videos` and `instagram/profile/reels`. Read `computed.vs_creator.ratio` on each video and `data.creator_baseline` on the page, then sort. If you would rather see your data before writing a single line, run the call in the explorer first.
## How did we measure this, and can you reproduce it?
We pulled everything on 1 October 2026 from the SocialCrawl API with caching off, so every harvest call was a live fetch. The endpoints and page depth were `tiktok/profile/videos` (10 videos per page), `youtube/channel/videos` (30) and `instagram/profile/reels` (12), at 1 credit per page. The whole study cost 85 credits in total, 4 of them on earlier probe calls. We also probed Instagram `profile/posts` and dropped it, since its first page for NASA had only 7 comparable posts and it mixes photos scored on likes. Reels give a views-only, video-only comparison.
**Sample.** It was fixed before we saw any result, from one hand-written list. We took 20 non-Korean creators per platform, a mix of mega brands, media, science, food, sport, tech and small niche channels, plus 10 Korean accounts (3 on TikTok, 4 on YouTube, 3 on Instagram). That is 70 attempted creators, and 62 were scored.
| Platform | Attempted | Scored |
|---|---|---|
| TikTok | 23 | 19 |
| YouTube | 24 | 20 |
| Instagram Reels | 23 | 23 |
| Total | 70 | 62 |
**Failures, reported and not replaced.** Four YouTube channels returned an empty page on the first call and on a retry: babishculinaryuniverse, tomscott, kitboga and BANGTANTV. We did not investigate the cause. Four TikTok creators stayed unscored. gordonramsay returned 1 video and ryan.trahan 4, while nba and washingtonpost still had fewer than 8 comparable videos after two pages, because pinned and under-48-hour posts eat most of a 10-video page.
**Page depth.** Everyone got page 1. If page 1 had no baseline, we read one more page. The scored page was page 1 for 49 creators and page 2 for 13.
**Formula.** The API's own `vs_creator.ratio`. A share is the number of videos at a threshold divided by scored videos, meaning those with a non-null ratio on a page that has a baseline. The 1,008 is scored videos, not returned videos.
**Limitations.**
- Not a random sample. It skews to large, active, English-language brand accounts.
- Windows differ by platform (a first page covers a median of about 38, 285 and 99 days on TikTok, YouTube and Reels, and TikTok second pages about 11), so the platform rows are not like-for-like.
- The baseline is per page, not cumulative. Page 2 gets its own baseline.
- A low page median inflates ratios, as the Tastemade example shows.
- "Views" means plays on TikTok and Reels and views on YouTube.
- Videos cluster within creators, and we computed no confidence intervals.
We also computed splits by views tier (9, 15 and 38 creators) and by market (53 English, 9 Korean). With groups that small, and tiers mixed up with platform, we do not report differences. The Korean sample is idol and label channels, not general Korean creators. For a sample-design sibling, see our [engagement rate benchmarks](/blog/social-media-engagement-rate-benchmarks-2026).
To reproduce it, or to check how many views is viral for a creator you follow, call `GET /v1/tiktok/profile/videos?handle=` (or `youtube/channel/videos`, `instagram/profile/reels`) with your SocialCrawl key, read `computed.vs_creator.ratio` on each video and `data.creator_baseline` on the page, and count the videos at 2x or more of that creator's usual.
## Frequently asked questions
### Is 100k views viral?
It depends on the creator. In this sample, 9 of 62 creators had a page median under 100k views, and 38 had a median of 1M or more. For those 38, 100k views is below their usual. The test is whether a video reached 2x its own account's median, and a fixed number does not answer that.
### Is 10k views viral?
The same relative test applies. The 9 creators with a page median under 100k ranged from 277 views (a BLACKPINK TikTok page) to 67,348. On Tastemade's TikTok page the median was 1,308, so 2x is about 2,600 views, and 10k is far past it. On the 67,348 page, 10k is below the usual. We did not measure a threshold for small accounts, because the sample holds only 9 of them. Compare the video with that account's own median.
### Is 2,000 views in one day good?
It depends on the account's usual, and a one-day count cannot be scored against it yet. The API leaves posts under 48 hours old out of the ratio, with the reason `too_young`. Once a video is two days old, compare it with that account's median: 2,000 views is a 2x video for an account whose usual is 1,000 and an ordinary one for an account whose usual is 2,000. We did not measure views per day in this study.
### How many views makes a TikTok viral?
There is no fixed number. On TikTok, 23.8% of 189 videos from 19 creators reached 2x their creator's median and 8.5% reached 5x, and all 19 creators had a 2x video. TikTok pages cover a shorter window than the other platforms, and a low-baseline account like Tastemade can post 161k views and show 125.7x, so the ratio needs the absolute views beside it.
### What is an outlier on YouTube?
An outlier on YouTube is a video whose views are a multiple of the median of the channel's other recent comparable videos, reported as `computed.vs_creator.ratio`. In this sample, 17.4% of 580 YouTube videos reached 2x and 4.1% reached 5x, and 19 of 20 channels had at least one 2x video.
### Is 1000 views on a reel a lot?
It depends on the account's usual. A reel is a lot if it is 2x that account's median plays, and ordinary if it matches it. In this sample, 15.5% of 239 reels reached 2x and 2.5% reached 5x. Photos and carousels show no public view count, so they are outside this study.
### Is there a free viral video finder?
Yes. The [free viral video finder](/tools/outlier-video-finder) flags videos at 2x a creator's median or more, skips pinned posts and posts under 48 hours old, and needs 8 comparable posts. Developers can get the same fields from `tiktok/profile/videos`, `youtube/channel/videos` and `instagram/profile/reels`, at 1 credit per page.