When do views stop coming in? TikTok, Shorts, Reels and Facebook
Our median TikTok post had 96.1% of its views by day two and Facebook 99.2%. Reels took 2 to 4 days. When views stop coming in, app by app.
By Abondy ResearchPublished Updated 10 min read
In short: Most of a post's views arrive in its first day or two. Across our posts with 100 or more views, a reading taken 6 to 24 hours after posting showed a median of 88.1% of the post's current views, and a reading at 1 to 2 days showed 97.6%. TikTok (96.1% at 1 to 2 days) and Facebook (99.2%) were close to done within two days, while Instagram Reels only caught up at 2 to 4 days, possibly partly because of how we read its counter.
This post answers a narrower question than how many views a short video gets. It's not about how many views, but when they arrive, so you know when to judge a post and when to stop waiting. The view counts behind it are the same ones we publish each month in the views report.
Timing matters because a count read too early understates the post. To estimate views for a planned number of posts, use the views calculator.
How we measured it
Abondy reads each post's public view count several times after it goes up and stores the count each time it has risen. By November 17, 2026 that was 230,710 stored counts of 40,237 posts, a median of 3 a post, counting the 0 we store when each post is recorded.
For this post we kept posts that were at least 31 days old on November 17, 2026 and had 100 or more views by then. For each reading of those posts, we worked out what share of the post's current views it had already reached. The table shows the median share for readings taken at each age.
| Reading taken | All apps | TikTok | YouTube Shorts | Instagram Reels | |
|---|---|---|---|---|---|
| 0 to 6 hours | 35.0% | 37.7% | no readings | 40.1% | 13.3% |
| 6 to 24 hours | 88.1% | 85.2% | 1 reading | 45.4% | 96.6% |
| 1 to 2 days | 97.6% | 96.1% | 92.6% | 55.4% | 99.2% |
| 2 to 4 days | 97.7% | 95.9% | 97.7% | 99.5% | 99.4% |
| 4 to 7 days | 99.1% | 97.6% | 99.8% | 99.5% | 98.9% |
| 7 to 14 days | 99.6% | 98.2% | 99.9% | 100% | 99.9% |
| 14 to 30 days | 99.8% | 99.3% | 99.9% | 100% | 100% |
| 30 days or more | 100% | 100% | 100% | 100% | 100% |
Read a cell like this: a typical reading taken at that age showed that share of the views the post has now. Across all apps, a reading at 1 to 2 days showed 97.6%, so typically only the last 2.4% (100 − 97.6) came after that.
Two cautions. First, the rows hold different posts, because readings aren't evenly timed. That's why a few shares dip slightly from one row to the next, such as TikTok's 96.1% then 95.9%. Treat small dips as noise between samples. Second, the first six hours are thin. The 0 to 6 hour row only counts readings that found views: 35.0% across all apps. Our first version also counted the 0 we store the moment a post is recorded, which isn't a reading at all. Those zeros were half of that row and pulled its median down to 0.1%. A reading that finds no views yet isn't stored either, so even 35.0% likely runs high.
The first day is most of it, but not for every post
The 6 to 24 hour row is where the curve does most of its work. On TikTok the median reading there showed 85.2%. The spread was wide, though. A quarter of TikTok readings at that age showed less than 62.4% of the post's current views, and a quarter showed more than 95.7%.
So a one-day check gets the typical post roughly right and some posts quite wrong. A slow starter can look weaker than it will be, so don't drop a version on its first-day number alone.
App by app
TikTok: fast, with the longest tail
In the first six hours, TikTok readings that found views showed a median of 37.7% of the post's current views. By 6 to 24 hours it was 85.2%, and by 1 to 2 days 96.1%.
After that TikTok crept up more slowly than the other apps: 97.6% at 4 to 7 days, 98.2% at 7 to 14 days and 99.3% at 14 to 30 days. It was the only app still below 99% in the 7 to 14 day readings. For the median TikTok post, about 1.8% of its views (100 − 98.2) came after that point. That's small, but it's the slowest finish of the four.
Facebook: the fastest
Facebook was the quickest to settle. A reading at 6 to 24 hours showed 96.6%, and at 1 to 2 days 99.2%. In our data, a Facebook post's count after one day was already close to its count a month later.
YouTube Shorts: we only see it from day one
We have no readings of Shorts in the first six hours and only one at 6 to 24 hours. That isn't because we read Shorts late: on recent Shorts, our first count above zero came a median of 3 hours after posting. It's because YouTube's rules let us keep a Short's stored readings for 30 days only, and every post in this curve was at least 31 days old, so most early Shorts readings had already been deleted. So we can't say how quickly a Short's first day goes, and the Shorts column rests on the readings that were kept. From there the curve was steady: 92.6% at 1 to 2 days, 97.7% at 2 to 4 days and 99.8% at 4 to 7 days.
Shorts also tend to land in one of two groups, pushed or not, which is covered in TikTok vs YouTube Shorts vs Instagram Reels. The timing curve here only covers Shorts that reached 100 views.
Instagram Reels: slow, partly because of how we read it
Reels looked slowest. Readings that found views in the first six hours showed a median of 40.1%, then 45.4% at 6 to 24 hours and 55.4% at 1 to 2 days, then the median jumped to 99.5% at 2 to 4 days. At 6 to 24 hours the spread ran from 36.7% (25th percentile) to 62.5% (75th).
A jump that sharp is hard to take at face value. Read the Reels numbers as a floor. Some of our Instagram counts come from each post's public page, and in a 16 Nov 2026 check of 37 posts that page showed about a third of the real count. So part of the slow start is likely our reading of the counter, not Instagram. Treat early Reels numbers as low.
When to judge a post
Based on these readings, this is when the median post with 100 or more views had nearly all of its views:
| App | Judge after | Median share at that point |
|---|---|---|
| 1 to 2 days | 96.6% at 6 to 24 hours, 99.2% at 1 to 2 days | |
| TikTok | 2 days | 96.1% at 1 to 2 days |
| YouTube Shorts | 3 to 4 days | 97.7% at 2 to 4 days |
| Instagram Reels | 4 days | 99.5% at 2 to 4 days |
These are starting points, not rules. Your posts may build faster or slower, and the Reels row partly reflects how we read Instagram.
When to stop waiting
By the 4 to 7 day readings, the median share across all apps was 99.1%. By 7 to 14 days it was 99.6%. For the typical post with 100 or more views, waiting past a week changed the number very little. That fits the main sample in our other posts, which only counts posts at least 7 days old.
Two limits apply. This curve only covers posts that reached 100 views, so it says nothing about a post sitting at 0. Whether a zero ever picks up is a different question, covered in why TikTok posts get 0 views. And "current views" means the count at our November 17 pull. A post that's still growing slowly would make the earlier shares look higher than they'll turn out.
What this means for a brand reading results
- Read the first report at day two, not hour two. Even the readings in the first six hours that found views showed a median of only 35.0% of a post's views across all apps.
- Pick winners on TikTok after two days. If you're choosing which versions to post more of, two days gave the median TikTok post 96.1% of its views. Allow for the slow quarter: one TikTok reading in four at 6 to 24 hours showed less than 62.4%.
- Don't write off a Reel on day one. A Reel read at 6 to 24 hours showed a median of 45.4% of its views.
- Use a week-old number for final comparisons. In the 7 to 14 day readings the median share was above 99% on every app except TikTok (98.2%).
On Abondy, a brand sees each post's views on its dashboard as these readings come in, and the brand pays per post rather than per view. UGC distribution explains how real people post a brand's videos on their own accounts.
How we counted
- Updated November 21, 2026: an earlier version counted posts whose views we never managed to read as 0 views. They're now left out of the zero share in the last answer below (9.7% of posts in the main sample). None of them was in the timing curve, which only covers posts with 100 or more views. The same check found that the 0 to 6 hour row counted the 0 we store when a post is recorded as a reading, and that our reason for the missing early Shorts readings was wrong. Both are fixed above.
- Stored counts: 230,710 of 40,237 posts, a median of 3 a post, as of November 17, 2026, counting the 0 stored when each post is recorded. A reading is stored only when the count has risen. Readings aren't evenly timed, so different posts sit in different rows. Not every post is tracked through every row.
- Curve sample: posts at least 31 days old on November 17, 2026 that had 100 or more views at that pull. That's 75,922 readings in all (8,153 + 21,198 + 12,971 + 8,908 + 5,432 + 7,946 + 8,024 + 3,290).
- Readings per row, all apps (posts in brackets): 0 to 6 hours 8,153 (4,658), 6 to 24 hours 21,198 (6,539), 1 to 2 days 12,971 (5,103), 2 to 4 days 8,908 (4,115), 4 to 7 days 5,432 (2,870), 7 to 14 days 7,946 (4,561), 14 to 30 days 8,024 (4,307), 30 days or more 3,290 (1,881).
- Current views: each post's count at the November 17, 2026 pull. A post still growing slowly would make earlier shares look higher.
- Early readings: the 0 to 6 hour row was worked out again on November 21 from the same stored counts (those stored before the November 17 pull), leaving out the 0 stored when a post is recorded. Readings that find no views aren't stored, so that row likely runs high. YouTube's rules make us delete a Short's stored readings after 30 days, so few early Shorts readings remain for posts this old.
- Campaigns: the posts in the curve come from 4 campaigns. The largest made 79.3% of them, and Abondy's own campaigns 18.4%.
- Counts only: this measures how much of a post's current count it had at each reading. It isn't built to show whether a count ever goes down, so we don't draw any conclusion about that.
- Exclusions: Abondy's own test accounts and posts whose views were confirmed as bought. Only posts that passed Abondy's check and were still up count.
- Who posted: real people on mostly small accounts, posting mostly Roblox game videos.
- Associations, not proof: this describes when views showed up on our posts. It doesn't show what made them arrive faster or slower.
- Instagram: read the Reels numbers as a floor. Some counts come from each post's public page, and in a 16 Nov 2026 check of 37 posts that page showed about a third of the real count, which likely shapes the slow start.
Questions people ask
When do TikTok views stop coming in?
In our data, mostly within two days. The median TikTok post with 100 or more views had 85.2% of its current views by a reading at 6 to 24 hours and 96.1% at 1 to 2 days. A small tail kept coming for weeks: 98.2% at 7 to 14 days and 99.3% at 14 to 30 days.
How long should I wait before judging a post?
About two days on TikTok and Facebook, three to four on YouTube Shorts and four on Instagram Reels, based on our readings. For a final comparison between posts, a week is safer.
Why do Reels views seem to come in later?
In our readings a Reel showed a median of 45.4% of its views at 6 to 24 hours and 55.4% at 1 to 2 days, then 99.5% at 2 to 4 days. Part of that is likely how we read Instagram's counter. So we'd wait longer before judging a Reel, but we wouldn't conclude that Instagram itself is slow.
Can a post with no views take off later?
This data can't tell you. The curve only covers posts that reached 100 views, so posts stuck at zero aren't in it. Every post in our main data was at least a week old, and 10.6% of the ones we could read still showed 0 views. For another 9.7% we never got a reading at all, so we don't count those as zeros.
Read next
- Short-video data · 17 Nov 2026 · 8 min read
How many views does a short video get? Data from 38,486 posts
The median short video in our data got 61 views and the average 517. The full spread of 38,486 posts on TikTok, Shorts, Reels and Facebook.
- Short-video data · 17 Nov 2026 · 10 min read
Why TikTok posts get 0 views (15% of ours did)
14.7% of the 8,070 TikTok posts we could read showed 0 views after a week. What the zeros were linked to, what TikTok says, and what to do about it.
- Short-video data · 17 Nov 2026 · 8 min read
TikTok vs YouTube Shorts vs Instagram Reels: views per post compared
35.3% of our YouTube Shorts passed 1,000 views vs about 1% on TikTok and Reels, but Shorts had the lowest median. Views per post across 38,486 posts.
