The Entire Shorts ALGORITHM Explained 2026

[00:00] How does the YouTube Shorts algorithm

[00:02] exactly work? Why do Shorts get a spike,

[00:04] but then the graph becomes flat? How

[00:06] will the algorithm decide whether your

[00:08] Shorts will get views in millions or

[00:10] thousands? The answer to all these

[00:12] questions will be found in this video.

[00:14] Let's begin. So, we know that big

[00:16] YouTubers keep getting views on their

[00:18] Shorts, but in your case, the views are

[00:20] 100 or 1,000. But after that, the graph

[00:23] becomes completely flat. So, let's

[00:25] understand why this happens. Todd

[00:27] Sherman, who is the product lead of

[00:29] YouTube Shorts, which obviously means

[00:31] that he can tell us a lot about the

[00:32] Shorts algorithm. In an interview, he

[00:35] told that whenever you upload a Short,

[00:37] the algorithm finds a random seed

[00:39] audience for a Short. This seed audience

[00:41] is basically a small group of viewers

[00:44] with whom the algorithm wants to test

[00:45] your Shorts. Because the algorithm

[00:47] doesn't know much about your channel and

[00:49] its content yet. The word random is

[00:51] important here, because it's possible

[00:53] that the seed audience isn't your target

[00:55] audience. The algorithm randomly

[00:57] selected these audiences to test your

[00:59] Short. So, let's say that seed audience

[01:01] doesn't engage properly with your Short.

[01:03] So, after a while, YouTube stops pushing

[01:05] your content to audience, which means

[01:07] there will be almost no views. Let's

[01:10] assume that by chance that seed audience

[01:12] is your actual target audience. And if

[01:15] your content is good, then naturally

[01:18] viewers will engage well with your

[01:19] Shorts. This is the positive signal for

[01:21] the algorithm. After this, the algorithm

[01:24] finds more viewers that match the seed

[01:26] audience. If this new group of viewers

[01:27] also engage properly with your Short,

[01:30] then your Short will be pushed even

[01:31] more, and this will keep on repeating.

[01:33] So, now you must be understanding how

[01:35] the algorithm of Shorts exactly works.

[01:38] Almost every Short creator has a

[01:40] question in their mind that if our Short

[01:42] was performing well in the initial

[01:43] phase, then why did the algorithm stop

[01:45] its views? You have to understand that

[01:47] the algorithm of long video and short

[01:49] video is not the same and can never be

[01:52] the same. In case of long-form video, a

[01:54] view means that the viewer is

[01:55] intentionally choosing to watch the

[01:57] video. However, this is not the case

[01:59] with shorts because here there is no

[02:01] concept of thumbnail. Here the viewer

[02:02] watches the video through the short

[02:04] feed. This means that if the algorithm

[02:06] wants to know if your video is good or

[02:08] not, then the algorithm has only one

[02:09] option. That is testing your videos to

[02:11] seed audience. The next question that

[02:13] will come to your mind will be what are

[02:15] the important matrices which algorithm

[02:18] decided that your short has performed

[02:20] well and has been shown to other people

[02:22] besides the seed audience. The first

[02:24] metric, how many choose to view? Just

[02:26] like in long video, we have CTR.

[02:29] In the shorts, we have how many choose

[02:31] to view metric, also known as viewed

[02:34] versus swiped away. This metric tells

[02:36] you how many percentage of viewers

[02:38] intentionally stopped to watch to your

[02:39] short and how many swiped it without

[02:42] watching the video. The percentage of

[02:43] viewers who stopped intentionally to

[02:45] watch your video, only their views will

[02:47] count. Your next question will be how

[02:49] much should this viewed versus swiped

[02:51] away percentage be? So, you must know

[02:53] Paddy Galloway. He's a YouTube

[02:55] strategist and he has worked with a lot

[02:57] of big creators including Mr. Beast. So,

[03:00] last year Paddy and his team conducted a

[03:02] study where they studied more than 5,000

[03:05] shorts across 33 YouTube channels in

[03:07] different niches. According to his

[03:09] study, shots whose viewed percentage was

[03:12] 60% and less than that, those shots

[03:14] didn't perform as well. The best

[03:17] performing shorts which went viral,

[03:19] their viewed percentage was on an

[03:20] average between 70 to 90%. Again, this

[03:24] is just a study. So, I recommend you to

[03:26] not obsess over these numbers. But yes,

[03:28] you can take a rough idea from this that

[03:31] if your viewed percentage is 80% or more

[03:33] than that, then there's a high chance

[03:35] that your short will go viral. Even I

[03:37] personally, all the viral shots I've

[03:39] seen, their viewed percentage is above

[03:41] 75%.

[03:43] Now, I know that the maximum creators

[03:45] who are watching this video, their

[03:46] viewed versus swiped away ratio is not

[03:48] that good. So, now let's understand how

[03:51] to improve this number.

[03:53] So, as I said, in the case of long

[03:54] videos, we have CTR. Similarly, in the

[03:57] case of shorts, CTR is equivalent to how

[04:00] many choose to view. If we want to

[04:02] improve CTR, we work on the title and

[04:04] thumbnail of the video. That is, the

[04:06] thing that the viewers see first. So, if

[04:09] we talk about shorts, then which is the

[04:11] thing that the viewers see first? You're

[04:13] right, the first 3 seconds. So, it's not

[04:16] wrong to say that the first few seconds

[04:18] of your short, that is, its hook, plays

[04:20] the same role as in the case of long

[04:22] videos of the title and thumbnail. Now,

[04:24] in a long video, we have 30 seconds to

[04:26] hook the viewer. In shorts, we have only

[04:29] 3 seconds. This is why in both these

[04:32] formats, creating a hook is very

[04:34] different. But even though long-form and

[04:36] shorts need different types of hooks,

[04:38] the foundation of both is the same. That

[04:41] is, you have to spark curiosity in the

[04:43] viewer's mind in some form.

[04:45] You can create this curiosity with a lot

[04:47] of things. For example, depending on the

[04:50] type of video you're creating, you can

[04:51] say some lines in your hook, which will

[04:53] immediately stop any viewer who is

[04:55] mindlessly scrolling. For example, in

[04:57] the start of the short, you can say,

[04:59] "I'm sure you don't know about ABC."

[05:02] Here, simply replace ABC with something

[05:04] about which you have a video. Example,

[05:07] "I'm sure you don't know about this

[05:09] YouTube setting. I'm sure you don't know

[05:10] about this recipe. I'm sure you don't

[05:12] know about this new Minecraft update."

[05:14] And this will work as a very good hook.

[05:17] But it's not necessary that you say

[05:19] something in your hook. Because the more

[05:21] powerful a line is to grab the viewer's

[05:23] attention, the more powerful and

[05:25] intriguing visual. Let's take a look on

[05:27] Mr. Beast's shorts. Can you slice a

[05:29] bullet with a katana? Yes, sir. In this

[05:31] short, Mr. Beast shows the visual of the

[05:33] first 3 second, in which bullet almost

[05:36] hit the katana.

[05:37] But just before the climax, that scene

[05:39] has been cut so that viewers can watch

[05:41] the video till the end. But if you

[05:43] really want to make your shorts viral,

[05:45] then just creating a hook is not enough.

[05:47] You also need a good audience retention.

[05:50] When we make long videos, the focus of

[05:52] YouTube is on watch time. Because the

[05:54] algorithm wants your content to keep the

[05:56] viewer on the platform for as long as

[05:58] possible. That's why if a video is 1

[06:00] hour long and an average viewer is

[06:02] watching it for just 15 minutes, even if

[06:05] its retention is bad, the algorithm will

[06:07] promote that video a lot because it

[06:09] keeps the viewer on YouTube for 15

[06:11] minutes.

[06:12] But in the case of shorts, it's

[06:13] different. Here, YouTube doesn't care

[06:16] about AVD. Instead, the algorithm looks

[06:19] at audience retention. If the audience

[06:21] retention is 25%, then it won't work.

[06:24] The best-performing shorts have a 90% or

[06:27] 100% retention, or even more if the

[06:30] viewer watches the short again. But it's

[06:32] not like the shorts algorithm doesn't

[06:34] care about AVD. According to a study by

[06:36] Paddy Galloway, shorts with more AVD

[06:39] performed better as compared to others

[06:41] with lower AVD.

[06:43] So, keep this in mind if you're thinking

[06:45] of making 5-second shorts to get 100%

[06:48] retention and go viral. I'm sure after

[06:50] hearing all this, you're confused as to

[06:52] what's the best length for shorts.

[06:54] Should we make 10- to 15-second shorts

[06:56] to get good retention? Or according to

[06:58] Paddy's study, should we make long

[07:00] shorts? I'll give you the answer that

[07:02] YouTube employees gave, and that is make

[07:05] a short that aligns with your story.

[07:08] Some niches will have quick, instantly

[07:10] rewarding shorts, while in other niches,

[07:12] you can use storytelling to build a

[07:14] narrative arc with a beginning, middle,

[07:16] and end, and you can make longer shorts.

[07:19] So, yeah, the best length for YouTube

[07:21] shorts is in which you can perfectly

[07:23] deliver your content.