I made 3,457 Shorts to learn this...

Summary

This video script delves into the intricacies of YouTube's algorithm, emphasizing its role in maximizing user engagement and revenue through content recommendations. It explains collaborative filtering as the primary method for short video promotion, using viewer data to match content with interests. The script illustrates the importance of crafting an engaging intro for YouTube shorts to attract the right audience and avoid misleading irrelevant viewers, ensuring the algorithm accurately promotes content. Tips are provided to optimize short intros, including studying successful niche shorts, using text overlays, and adding AI voice intros to repurpose existing content effectively.

Takeaways

• 📈 The YouTube algorithm's primary goal is to keep users on the platform by recommending content they want to watch, thereby generating revenue through user engagement.
• 🔍 Collaborative filtering is a key method used by the algorithm to understand user preferences and match them with suitable content, including YouTube Shorts.
• 🎯 For Shorts, the algorithm tests content with small, diverse audiences to gauge interest and engagement before promoting it more broadly to a relevant audience.
• 🔑 The intro or 'hook' of a YouTube Short is crucial as it needs to quickly and clearly communicate the content's theme to attract the target audience and filter out those who aren't interested.
• 📹 The quality of a video or Short is important, but it's equally important that it resonates with the audience it's intended for to ensure algorithmic promotion.
• 📊 Initially, a Short may not gain many views as the algorithm is learning and testing its appeal with different audiences, but this can change as it gathers positive feedback.
• 📈 If a Short resonates with its intended audience, the algorithm will recognize this and promote the content to a larger audience, potentially leading to a significant increase in views.
• 🎨 To create compelling intros for Shorts, consider mimicking successful formats in your niche, such as using text overlays or showing the main subject matter upfront.
• 📝 Adding text on screen at the beginning of a Short can summarize its content and entice viewers to continue watching, similar to a YouTube video title.
• 🛠️ Repurposing existing content into Shorts can be enhanced with an added intro or an AI voiceover to set the context and attract the right audience from the start.
• 🔬 Analyzing successful Shorts in your niche can provide insights into effective formats and techniques that can be adapted for your own content.

Q & A

What is the primary goal of the YouTube algorithm?

The primary goal of the YouTube algorithm is to make YouTube money by keeping users on the platform for as long as possible by continually recommending them content they want to watch.

How does the YouTube algorithm determine which videos to recommend to viewers?

The algorithm uses a method called collaborative filtering, which looks at the viewing history of users on the platform and then uses that data to figure out what other people would also like to watch.

What is the purpose of collaborative filtering in the context of YouTube Shorts?

Collaborative filtering for YouTube Shorts helps the algorithm to understand the content and promote it to a small, targeted audience to see how they respond, and if the response is positive, it will promote the short to a larger audience.

Why is it important for a YouTube video to have a good intro or hook?

A good intro or hook is important because it not only keeps people watching but also serves the same purpose as a title and thumbnail in filtering out people who aren't interested in the content, ensuring that the video is promoted to the right audience.

How does the algorithm test a new short video to determine its audience?

The algorithm tests a new short video by promoting it to a small portion of various audiences and observing their response. It looks at engagement metrics such as clicks, watch time, and completion rates to determine if the content resonates with the audience.

What happens if a short video is not getting views initially?

If a short video is not getting views initially, the algorithm continues to test it with different audience segments, gathering data and feedback. If it eventually receives positive engagement from a relevant audience, the algorithm may start promoting the video more broadly.

Why is it a problem if a short video's hook is too vague or general?

A vague or general hook can attract viewers who are not the target audience, leading to low engagement and negative signals to the algorithm, which may then decide not to promote the video further due to the assumption that it's not relevant or good content.

What can be done to improve the intro of a short video that is not performing well?

To improve the intro, one can script a clear and enticing intro, analyze successful shorts in the niche for common patterns, add text on screen that summarizes the video's content, or add an AI voice intro to clearly state the video's topic and purpose.

How can analyzing successful shorts in a niche help in creating a better intro?

By identifying common patterns and elements in successful shorts, such as the presence of the UI, text hooks, or the format, one can emulate these successful strategies to create an intro that resonates with viewers and signals the content's relevance.

What is the significance of the 'vibe' of successful shorts in relation to creating a compelling intro?

The 'vibe' of successful shorts, which includes visual elements and presentation style, helps in creating an intro that is instantly recognizable and appealing to the target audience, increasing the likelihood of viewers staying engaged with the content.

Outlines

00:00

🔍 Understanding YouTube Algorithm and Collaborative Filtering

This paragraph explains the YouTube algorithm's role in maximizing platform revenue by retaining users through relevant content recommendations. It introduces the concept of 'collaborative filtering,' a method used by YouTube to match content with viewers' interests based on their viewing history. The speaker uses an illustrative example of categorizing viewers by their interests and how the algorithm tests content with different audiences to gauge engagement. The importance of audience relevance to a video's success is emphasized, as the algorithm's promotion of content is dependent on positive user interactions.

05:02

🎥 The Importance of an Engaging Intro for YouTube Shorts

The speaker discusses the critical role of an engaging introduction in YouTube Shorts, as it serves a dual purpose of attracting viewers and filtering out those not interested in the content. The paragraph uses the analogy of a title and thumbnail for regular videos, which is absent in Shorts, to highlight the need for a compelling start. The speaker shares personal experience with a clip that lacked a clear hook, illustrating the potential issues of vague intros, such as attracting the wrong audience and sending negative signals to the algorithm. The paragraph concludes with examples of view growth patterns for Shorts, showing how the algorithm learns and eventually promotes content to a wider audience once it identifies the right viewer segment.

10:02

🛠 Techniques to Improve Short Intros and Attract Views

The final paragraph offers strategies to enhance the introductory segments of YouTube Shorts. It suggests scripting the first few seconds to clearly convey the content's theme and hook viewers. When scripting isn't possible, the speaker recommends observing successful Shorts in the niche for common patterns, such as using text overlays or subtitles. The paragraph also advises testing text on screen at the beginning to summarize the video's content and entice viewers. Lastly, it proposes adding an AI voice intro to repurpose existing content, providing examples of how this can be effectively done. The speaker encourages applying these techniques to help the algorithm understand the target audience and potentially lead to a surge in views.

Keywords

💡YouTube Algorithm

The YouTube Algorithm refers to the system that YouTube uses to recommend content to its users. It is designed to keep viewers engaged on the platform by suggesting videos that align with their interests. In the video, the algorithm is central to understanding how certain videos gain traction and views, as it uses user engagement data to determine which videos to promote to which audiences.

💡User Engagement

User Engagement refers to the interaction between users and content on a platform. It can include clicks, watches, likes, and comments. In the context of the video, the algorithm uses engagement metrics to assess how viewers respond to content, which in turn influences the visibility and promotion of that content on YouTube.

💡Collaborative Filtering

Collaborative Filtering is a method used by the YouTube algorithm to recommend content based on the viewing patterns of similar users. The script explains that this method categorizes users by their interests and uses this data to predict what other users might enjoy watching. It is particularly relevant for YouTube Shorts, where the algorithm needs to understand and match content with the right audience.

💡YouTube Shorts

YouTube Shorts are short-form videos that are typically under 60 seconds long and are designed to be easily consumable on mobile devices. The video script discusses how the algorithm promotes these shorts to different audiences and how the initial engagement can lead to a significant increase in views, as seen in the examples provided.

💡Intro/Hook

The Intro or Hook in a video is the opening segment designed to immediately capture the viewer's attention and set the context for the content. In the script, it is emphasized that for YouTube Shorts, having a clear and engaging intro is crucial for the algorithm to correctly identify the target audience and for viewers to understand what the short is about.

💡Retention

Retention in the context of video content refers to the extent to which viewers continue watching a video after they have started. High retention rates are a positive signal to the algorithm, indicating that the content is engaging. The video script uses the concept of retention to explain how viewer behavior affects the promotion of content by the algorithm.

💡Target Audience

A Target Audience refers to the specific group of viewers that a piece of content is intended for. The video script discusses the importance of creating content that resonates with the target audience and how the algorithm learns to identify and promote content to the right viewers based on their engagement.

💡Viral Content

Viral Content is media that spreads rapidly and widely across the internet, often resulting in a large number of views in a short period. The video script uses the term to describe the end goal for many content creators, where a video or short gains significant traction and views due to the algorithm's promotion.

💡Relevance

Relevance in the context of the video refers to how well a piece of content matches the interests of the viewers it is being promoted to. The script explains that even high-quality content will not perform well if it is not relevant to the audience it is being shown to, as indicated by their engagement or lack thereof.

💡AI Voice Intro

An AI Voice Intro is a technique where an artificial intelligence generates an introductory voiceover for a video. The script suggests this as a method to enhance the intro of a short, providing a clear and concise summary of the content to engage viewers and guide the algorithm in targeting the right audience.

Highlights

YouTube algorithm's primary goal is to keep users on the platform for as long as possible to generate revenue.

The algorithm recommends content based on user preferences to prevent frustration and maintain viewer engagement.

Collaborative filtering is a key method used by the YouTube algorithm to match content with viewers.

The algorithm tests content with different audiences and measures their responses to optimize recommendations.

A video's success on YouTube is highly dependent on the audience's interest and engagement with the content.

YouTube Shorts lack traditional titles and thumbnails, requiring a strong introductory hook to attract viewers.

The algorithm initially promotes Shorts to a general audience to gather data on viewer preferences.

An engaging intro in a Short is crucial for signaling its content and retaining viewer interest.

Vague or general intros in Shorts can lead to poor viewer retention and negative feedback for the algorithm.

The algorithm may take time to learn and correctly promote a Short to its target audience.

A slow initial growth in views for a Short can be followed by an explosion of views once the algorithm identifies the right audience.

Creating an effective intro for a Short involves scripting, understanding the niche's successful Shorts, and using text or AI voiceovers.

Successful Shorts in a niche often share a similar vibe or format, which can be emulated for better engagement.

Using text on screen at the beginning of a Short can summarize its content and entice viewers to continue watching.

Adding an AI voiceover intro can provide more freedom to explain the Short's content and hook viewers.

The video offers a study analysis of over 5,000 Shorts to understand what keeps viewers engaged until the end.

Transcripts

take a look at some of these shorts this

one 9.9 million views here's another

short 3.7 million here's another short

1.4 million views here's another one 1

million views here's another one 16.2

million views and without doing the

thing I'm going to talk about in this

video these shorts wouldn't have

anywhere near the amount of views that

they currently do but before I can

reveal exactly what I'm talking about

you need to understand a little bit

about how the YouTube algorithm works so

the algorithm's job is to make YouTube

money which it does by trying to keep

users on the platform for as long as

possible and it keeps users on the

platform for as long as possible by

continually recommending them content so

it could be long form videos or YouTube

shorts that they actually want to watch

and so recommending the right videos to

the right viewers is essential in the

algorithm's job description because if

it doesn't do this people are going to

get frustrated because they're not going

to be served videos that they actually

want to watch and so they're going to

leave which is bad because YouTube

doesn't get money but how does this

recommendation system actually know what

each P piece of content is and how to

connect it to the right viewer well

there are a bunch of different ways but

the primary one especially for shorts is

a method called collaborative filtering

it's probably going to be easier for me

to explain collaborative filtering to

you through an example so let me pull up

a whiteboard and actually draw this for

you I want you to imagine that all of

the viewers on YouTube have been

categorized by the type of content they

like to watch their interests so for

example we might have cat lovers over

here we might have fortnite players over

here we might have crime lovers over

here like True Crime maybe we have small

YouTubers over here maybe we've got

business owners over here all these

different audiences you get the idea and

in the middle here is our piece of

content and how collaborative filtering

actually works is it looks at viewing

history of users on the platform and

then uses that data to figure out what

other people would also like to watch

those types of videos so for example

what YouTube might do for this

particular piece of content that we've

created is it might promote it to a

small amount of True Crime lovers it

might promote it to a small amount of

cat lovers it might promote it to a

small amount of small YouTubers and when

it's promoting or rather I should say

testing our content with all the

different audiences what it's looking

for is how do these audiences respond to

your content and by respond I mean do

they click on it do they watch it all

the way to the end do they click on it

and only watch it for a little bit do

they not click on it at all etc etc so

for example let's say that this piece of

content we have in the middle is

actually a video or short about how to

get YouTube views so YouTube's going to

do its best to figure out what this

content is so it can recommend it to

what it thinks are relevant audiences

But ultimately it's just going to test

it with a bunch of different people but

what happens is when it promotes our how

to get YouTube views video to cat lovers

they're not interested when it promotes

out how to get YouTube views to crime

viewers they're not interested fortnite

viewers are not interested business

owners not interested however when it

promotes it to our small YouTuber

audience our small YouTubers want to get

more views and so they respond

positively to this video they click on

it they watch it they engage with it and

the algorithm is like oh there's

something positive going on here and so

what it does is it starts to promote

this video even more to small YouTubers

and to a larger percentage of this

audience and if they continue resonating

with this video it will promote the

video again to an even larger portion of

this audience and that's how a video

gets more and more views so how good a

video is is completely relevant to the

audience who is actually watching that

video Even if we have the best how to

get views video in the world if it's

being watched by people who don't want

to learn how to get views on YouTube

they're not going to click on that video

or they're going to leave that video and

the algorithm is going to kill our video

so now let me clear this up and talk

about how this works in the context of

YouTube shorts now earlier I said to get

some data initially the algorithm is

going to promote OTE your video to

slightly more General audiences and for

long form videos the way it can do this

is just by having this video show up in

the homepage having this video sh been

recommended and then use the title and

thumbnail of that video to work out what

kind of viewer would actually be

interested firstly in just clicking on

this video but for shorts we don't have

a title and thumbnail all we have is a

video usually cuz people are just

scrolling on the YouTube shorts feet and

so the way collaborative filtering works

with shorts is basically you'll create

your how to get views on YouTube short

YouTube's going to push this short out

to again a very small amount of people

in some general niches that it thinks

are related to the content and if one of

these small test audiences when they

recommended the short actually really

watch it and enjoy it the algorithm will

look at that be like hey let's send this

short to more similar users so it's

going to promote it to a larger portion

of this audience and it's getting it

more views and that Short's going to

grow and grow and grow like we talked

about earlier but in order for this

audience to actually watch and engage

with this video this short has to have a

pretty damn engaging intro that screams

to them basically like hey this short is

about how to get views on YouTube so all

the small YouTubers see that and like

I'm going to spend a bunch of time

watching this so basically the beginning

the intro of your short or the hook we

call it not only does it have to keep

people watching but it has to do the

same thing that a title and thumbnail

would do it has to clearly State and

represent what this short is about so

that it filters out people who aren't

going to be interested in this short and

a people who are so let me give you an

example that most of you guys watching

can probably relate to I'm going to play

a clip from a longer piece of content

that I've been considering cutting down

into a YouTube short now give this

example a watch one of the things that I

really struggled with to be honest was

embarrassment I was terrified that my

friends were going to find out about my

channel and be like you know this dude's

just some loser who's desperate for

attention and social validation and

believe it or not that's not the kind of

thing that makes people invite you to

parties now I'm not sure if you could

tell but this clip is basically Al about

how I overcame the embarrassment of not

getting very many views on my videos

when I first started my channel now in

this clip I go on to tell the story of

like how I felt and then share the

mindset shift that helped me get over

that embarrassment I think it's an okay

clip for a short but the problem is our

hook our intro the first like 3 to 5

Seconds they're very vague and general

one of the things that I really

struggled with to be honest was

embarrassment I was terrified that my

friends were going to find out now since

you're watching this video you might

have a rough idea of who I am and so

maybe you would have guessed the context

of this clip but if you didn't have that

context you would have had no idea what

this short is actually about and that's

a massive problem because if we come

back to our whiteboard if I did take

this clip and post it to my channel as a

short one of two things would probably

happen either my hook or the beginning

bit of this video is too vague in

general just doesn't grab people well

enough and so when the algorithm tests

promoting this to lots of people no one

actually watches it in which case it's

just going to die completely or the

other thing it could do is if this Hook

is interesting enough that it catches

people's attention because it's so

General and vague this short is going to

start getting views from cat lovers from

True Crim lovers from fortnite

enthusiasts from business owners and

from our small YouTuber audience who we

actually want to Target and the problem

with this is that as soon as our cat

lovers our true crime lovers our

fortnite lovers Etc discover that oh

this is actually a short about how to

grow on YouTube something that is not

relevant to me at all they're going to

leave that short they're going to swipe

away it's going to kill my retention and

it's going to send really negative

signals to the YouTube algorithm because

the algorithm is going to think well

this short is relevant to cat lovers and

True Crime lovers and fortnite

enthusiasts Etc because they did watch

the first bit of it but then after they

watched the first bit they left

immediately so maybe it's just a bad

piece of content and I shouldn't promote

it to anyone your shorts have to be

decent quality regardless to actually

get people to watch them but if you have

decent shorts and they're not getting

views this is the mistake you're

probably falling into so remember these

examples I I showed you earlier

something interesting you can actually

see if we come down to the Views graph

we can see that for the first period of

this short being live it didn't get very

many views relative to how many views it

ended up with this one only got about

50,000 views after 160 days before

exploding to almost 5 million views this

one grew really slowly as well until

bang we see this explosion here we see

the same pattern with this shot so

what's actually happening here is for

this beginning period where the short

doesn't get very many views the

algorithm is learning what it's doing is

it's promoting this short to a tiny

portion of the small YouTuber audience

and some of those people are clicking

and watching which is a big green tick

but then it goes on and it promotes to

the cat lover audience and they don't

watch they swipe away and see how's like

oh okay is this actually a good video so

it goes back and it promotes it again to

our small YouTuber audience slightly

larger portion they reciprocate they

watch that video the's like oh okay

what's going on here let me test this

again it tests it with the True Crime

audience True Crime doesn't watch cuz

they don't care and so what we can see

here is the algorithm going through this

process of testing promoting our short

to small segments of different audiences

and getting feedback getting data and it

gets enough positive data from our

relevant audiences to tell it hey

there's something here I should keep

promoting this but it's not getting a

clear enough signal to know who to push

this short to on mass until round about

here you can see we go from like 1,000

views in 100 days to getting like 50,000

views and then from here the algorithm

just goes all out and just promotes it

to like 5 million people which is

basically it's promoting to to a smaller

group here to prom to a bigger group a

big group and it realizes hey this video

is meant for small YouTubers and so then

it just promotes it to all small

YouTubers all in one go and that's why

we see this sort of explosion of views

and so if you've got a really good short

it's okay if it doesn't get a huge

amount of views in the beginning but

what's important is that short needs to

have an intro that is really appealing

to your target audience but that also

correctly filters out people who would

not be interested in watching that short

and if you do that correctly the

algorithm will learn and event

eventually you should see an explosion

of views and so to fix this we're going

to make some changes to our short intros

or hooks now obviously the easiest way

to fix this is to just script your short

in a way that the first few seconds of

it really clearly tell people what your

short is going to be about and hooks

them to keep watching it but that's not

always possible going back to our

example here if I'm going to use a clip

from a talk I gave I can't go back in

time and like rescript what I actually

say in the talk so what we can do

instead is use three powerful techniques

the first one is to figure out the Vibe

of other really successful and popular

shorts in your Niche or in very related

niches so what I mean by that for

example if we come and look at shorts

that are about PowerPoint presentations

this is the first second of this short

it's done pretty well 66,000 likes what

we can see is there is a PowerPoint

presentation full screen and there is

text that says don't make slides like

this so immediately I can see basically

you know this is about PowerPoint and

it's probably going to be a tutorial

which I can gather from this text or

here's another one again we can see the

PowerPoint UI and we can see text that

says me before I understood PowerPoint

PowerPoint UI descriptive SL hooky text

another one same thing you get it you

can see the pattern happening here so if

I'm going out to try and create

successful PowerPoint shorts I'd

probably want to start my short by

showing the PowerPoint UI and having

some sort of text hook on screen and

probably a very basic looking slide in

that UI as you can see all of these

examples here they're not showing like

some really fabulous amazing

presentation right off the bat they're

showing something bad so this is clearly

a format that works and if I was to

model it and emulate it for my own

PowerPoint shorts it would give me the

best chance of succeeding because when

people see my PowerPoint tutorial short

they're going to recognize the vibe from

other similar shorts and probably get a

feel as to what that short is actually

about if I was creating a short about

how to grow on YouTube I'd Do a similar

thing I'd go to my competitors who are

creating shorts that get views I would

go through all of their shorts and see

which one which are getting the most

views what the format is how it's

structured here for example if we go to

my friend Ed's shorts we usually see a

face front and center and instead of

like a static hooky piece of text on

screen we just see subtitles that

sharpen change as Ed's talking so if I'm

trying to create shorts about how to

grind YouTube I'd probably want the

beginning of my short to start with me

on screen and then I'd have subtitles

show up as I'm talking and then maybe

start introducing some overlays and

elements like Ed did but let's move on

to our second trick because sometimes

you might be in a situation where your

Niche is doing all kinds of different

types of shorts or maybe you're creating

shorts that are very similar to the vibe

of a lot of the successful shorts in

your space but they're still not getting

results and so I usually try to test

having some sort of text on screen right

at the beginning of your short that sort

of summarizes what the video is about

while also enticing people to continue

watching think about it like you would a

long form like YouTube video title so

coming back to my example here if I was

to turn this into a short maybe I could

have some sort of like text title on

screen that says how to instantly

overcome embarrassment as a small

YouTuber or something like that that

basically encapsulates what's going to

be delivered to them and tells them why

they should keep watching but the last

trick you can use and this is especially

helpful for shorts like this one where

you're just repurposing content that you

already have is to just add an intro to

it so I could actually just record an

intro then use my editing software to

stitch that onto the beginning of this

clip or if you don't have the time to do

that is actually to have have an AI

voice intro you're short before and I

want to show you a couple really cool

examples of that now the winner of my

latest lineup competition I played with

a young fan in Immortal so if I was to

write an AI intro for this particular

clip maybe I could have the AI say

something like how I overcame the

embarrassment of not getting views on my

YouTube videos when I first started so

with this you kind of doing a similar

thing that you did when you were using

text but using an actual intro gives you

a bit more freedom because it lets you

say a bit more cuz text can be a bit

limiting because there's only so much

text you can actually put on a screen

and still keep it readable and so now

you understand how the algorithm works

you can go out there create great intros

for your shorts that are enticing yet

filter out the people who shouldn't be

watching your short from the people who

should all in the first couple of

seconds and I'll do that the algorithm

will learn eventually you'll figure out

exactly who your short is for and if

it's a good short and a good piece of

content that's when your short blows up

and starts getting thousands tens of

thousands hundreds of thousands maybe

even millions of views like these on but

if you're still struggling with like how

do I create good shorts in the first

place I have a video on screen where I

analyzed a study of over 5,000 400

shorts to show exactly what matters most

and what's going to keep people actually

watching your short all the way to the

end after you've hooked them using the

techniques we've shown in this video so

that video is on screen click it now and

I'll see you there