Transcripts
1. Introduction: Class, I want to break
down exactly how the YouTube algorithm
operates and works in regards to pushing
out your content and videos to viewers on
the YouTube platform. On YouTube, I have over
500,000 subscribers on my main channel and
over 500 million views across the Internet. And over the last few years of being a full time
content creator, I have learned quite a lot when it comes to viral marketing, creating content that actually gets millions and
millions of views, and also the intricacies and details of how the
platforms operate and the types of metrics you
need to target in order to tick all of the boxes for
those videos to actually out. So if you would
like to learn more about my exact approach to creating
successful YouTube videos, Johny here for this
Skillshare class.
2. What is the YouTube Algorithm: YouTube algorithm is one of the most misunderstood elements
of the YouTube platform. Inside of this
class, I'm going to explain and demystify how the YouTube algorithm
actually works and how you can get your
videos to perform better, get more views, and
actually grow your channel. There are a ton of videos on the YouTube
platform that explain the YouTube algorithm
and how it works and all the different elements that you need to understand. But a lot of them don't actually understand what
the YouTube algorithm is. The YouTube algorithm
to put it incredibly simply is just a
recommendation system. It should really
be referred to as the YouTube
recommendation system. All the algorithm does is it
is designed to just serve viewers users of the
YouTube platform with videos that are
relevant to them. There's no big hacks of how to get the algorithm to do this, and that and the algorithm
destroys smaller channels. I punishes smaller channels. It's all absolute rubbish, some of the stuff
that's actually out there on the Internet. And I'm going to break
down the various elements of a video that you
can control that will increase your likelihood of being picked up
by the algorithm. So when you're browsing
on YouTube, for example, we are here on the home page, YouTube is going to present to you videos that it
thinks you will like. And all of this
recommendation is based off of previous activity and
your habits on YouTube. So stuff you've
watched in the past, things you've watched for
a very long time, like, you've got a high
engagement rate on there, you've
maybe left alike. You've maybe commented a
lot on particular content. In your YouTube account, you'll get recommended content that YouTube will think, right. Ben really likes watching
this type of content, so recommending
some gaming stuff, so he'll stay and watch
that video tonight. Or after that, he might watch some camera stuff because he's obsessed with YouTube
camera things. And then you can see all of the other stuff up here
that it thinks I like. It thinks I like video
editing, obvious, 'cause I create YouTube
videos, formula one, supercar, sort of other things, video games, cameras, et cetera. So YouTube constantly
is profiling its users. So when you create a
Google account and you start watching the videos
on YouTube, instantly, it starts collecting
all these meta data off of its YouTube users, and it starts profiling
them. So it goes, right. Ben, is aged 22. He is based in the UK. He likes to watch videos
of Supercars, formula one. He loves video games. In particular, he really likes, you know, battlefield,
caller Duty, He likes consumer electronics, so we'll show him
iPhones, iPads. He loves TVs, because
he likes gaming, so we'll show him
gaming specific TVs. It's incredibly intelligent. It's constantly collecting
all of this data. And as you grow as a human
being, it will then go, Oh, Ben now is
interested in this, because he's matured
a little bit. He doesn't play
Minecraft anymore. He now plays whatever. So it's constantly
evolving across all of the decades that its users
have been on the platform. And then from all of
this information, YouTube will then think,
who else is like this user. So 22 based in the UK,
likes video games. So he uses all of the data of 22-year-old males in
the UK, and it figures similar trends that these
guys are interested in. So they really like video games, they're like, sports, fitness, all these different
types of things. So YouTube has
this extra data in its algorithm or
recommendation system of other 22 year olds in the UK. And that's why sometimes you get a great recommendation out of nowhere because you might
just be browsing on YouTube. Looking for gaming content
because that's what you like. And then suddenly out of
nowhere, YouTube's like, you're going to love
this video because other 22 year olds in the
UK and 22 year olds in the USA really
enjoyed this video on personal fitness to go get ripped with
that in your thing, and you'll click on it and go,
Yeah, this video is great. So in really basic terms, that's what the YouTube
algorithm is doing. It is designed by Google
to recommend content to its platform users to in
turn keep them on the app, because the more time
on the app that they have on Google on
the YouTube website better the Google is is the
business, to make more money. You're watching more
adverts, you're spending money on
YouTube premium. So much stuff going
on, there collecting all the data off of you that they can then go and
sell or whatever, and they can sell
advertisements. It's huge money make machines. So it's in their
interest to design an algorithm that
recommends the best of the best content on the YouTube platform to keep the YouTube
platform relevant. A huge problem is, it's
all of this fake advice, stuff like this, so it's like 1 million plus views
every time you post? It sets unrealistic expectations
for small YouTubers, where they watch this video and they think, I'm doing
something wrong, and it gives them a
huge mental block when they're trying
to create content. They feel like they should
be achieving stuff cause some guy shows them how to
get rich quick on YouTube. And this class is going
to demystify all of that rubbish and really
help you understand how is a YouTube
creator that wants to have a proper
longevity and build a proper channel can
understand how to create and craft
incredible videos that the algorithm loves.
3. Why a Video Get’s Views: Things off. I want
to prove to you that the YouTube algorithm or the recommendation
system isn't designed to favor larger
creators and keep the smaller creators small and punish them for being
a smaller channel. All the things that people have maybe told you or
you maybe convince yourself are true because you're struggling to grow your
channel and get views. Now, there are key
metrics that you need to really hone
in on and focus on and increasing as high
as possible in order to ride the wave of the
Tube recommendation system. You almost need to think of
YouTube as your employer. They wouldn't hire
a small channel who hasn't got good
recommendations. You know, Your metrics are low. You've got low watch retention. You've got a rubbish
click through rate. You thumbnails out very good. Like a normal employer
wouldn't hire someone with rubbish
recommendations from past people who have
worked for them. If you're hiring
someone as a manager, and you're looking at the
metrics and the data, and you go, ah, they
showed up for work late. They're very inconsistent,
didn't plod very frequently. They didn't commit
to the content much. No many people really
watch that content. Your YouTube's not going to hire you as the channel to
give you an opportunity. Not going to give
you an opportunity because your data is rubbish. So you need to focus
on these key metrics. So let's take a look
at my channel here. This is my channel,
Ben Rowland's tech. I've got multiple channels, got 70,000 18,000 subscribers. And this channel, in particular, in one year has amassed
almost 4.5 million views. Pretty pretty huge achievement
for one year on YouTube. Now, if we take a look
at my oldest videos, I want to prove to you
that it's rubbish that small channels can't ride
the algorithm or grow. You can see right here
looking at my videos. First few ones performed, all right, third
ones done all right. The other ones
underperformed performed. But my sixth video is that. Almost 300,000 views. But that's not where the
interesting stuff lies. My tenth video. I haven't deleted any videos
on this channel. My tenth video 610,000
views in 11 months. And this success is all down to understanding how to provide YouTube great data so they think this person is worth
giving an opportunity, almost like a
promotion in a job. They're a small YouTuber,
but let's give them an opportunity in
the Big Leagues. So what are these metrics? These magical metrics
that I'm talking about? The first one is average
percentage view, to your watch percentage view. This is probably the
most important metric that you can obsess
about as a YouTuber. I can guarantee probably 95% of the people watching the
Skillshare class probably have maybe 25-35%
average view percentage. That's what you're getting. You maybe get about 30 to
35% tops on your videos. You videos maybe maybe 8
minutes long or whatever. Not terrible, but definitely
room for improvements. So according to YouTube, and this is data directly
from a YouTube employee, these are the top
10% videos versus the average median
for YouTube creators. So let's say you're creating ten minute videos on YouTube. For the top 10% creators, the top 10% videos
on the platform, for 10 minutes long, they were getting 61% percentage viewed. The guys that were medium average just you sort of
still performing well, but they were in that
sort of middle area. For a ten minute video, they are lucky to
get 44% viewed. Still fantastic, but
it's a huge difference. Almost 20% difference between
the top 10% and the median. If we then compare this to six
minute videos, on average, the median is getting around
50% watch percentage, and the top 10% is are
having people watch their videos and have almost
a 71% completion rate. So out of the people that
click on that video, basically 71% of them are
completing that video, which is incredibly high. Now, for me, at the moment, my channel is a median channel. I create videos 3-6 minutes, and we are getting
between 50 65% viewed. Now, there's still a huge
room for improvement. If I want to be a top percenter, I should really be getting
70-82% watched on my videos. And I do have a
few that have got maybe 68% watch attention, like around six minute
video, which is fantastic. I nailed it, but it is very difficult to
getting that top 1%. I'd be trying crazy
to try and do that. And there's a few extra
things that I could try, but it's hard. So I highly recommend If
you're small YouTube, focus on this median. How can you get to
be in that median, get your videos from 35% to 44%? Cause that is going
to be pivotal, which I'll explain now. But first, let's discuss
a few other metrics that contribute to your
success on the algorithm. One of the most important
ones is obviously, as we've established,
the average percentage, the more time people spend
watching your video, the longer you're keeping
them on the YouTube platform. So YouTube likes
that. They think, Well, Ben's keeping
people on for, you know, basically 2
minutes and 10 seconds out of this three
minute, 50 video? I like that. So they're
going to push that out. I like that a lot.
They're enjoying Now, other key elements are your click through rate
in your impressions. You see here, this video has had almost 5 million
impressions. So 5 million people
on YouTube had this video pop up
on their newsfeed or whatever when
they're scrolling. And around 8.9% of those have
actually clicked through. If we went to the
advanced mode and we took a look at the actual
click through rate, Traffic sources here. You can see in YouTube search, it's got a 12% click through
rate and in YouTube brows. It's got a 9% click through
rate in the last 28 days. Since lifetime, it's, yeah, about 10%, 10% click through rate in both
of those things. So the higher you can get
this click through rate, again, this is
another metric that the YouTube algorithm
likes. It goes right. Ben has got this great video
that's getting people to watch for about 60% of
its duration, like that. And then it goes right. He's also getting 10% of
people to click through. That's a great conversion
rate coming through. Now, ideally, you want
to maybe get 15% or 16% people clicking through in your sort of first few months of the
video being a live. The longer the videos
on the platform, the lower the click
through rate becomes over time because it gets shown
to more and more people, and YouTube just
keeps pushing it out to a less relevant audience. So, the fact that this
video after a year, and 5 million people have
seen it still has basically a 10% click through rate is
quite difficult to achieve. Because traditionally, the more the content
gets pushed out, the more the click
through rate decreases, because it's going to
less relevant people that aren't really
going to watch it. So those are the two key metrics that you
want to focus on, your average percentage viewed, and your click through rate. The higher you can get
the click through rate, it shows it's interesting
and exciting content. It's got a good
title and thumbnail. So YouTube thinks quality. The longer you can keep people
on the platform watching with your average view
percentage viewed, it's pivotal. What YouTube then does with this data is very interesting. It will compare the data against other channels that have
created similar videos. So, for example,
this video is about a paper like screen
protector for your iPad. So YouTube will go,
what other channels have created reviews of a
paper like screen protector, and what was their data? So YouTube will
then take this data and compare it almost
like a league table. So you'll have Ben's channel, and he's got a 55%
watch retention on this video category. Then you've got C channel B. And also channel C, who
have maybe got, you know, 35% watch retention, and this person's maybe got a
44% watch retention. Then what YouTube
will do is right. It will go, Let's
rank these videos as which is the most
important one to promote. So clearly, it's Ben's video. Ben's video is
keeping everyone on the platform the mo
versus channel B. So Channel B on the algorithm will not get
pushed out because they will think this video is not
good enough to get promoted because it's not got a good
enough watch percentage. And then it will give Channel
C the second opportunity. What will then happen is when you're watching
your videos, you will see Ben's video first, And then it may recommend
Channel C as an up next, so it might say this
one as an up next, because it's not as
good as Ben's video, but it's good enough
to also keep people on the platform because it's
better than C channel B. Channel B hasn't
worked hard enough, and they haven't created
a good enough video to compete with these
other channels. So the algorithm is more about your videos data and how that is ranking
on YouTube system. So they'll then promote
the videos that have got the best watch
percentage that have got the best click through rates,
and all these other things. And all these factors
combined together, influence whether your video
basically flies or dies. The same is also true with
the click through rate. So an interesting thing
that I've often seen, and I see this loads with
Apple products because Apple is so competitive
on YouTube, and everyone's making
videos about Apple iPads. Apple this, Apple that. It's
you uploaded video first, so I'm Ben, and I've
uploaded my review first. So this is my review of
an iPhone or whatever. So I've uploaded my video
first before other channels. If a bigger channel
then comes along, a bigger channel shows
up 48 hours later, and they upload their review. What I've also noticed
happens with these metrics, even though I can
guarantee my video has got the higher watch percentage
just by judging them and my experience of YouTube
and graphs and things that. I can guarantee my video has got the better
watch percentage. What I have noticed
is the larger channel may have a much better
click through rate. So their CTR on the video
may be much higher than me. So I may only be getting 8.9% of people clicking
onto the video. Was they are getting 18% because they've got
quite a loyal audience, because they've
got maybe 120,000 followers or 200,000 followers. Because they've got that
loyal audience that sat there waiting for
their next upload, they get that huge
initial bulk of traffic, which then skyrockets the
video on the algorithm, because YouTube again sees
this and thinks, Wow, that's creating a lot
of interest versus Ben tiny little
click through rate. However, over time, my video may perform
better because after they've had that
initial boost of that huge CTR and that huge click throu rate
on the thumbnail title, because that dries up after 48 hours because the
audience has seen it, then my video ticks away in the background because it's
actually the better video. And this is where that
misconception comes where small YouTubers
blame the algorithm. They see all these
larger channels getting more views in them
because they upload a video, and it goes like this, absolutely to the sky. Whereas, when they
upload their video, they upload it, they go,
I'm uploading my video, and they've got a little graph, and it just goes something like that. Nothing ever happens. This is because of this
channel authority here. They've got huge returning
viewers which YouTube likes. So they've got a lot of
returning viewers coming in, massive click through
rates because of their loyal audience
versus a kid with a couple
thousand subscribers or whatever with these
tiny click through rates. But over time, the
video with the biggest what percentage will always win. The one that's keeping
people on the platform more is the one that's always going to beat
in the long room, whether that be
through YouTube search or even YouTube brows. So to simplify everything
that I just said, I want you to start thinking
of the YouTube algorithm as a ranking system for the
best of the best on YouTube. And videos constantly
fluctuate around. So videos might have the
number one top spot. So these are the ones
that are absolutely rocketing at the moment, so they might take the top spot, and then a new channel may rack up and upload
a great video, which then takes
that spot from them. So they now move down one, and they now become
the second best video versus this one that
then becomes number one. And it's just a constant
influx of up and down, depending on how
good the content is around the competitiveness
of the topic. And this video here is proof
of that exact concept. I uploaded this video
and it didn't even have 100 subscribers on
the tech channel, and somehow it managed
to get me 600,000 views. And that's for all of those factors that
I just explained. And it was a contribution
of multiple things. It wasn't a competitive topic. No many people were doing reviews of a Paperlike
screen protector, but it was a hugely popular
product, among students. It gets promoted everywhere on every iPad video. It's like. This video is sponsored
by Paperlike. So they promote the
product like crazy. So everyone knows about it. There was hardly
any reviews on it, because most people
think no one's gonna watch a review of this
iPad screen protector. That's exactly what I thought, but I made the video anyway. God job, I did. But that's
the exact thing there. It's the competitiveness
of the area that then dictates whether the algorithm gives
you the top spot, the second spot, the third spot, or the 100th spot in the
almost Premier League, the league table of YouTube.
4. Problems Small YouTuber’s Face: That isn't discussed enough
is the problems that small YouTubers face when they just begin their
YouTube channel. Now, one of the
largest problems that a lot of creators
have the challenge of overcoming is the lack of authority over keywords
within their niche. We're told a lot by a lot of the YouTube channels that you should be optimizing for search, you should be using Cube buddy, VDIQ, all these different tools to get the right
keywords for SEO, to get you to rank number one. But what every single
video doesn't seem to mention is the fact that when
you're a brand new channel, you don't have any authority
at all over those keywords. What I mean by this is basically
You may want to rank for the keyword like iPad or
iPhone or something like that. Maybe making reviews
of these products. So you want to rank
for the keywords iPad? Now, the problem you have is, there are a bunch of other
channels that have done hundreds if not thousands of
videos on Apple products. You've got MKBHD with
16 million subscribers. You've got over 1,000 videos. You've got Unboxed
therapy that's got once again
over 1,000 videos. Lines tech tips, that has thousands and
thousands of videos. So those mega channels have huge authority
over the keywords, Apple iPad, whatever it is
that they want to review. They can crank out a video on rank pretty high for it because of all of that data they've
generated over decades, over a decade, on YouTube. Whereas if you're a
brand new channel, and you're using
the keyword iPad. You have no data to supply
to basically YouTube, say, Look, I make iPad videos, and this is my track record. I've got one that's
got 100,000 views. This one's got 80,000 views. I've got one that's
got 300,000 views. You have none of this data
where YouTube thinks, Oh, this channel right here is a great fit for
the keyword iPad. So we'll let them
rank number one on YouTube or number three
on YouTube for iPad, review blah, blah, blah, whatever year it is, 2020. 30, 2030, let's
say, in the future. So this guy's ranking for
this keyword because they have all of this
data to prove to YouTube that they
are an iPad channel. So my first recommendation
to any small YouTube is, don't get distracted
by using buddy VDIQ and wasting
all this time SEO. Think about how
you can build out your content library with
a strategy to go right. What is my channel
about? I'm going to be doing iPad reviews. So I'm going to make
videos that are so intentional to
the keywords iPads, so I can start generating all that data for
YouTube to think right. This is an iPad channel, or this is a tech channel
that does Apple products. Very, very intentional
with your content. That's what's really going
to help you rank in search. So this then transitions
us perfectly onto the next issue that
small YouTube is face, and that is having a
scatter gun approach with the content that
they're actually producing. And every single
channel does this. When you start off on YouTube, you're not 100% sure what your niche is or what type of
videos you want to create, and you're trying stuff out.
And that's perfectly fine. You're going through
a testing phase, basically you're
throwing stuff at the wall and you're seeing
what sticks and you go, A, I'll do more of that
because that work. I've gone through this on every single YouTube channel
that I've done. When I started the tech
channel last year, I was testing out apple
stuff predominantly, with my keyboard reviews, best keyboard for Mac Mini, my desk setup, awful thumbnail. So there's a reason why
this video underperformed. Best monitors, best
accessories, iPad review. Again, why did this video
only get 2000 views? Because it was Ben's
fourth ever video. He was never in 1
million years going to rank for the M one iPad pro. But when you type this
into Cube buddy or VDIQ, they're like, this is a
100 out of 100 keyword. 100%, you should be
using the keyword M on iPad pro review because this is searched
a lot on YouTube. Yes, it's search
a lot on YouTube, but your channel has zero
chance in ranking for it. That's the reality of
the situation that we're facing because of how
competitive this keyword is. Then you can see stated to
figure a few things out. I found that yes,
the M on iPad pro. I couldn't rank for the review, but I could do a
video on the gaming because this was an
underserved keyword. So people at the M one PAP
pres a very popular product, but there wasn't
that many people doing gaming videos on it because no one really
wants to game on an iPad, but I saw sort of an
underserved area. So that was me learning a bit. And then we started to get
a bit more generic here. We bought the M on MAC, and we were switching from
a WdpeC. Terrible title. If I would do this again, I would Why I switched from Windows PC to M MAC
rather than all of this, M IMAC 2021,
Unboxing, and review. But do you know why I
structured the title this way because of VDIQ Q buddy, all of these tools and this
advice for ranking in search. What was the best keyword
for the M one MAC? It was M IMAC 2021. That's what was the 100 out of 100 keyword term that was recommended to be
used from those tools. So I used it. What was the
next best one M MAC boxing, blue, first impressions,
the colors of them? So all of this was
driven by poor data from these plug ins that we're told solutions to the
problems that we face. And I'm not criticizing
these plug ins because they are
incredibly useful once you've established some
form of authority and data on your channel to actually use them,
where you go, right? I've got 1 million
views on my channel now, I've got loaded data, and then you actually
know whether you can use these keywords or not from
your own personal experience, and then they are useful because they point you
in the right direction. But when you have views,
zero subscribers, they actually point you
in the wrong direction more than anything because you
end up chasing things that Too competitive. So we keep scrolling through. And you can see because this was my second time
doing YouTube, there wasn't too much of
a scattergun approach. I was predominantly doing
apple stuff and iPad, so it was quite focused. But you can see
there's the odd video like this drone video. I was a fantastic video. The Stat said it
was a great video. But it was a bit random. You know, we were doing iPads, iPads, iPads, drone.
Didn't really fit in. And at the time when you
don't have much data, and you're trying to explain
to YouTube who you are. You try to say, Hey, YouTube, I'm a tech reviewer that
does apple review videos. And then you suddenly
do a drone video. There's no connection
there within the data. So it just thinks
that's a bit random. What on earth is this channel? It just confuses the
algorithm more as to where it should be put in your
content in you as a creator. So I realized that that
was a bad mistake. We went back to Applestuff,
Applestuf, Applestuff. And then boom, we had
another video stick. And so on. And then tried out
some gaming stuff because that's predominantly
what I enjoy doing more than the Apple stuff. And now we fast
forward a few months, a few months to this point do apple stuff blah, blah, blah. And now we predominantly
do gaming stuff. I'm now a tech gaming
channel, X boxes, steering wheel, Xboxes, iPads still because
there's still audience. It's predominantly a
gaming tech channel. And that's because I had
a lot of videos that performed well in gaming
tech and I thought, well, there's clearly a gap here for me to do this type of stuff. It was more the tech I actually
enjoyed talking about. I really enjoyed talking about that more than the Apple stuff. So it all sort of
connected really well and we were getting a
lot of positive data to YouTube because of
all the Xbox videos that were performing well
and things like that. So it's just being clever
with the direction you take your channel in
and not being too sporadic with all your ideas. Because it's being all over
the place that confuses YouTube as to what your
channel identifies as. Is it a tech channel?
Is it a beauty channel? Is it an Apple tech channel? Is it a Windows PC tech channel? Is it an Android
smartphone channel? Whatever the subcategories
are within your niche, how do you identify
with in nose? This is one of the most
important things you can do when you're
starting your channel. It's called Niche down, but I think it's much more
than just that generic term. You're actually figuring
out your category within the more generic category of tech consumer tech or beauty or whatever it is
that you are trying to do. The next problem I
see small channels doing is a lack of originality. And I'm going to use myself as an example for
this once again. When I launched
the tech channel, I was learning a brand
new space on tech. I was trying to figure out
how does this space work? How are the best YouTubers
making their videos? How are they titling them?
What are they doing? So I could learn and
sort of educate myself on the eticacies of
this niche on YouTube. So I was seeing
very generic videos that millions of
people are doing, the best M one MAC accessories. 4.2 thousand views. So if we type into the
YouTube search bar, best M on MAC accessories, and we search for this, you'll see there are a ton of videos. You've got best accessories, the cheapest M on Mac
Mini, the everyday dad, 400,000 subscriber channel,
400,000 subscriber channels, a load of other big channels, all dominating this key term. There I crop up for
some bizarre reason about sick the search, but
no one cares at that point. They've watched this video and learned everything that are the best accessories because we've created a
very similar video. There's only so many accessories
that are the best that you can buy for your
M one Mac Mini. So my word of advice to Small
YouTubers is from the off, try and figure out how
can you do something that's different than
just obvious video ideas, which we were trying
out here, like M one iPad pro review on
boxing impressions. It's just so basic ideas, but then suddenly we did a
gaming video on the iPad. It was something a little bit different that no
one else was doing. It was an original idea. We keep scrolling
through desk setup, a very generic idea. Triple monitor setup. This was a little bit more
niche within the topic, but not a huge audience, it
didn't perform very well. And you keep
scrolling through all generic generic
generic until, like, months into the whole equation, where I started this is
generic video, generic video. And it isn't until
almost December time, I would say I started to rock out some half
decent things, but just doing reviews, reviews, reviews. And then boom. This is where we start
to get more original, 600,000 views, 100,000 views. 80,000 views, 300,000 views. This is when I figured it
out and I thought, Oh, I need to do something original and different to stand out, 30,000 views, 40,000 views. Things were just so
much better when I stopped trying to
be a M two copycat, not really copycat, but just something that there
isn't space for. There isn't space for another person to come along
with generic review videos, because we've got
Mr. Who's the boss, we've got MKBHD and Minis.
We've got all these beasts. Create the best of
the best videos, no one needs another
person doing that. They need something that
accommodates us alongside that. So they've got their
favorite YouTube and Mr. Hud Boss and MKBHD, who are going to be there
for decades because they're dead young,
they're like in their 20s. They're going to be there for decades and decades and decades. There's no point
me cropping up and trying to take a space
that's already filled. There's only so many spaces
at the top of each niche. You've got your top
ten, top five channels. And for you to try
and break in to a gap that doesn't really exist
is really impossible. But if you can
position your content around this top thing and you maybe create
a new subniche, or you compliment it in a way where all of these big channels can assist you in your
growth because you're doing something that complements it rather than competes with it. This is where the real
growth exists on YouTube, creating content alongside
the more generic top channels that don't really need
someone to replace
5. The Algorithm Explained: Next, let's move
on and dive into how the algorithm
actually works. More specifically, how does this recommendation
system operate and create links between videos. Hopefully, so far, you
are aware that there are multiple different traffic
sources on YouTube. You've got brows, search, suggestive views, all of
those types of things. Each of these traffic source has their very own algorithm. Depending on where your
video is placed on YouTube, it will have its own
independent click through rate. It will be catered towards
a different type of viewer because they have
different viewing habits compared to someone
that uses search, someone that uses brows, predominantly likes content
to be recommended to. Their habits on YouTube
will be different. So as a viewer, you have to cater your
content differently. So each of these
traffic sources, YouTube has their own
algorithm for them and their approach to what causes
content to be pushed out. And as a creator, you have to be mentally aware of which traffic source you
want to target? Do you want slow
evergreen traffic, then target YouTube search? Do you want quick
and fast views, and trying to create
viral content? You want to think more about
the browse traffic source. And all of this influences
how you title your videos, whether you go a bit more
click bait for browse or whether you go more keyword
optimized for search. But more importantly, I
want to talk about how the recommendation
system knows which are the good videos and creates the link between
all of the data. So let's say that our
keyword is going to be Xbox. This is the topic of the
video that we are creating. Now, what happens on
YouTube is you have lots of different videos
on the same thing, Xbox review, Xbox boxing video, what are the best games on Xbox? Xbox controllers, best
gaming TVs for Xbox, all of these different videos. On YouTube, but all
over the algorithm, millions of videos
filling up the algorithm. What happens is
YouTube basically creates links between
the relevant content to. This is an Xbox review, and this video here is
also an Xbox review, and this video over here
is also an Xbox review, and this is an Xbox review, and this is an Xbox review. The biggest misconception
is that YouTube derives this decision for the link from the keywords in the title
and the description. Sure, that helps
the description, especially, but the
title not so much. You need to start thinking about optimizing your videos for humans when you're
doing the titles rather than for the algorithm. What actually happens
is YouTube AI knows exactly what you're
saying in your video. Hence why it has
autogenerated captions because it detects
what the human being and the video is saying. So it knows exactly every single keyword you're
saying in the video, X box, X box 360, series S, series X, X box one. It knows all of the different keywords that you're
seeing in the video, which is why it knows
where to place it, and then it creates
the auto generated captions on all those
kind of things. Addition to this, YouTube also analyzes all of the
footage in your videos, so it knows when there's text
popping up on the screen, what that text says, what object you're holding in the video, all because of the
AI intelligence that's analyzing everything
on the platform, because one, it helps with the
algorithm, but also two, it helps with the moderating the platform for things that
are also inappropriate. From this AI driven
data that YouTube has, they create all of these
links automatically, regardless of you spamming the keywords like
this in your title and descriptions. But what
can happen is you can create a chain between your content and between
other people's content, but it is possible that
a chain can be broken. And when a chain is broken, this is often the explanation
as to why maybe a video you previously had performing really well suddenly stopped
performing well. It was maybe getting
loads and loads of views over a few months, and then suddenly on nowhere it completely slowed up,
and nothing happened. Well, a chain got broken
in the algorithm. And there's a lot of reasons
for a chain being broken, and one of the biggest is a better video got uploaded
on the same topic. So let's say maybe a
slightly bigger channel, maybe stole an idea
or got inspired by a smaller channels idea because they saw it
got performing well. They went ahead and uploaded their video that
had better stat. It had 70% watch retention, you know, 18% click
through rate. It was really good
status on the video. And your video only had, let's say 40% watch retention and 8.9% click through rate. Well, YouTube now thinks, well, let's create a link here. And a link here, and then we'll create this new
path for the content. So you are no longer
included in this path. So you've been completely
raised from here because your video wasn't as good as this new one that just
arrived on the block. Now, this is constantly
happening on YouTube, and this white it's
better to refer to the YouTube algorithm as a YouTube recommendation
system because constantly new videos
are being uploaded and new chains are being
created between content, new recommendations
for the end viewer, the person sitting on their
iPhone or their iPad. Constantly, there's new content being created for these people. New links between content
content links are being broken. To create the most
efficient route to the best videos
for the viewer. Because YouTube aren't
going to overcomplicate the path to the perfect
answer for the viewer. If someone's coming
onto YouTube, and they want to know
whether to buy an X box, YouTube wants to find
the quickest path with the best videos
to get to the outcome, which is the viewer being
satisfied and buying the X box. So they go away, and they're
happy with their experience, and they use YouTube
to achieve that. So if YouTube can do that in as minimal
steps as possible, rather than a long winded path, they're going to go a right, bang, bang, bang, bang, straight to the solution. Rather, an going. We're
going to recommend This video, then that video, than this video,
then that video, and this video than that video. And then finally, you've got to the golden video that
was really good. The YouTube algorithm and
recommendation system is all about optimization
for the user, the human being using the YouTube app on their
phone or on the computer. That's who you have to
start thinking about, not optimizing for this
mystical algorithm, thinking about the videos for the human being that's
going to watch them. That way, you'll create
a better connection with the viewer in
the video, but also you'll think more creatively
about how to create a better video for humans rather than one crammed with
keywords and rubbish titles, crammy keywords and all
those sort of things. So let's take a look
at a few examples of these chains being created
and also being broken. I can x best the amount of times that this
has happened to me, especially in the technique, because a lot of the
videos are very similar. It's like iPad review
or iPhone review. There's constantly
new videos being uploaded essentially
on the same topic. So you're constantly losing
your spot on the algorithm. I first want to show
you two examples with this LGTV that happened
to me recently. First, let's take a look at the I bought the smallest OLED TV the world. Now, if
you remember back to when I was saying
that keywords aren't really that important
to YouTube placing the content because it knows what you're
saying in the video. Well, this is a prime
example of that. The title is, I bought the
smallest OLED TV in the world. I do not say LGC two OLED review or even state what
the product is. It's very vague the title. But when you look inside
of the analytics, on this video, in particular, you can see that YouTube's
placing it in search, it's getting traffic from
elsewhere on the platform. Cause, it knows exactly what
I'm saying in the video. I haven't mentioned any of
the keywords in the title, but somehow YouTube
knows where to put it. So if there's any
part of you that sort of thought that wasn't
true about the keywords, I hope that puts
your mind to rest. But now let's move on to the biggest thing that
I want to show you, which is videos that
were performing well, but lost their momentum. Al right here, I've got a
video called switching to a 42 inch four K ODTV
as a computer monitor. You can see, I uploaded
this video two weeks ago, and it has 41,000 views. Now, if we take a
look at this channel from a guy called Spawn Point, He's a fellow Tech
YouTuber here in the UK. You can see that he did
this video right here, LGC two review, a perfect gaming and
productivity monitor. That's what it was called. And you can see that
he uploaded that nine days ago and
has 120,000 views. Obviously, it's a biggerchannel, so he has a lot of returning viewers but he uploaded that
video a week after I did. So I uploaded my monitor video on using the LGC
two as a monitor. And then a week later, he released his
version of that video. I was quite a common
idea for people to do this because quite a popular
trend with the LGC two. And I lost a lot of momentum on my video
when his went live. And this is an example of
a chain that got broken. So if we jump into
the analytics here, you can see that the video
was performing really well. It was rocketing
up in the views. It was outperforming
the typical period in the gray line, the blue
line is obviously the views, and it was performing really
well in the first five days, seven days, and Spa Point uploaded his video
roughly a week later. You can see how quick
this momentum was going. We got basically 30,000
views in one week. Now, you can see that I
uploaded my video on 1 May, then Spa Point uploaded his roughly seven days,
eight days later. So the following weekend, he uploaded his video on
a very similar topic. And you can see at this point in the algorithm around the
eighth day when he uploaded, how the video slows up. See how we got 30,000
views in a week. We then fast forward
another week, and we only got like
maybe 10,000 views. Considerably lower. We lost almost more than 50% of the traction this video
was previously getting. Because in the 48 hours, it was getting like 4.5
thousand views every 48 hours. Spawn Point uploaded his video. It went down to like two
and one half thousand, and now it's at 1,300
views every 48 hours. And that's because we
lost a spot in the brows. We lost a lot of the traction
there because Spa Point clearly took our spot on that recommendation
in the links. Span Point took it, so that was getting pushed more the M video, bigger channel, longer video, as well, so more
watch time on there. And same with the search. We probably lost our
position in the search because Spawn Point had
uploaded his video. There's probably other
channels as well that uploaded their video on a similar topic that
I'm not aware of, but I constantly track what Spawn Point's doing because
we're very similar. O viewers watch each
other's content. And that is an
explanation as to why this video went from flying
to completely flat lining. Now, a more frustrating
example that really did irritate me was
this video right here. Can the Xbox Series S
replace a budget gaming PC? Now, I uploaded this
video on 5 January 2022. Just to be clear, I
wasn't saying that Sporm Point copied my
previous video idea. It's a pretty generic idea. It's just I did it first, and then his video killed my video because of
just how it goes. But this video right here is
an incredibly original idea. I thought about it a
lot, and I talked about this in other courses about
how it was a unique idea. Hence why it got so much
traction so quickly, and like a month or two, it
got all of this traction. What happened was a month later, this larger channel with 100,000 subscribers basically
copied the same idea. That's a bit in
nutshell. What happened. They basically copied
the exact same idea on the Xbox Series S, and you could see that there's
got 1.3 million views. Now, this is where it's
frustrating because my video was absolutely rocketing
on my channel. We got 1 million
views in January. Everything had huge momentum. And then in February, the link got broken for this Xbox video. It completely got chopped
off because this guy, a larger channel, uploaded
the same video idea. And basically, because his video was significantly longer, 8 minutes and 30
seconds versus mine, that was just under 3 minutes, because at the time
I was creating shorter videos because
that's what I wanted to do. It just completely got chopped off on the algorithm,
because this, even though Mine probably had
a higher watch percentage, this had significantly
more watch time. It had way more watch time
because it's 8 minutes. So let's say people may
watch 35% of this video because it isn't crazy engaging
in terms of the burrol. It would still probably be
more than the maybe 60% viewed that I would get from my three minute video in terms of watch time
on the platform. So, again, the chain got broken on the YouTube algorithm
with all them dots in them lines because this was the
video that was keeping people longer on the platform
than Ben Rollins is, even though I came up with the idea first for
this type of thing and then this kicked
off a whole trend of people doing these videos. So I did this and this guy
did this 11.3 million views, and then there was a ton
of other huge channels that started doing Xbox, C your Xbox replace
your gaming PC? And like acting like they
came up with this idea? Hopefully, that explains how the YouTube algorithm and
recommendation system is working for
pushing out content. It's basically
looking for the best videos with the best watch time to create links between them
to result in a happy viewer. Unfortunately, sometimes
this results in the original creator who came
up with an idea losing out on all of the views because
a bigger channel comes along has more watch time on their video because
it might be longer, and they get more views
as a result of that. But that's just how the
algorithm works at the moment. It almost doesn't almost reward the original creator
of the idea or the videos that were
first with the idea, not necessarily whether it
was their original idea, but if they uploaded
the video first before somebody else coming along,
doing the same thing.
6. Don't Do This: Like to share something with
you that will hopefully blow your mind when it comes to
creating YouTube videos, because it especially did for me when I tried it the
very first time. And that is that you should
create your thumbnail first. Before you go ahead and
even script out the video, film the video, and
mess around doing all of that editing it and
getting it ready to upload. You should create the
thumb nail first, and that's for a few reasons. First reason is it will let you see whether the
idea is clickable. So you may come up with
a great video idea like the World's Cleaners
desk setup or something. Then when you actually
create and make the video, you can't find a good enough
thumbnail that's clickable. You maybe not clean enough,
eye catching enough, so all them days
wasted were literally wasted because the thumbnail was never going to
be good enough. Whereas if you created
the thumbnail first for your idea or at least had a solid idea for how
that would work, then you would know
whether the idea is going to be successful first. The second reason why
you should create thumbnail first is
because traditionally, people are dead lazy when it comes to making
their thumbnail. They've gone through
days and days of hard work of editing, filming, and doing all that stuff and then they finally
get ready to upload, and they just want to
get the video up and out the way so they can move
on to the next one. So they either just grab a quick screen grab
and throw that up, and it doesn't give it much thought into what the
thumbnail is going to achieve. So if we take a look at
this video right here, it's a couple of months
old, and it's got just under 300,000 views. Now, this is the
thumbnail for this video. I created this thumbnail before I even had any idea what the
video was going to be about. I came up with a really
cool title idea, and then I made a thumbnail
that would match that before I even had any idea what I was going to
say in the video. I had no clue what the tips and tricks I was going to share in this video was until I
done this part first. So this thumbnail
really dictated the direction that the
video ended up taking. And then the final
video was called, If you own an Xbox
series S, don't do this. So you've got the thumbnail with the X box with a
red ring of death, say, this X box is
broken for some reason. Then you've got the
video, so don't do this. Then we go through different
tips and tricks of hidden features on the X box that people maybe
don't know about, and maybe some things that won't destroy your console, but you
should be aware of, like, you maybe save more money
if you buy 12 months of game pass rather than six months and all
those type of thing. Just little tips and tricks
that get more out of your X box than just
not knowing about them. So it's almost like
a getting started guide for Xbox users. Really clever title and a
really clever thumb now. Now, if I'd just gone ahead and came up with the idea
first and gone right. I'm going to make a
title and be like Top ten hidden features on the Xbox Series S. Much blander, much more boring, way more formalic in terms
of how it's out. Tip number one, tip number two, Timber things you've seen 1
million times on YouTube. And then I had A Thumbnail with me holding the Xbox going. Or something silly like that. So, instead, I took a time to actually conceptualize what
would make someone click. And it was a thumbnail that
was incredibly eye catching. It's got a bit of fear factor
there because it's like, Oh, we destroyed his Xbox
and never seen that happen. Sure there was an element of photoshopping in there
to make that happen. But it was successful. So this slight tweak and
pivot in my workflow completely revolutionized
my approach to creating YouTube contents. Now, there's this
super helpful tool called thumbs up dot TV that lets you upload your thumbnail
and preview it on YouTube. So this is how it would
look like a web browser, on the home page, the side bar, channel pages, subscription
feed, history. This way, you can see
how your thumbnail and title looks in combination, and you can view it in
both light mode and also dark mode to really see
whether the thumbnail pops, whether it's too
dark, whether it needs to be brighter,
whether it's too bright, and a bit over exposed, all those types of things, and then you can go ahead
and try different titles. You go right here, it
might be something like the coolest desk setup. And then you'd have a
look and you'll be able to see how that
looks as a title. And then you'll be able to try out different combinations, top ten tips for new
desks. Setups or whatever. Terrible title, but
you see what I mean. So you've got these
different type of things. Then you can see how the
title gets cut off here. So it says the
coolest desk set up, the top ten tips for new desk. And then it gets cut off. So you think titles not very good, and then you can
see it's okay here. So this lets you preview how
it will look on YouTube, and whether your
title's too long, too short, whether
it's actually eye catching enough to warrant
someone clicking on it. And you do all of this first before you go ahead and
actually film the video. Sure, there'll be
certain instances where you may not be able
to make the thumbnail because you need to
build something like a desk setup to get
that thumbnail. But you can still upload
pictures off Google or whatever, just to see how it might
work and conceptualize the idea before you go ahead and actually create
the final thumbnail. And if you would like
to see some data behind this workflow, check
out the click through rate. This is the first 24
hours of this video. You can see it had a
15.8% click through rate, 14.5% click through rate. Very high click through rate. You know, traditionally,
people might be hovering around five to 8% click through rate with the
thumbnails that they're using, so very great
performing. Thumbnail. And if we go ahead and actually take a look at since published, it's continued to sustain a high clickthrough rate at
around sort of eight to 9%. These are obviously
the averages, and it changes depending on. And traditionally, when your
content gets pushed out, the further it gets pushed
out in the algorithm, the more the click through
rate will decrease because it goes to a less
relevant audience because, you know, YouTube is trying
to get it more views. So for it to sustain quite
a steady clickthrough rate while YouTube's gone
through showing almost 3 million
people this video, shows that this really
made an impact. And this work flow is not too dissimilar to what you
would do at college, school, or at university when
you're writing an essay. You don't go ahead and
just write a random essay, and then come up with a
catchy title for at the end. You have a clear idea of
exactly what you're going to research and write about in the essay before you
go ahead and do it. It's the same with a HTue video. You come up with the title the thumbnail in this
instance first, so you have a clear vision of what's going to get
someone to click on this. That then dictates
the story that you're going to tell within the
video that you create. Then it results in
much more success. This video right
here, if we take a look at the analytics, you can see that it has
generated 300,000 views, and it brings in almost
4,000 views every 48 hours, and sometimes it brings
in 6,000 views when the algorithm is giving it
a good push every 48 hours. And you can see how it's
performed well, performed well, and then had a great shelf life for Evergreen on this channel. It just kept ticking
away and ticking away, bringing in new viewers and continuing to
grow the channel. The biggest takeaway
you should take from this video is that you
should start coming up with the coolest
possible title and thumbnail that you can
create for your niche. Because traditionally
what people do is they create
a fantastic video, but then it has a rubbish
title and a rubbish thumbnail. Then because the video
is underperforming by no surprise due
to those factors, they then panic and they start changing the title
1 million times. They maybe make a
different thumbnail, that's more clickbaity. But the problem is the
video doesn't justify the click bait because you never plan for how you would do that. So this then leaves the
viewer a bit dissatisfied. They click on your top ten, whatever video with
your new title and thumbnail come and think, well, that was a waste of time. It didn't really
deliver on the promise. Whereas, instead,
if you can come up with the click
bait title and click bait thumbnail first and then create a video
that justifies that, the viewer will feel
way more satisfied. They'll go away with value,
and they'll think, actually, that video that really
attracted my click delivered on that huge promise
that I never anticipated. And that's how you
can start to begin to get more clicks
on your videos, but leaving the viewer satisfied rather than dissatisfied, and it will leave
your brand with a much better feeling than just being desperate for views.
7. Thanks for Watching!: That I've broken all
of the fundamentals down when it comes to
the YouTube algorithm. I want you to go ahead
and complete a task, which is where you will
completely change your workflow. As I mentioned, in this class, I like to make the thumbnail
first before actually producing and refilming and
editing the YouTube video, so I want you to try and apply my workflow to your very own
content production process. So go ahead, come up with an incredible video title that's very exciting video
idea and create a thumbnail that
matches that. And then go and create the
video, make the script, and make the piece of content
that will marry up with that thumbnail to
ensure that you have packaged the
video correctly, and also a great way to
disregard any ideas. You come up with a
great idea, don't know how to quite package it. So obviously, you will scrap that entirely and move
on to something that you know may have a high
successful chance of working. That's exactly what
I do, and I would recommend doing that
as the class project. So you get a little
bit of an idea and understanding of how that
process would actually for you going forward
in the future. But if you actually
want to learn more about how to grow on YouTube, I have a class here
on SkillShare, which is my YouTube
road map to success. This is one of the
most comprehensive courses on the entire platform. It's a series of multiple skill share classes
combined together in a sequential manner that you can follow one by one all
the way from A to Z, that will show you
everything you need to know about YouTube growth, both from the back
end, managing, time managing, making
videos, filming content, getting views, and
also sustaining the success over a
long period of time. As always, Abby Ben Rollins, thank you so much for watching. Be sure to follow me here on
Skill Share 'cause we upload brand new classes pretty much every single week on here now, and I'll see you
in the next one.
Ben Rowlands, Content Creator with 800,000 Followers
