YT Algorithm Working

The document discusses how the YouTube algorithm works, including a brief history of changes to the algorithm over time and how it currently functions. The algorithm aims to recommend videos users will find most relevant and engaging based on their interests and watch history as well as videos' performance metrics.

Uploaded by

Manjeet Rana

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

93 views

YT Algorithm Working

Uploaded by

Manjeet Rana

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

You are on page 1/ 9

How YouTube Algorithm Works

Submitted by: GAURAV MEHTA & MANJEET SINGH RANA

SUBFMITTED TO: Dr. N. S. Bohra

[BBA1F]
Table of Contents
 A brief history of the YouTube algorithm
 How does the YouTube algorithm work in 2023?
 How YouTube determines the algorithm
 What is the YouTube shorts algorithm?
If you believe in free will, we have terrible news — well, at least when it comes to
YouTube. Because YouTube’s algorithm for recommendations drives 70% of what
people watch on the platform.

Seventy percent! That is some seriously staggering influence. Youtube’s 2.5 billion
users watch 694,000 minutes of video each minute. And the platform’s
recommendation system is responsible for the majority of those views.

So it’s no surprise that marketers, influencers, and creators are obsessed with
unlocking the secret of the YouTube algorithm. Getting recommended to the right
viewers at the right time is the ticket to YouTube stardom, but how does it work?
What makes it tick? And, most importantly, how can we take advantage of this
mysterious formula?

Well, ponder no more, my ‘Tube curious friends because in this blog post, we’ll
cover everything about the YouTube algorithm that you’ve been dying to know.

 What the algorithm is (and isn’t)

 The most recent changes to the YouTube algorithm
 Pro tips for getting YouTube’s search and discovery systems to work for
you

A brief history of the YouTube algorithm

 What is the YouTube algorithm? To answer that question, let’s do a quick
overview of how YouTube’s Algorithm has changed over the years. And
how it works today.
 2005 – 2011: Optimizing for clicks & views
 According to founder Jawed Karim (a.k.a. the star of Me at the Zoo),
YouTube was created in 2005 in order to crowdsource video of Janet
Jackson and Justin Timberlake’s notorious Superbowl performance. So it
makes sense that YouTube’s algorithm started off by recommending videos
that attracted the most views or clicks.
 Alas, this led to an increase in misleading titles and thumbnails (a.k.a.
clickbait). User experience plummeted as videos left people feeling tricked,
unsatisfied, or plain old annoyed.
 2012: Optimizing for watch time
 In 2012 YouTube improved its recommendation system to support time
spent watching each video. It also included time spent on the platform
overall. When people find videos valuable and interesting, they watch them
for longer. Or so the theory goes.
 This shift to reward watch time was a game changer. According to Mark
Bergan, author of Like, Comment, Subscribe: Inside YouTube’s Chaotic
Rise to World Domination. Early YouTubers were basically making TikTok
videos…but watch time created gaming, beauty vlogging, alt-right
podcasts… all these verticals we now associate with YouTube.”
 Accounts that were big performers previously (like videos from eHow, or
MysteryGuitarMan) dropped off almost immediately.
 YouTube’s algorithm change led some creators to try to make their videos
shorter in order to make it more likely viewers would watch to completion.
Others made their videos longer in order to increase watch time overall.
YouTube didn’t comment on either of these tactics and maintained the party
line: make videos your audience wants to watch, and the algorithm will
reward you.
 That said, as anyone who has ever spent any time on the internet knows,
time spent is not necessarily equivalent to quality time spent. YouTube
changed tack again.
 2015-2016: Optimizing for satisfaction
 In 2015, YouTube began measuring viewer satisfaction directly with user
surveys. It also prioritized direct response metrics like Shares, Likes, and
Dislikes (and, of course, the especially brutal “not interested” button).
 In 2016, YouTube released a whitepaper describing some of the inner
workings of its AI: Deep Neural Networks for YouTube Recommendations.
 In short, the algorithm had gotten way more personal. The goal was to find
the video each particular viewer wants to watch, not just the video that lots
of other people have perhaps watched in the past.
 As a result, in 2018, YouTube’s Chief product officer mentioned on a panel
that 70% of watch time on YouTube is spent watching videos the algorithm
recommends.
 2016-present: Dangerous content, demonetization, and
brand safety
 Over the years, YouTube’s size and popularity have resulted in an
increasing number of content moderation issues. And what the algorithm
recommends has become a serious topic not just for creators and advertisers
but for journalists and the government as well.
 YouTube has said it is serious about its responsibility to support a diverse
range of opinions while reducing the spread of harmful misinformation.
Algorithm changes enacted in early 2019, for example, have reduced
consumption of borderline content by 70%. (YouTube defines borderline
content as content that doesn’t quite violate community guidelines but is
harmful or misleading. Violative content, on the other hand, is immediately
removed.)
 This issue affects creators, who fear accidentally running afoul of ever-
changing community guidelines. Or being punished with strikes,
demonetization, or worse.
 (Former CEO Susan Wojcicki said one of YouTube’s priorities in 2021 was
increasing transparency for community guidelines for creators).
 It also affects brands and advertisers, who don’t want their name and logo
running alongside white supremacists.
 Meanwhile, American politicians are increasingly concerned with the
societal role of social media algorithms. YouTube (and other platforms)
have been summoned to account for their algorithms at Senate hearings.
And in early 2021 Democrats introduced a ”Protecting Americans from
Dangerous Algorithms Act.”
 Next, let’s talk about what we know about how this dangerous beast works.

How does the YouTube algorithm work in 2023?

The YouTube algorithm in 2023 delivers distinct recommendations to each user.
These recommendations are tailored to users’ interests and watch history and
weighted based on factors like the videos’ performance and quality.

When deciding what to recommend to each user, the YouTube algorithm takes into
account the following:

 What videos have they enjoyed in the past? If you’ve watched a 40-
minute video essay about the flags of the world or gave it a like or
comment, it’s probably safe to say you found it interesting. Expect more
flag content coming your way.
 What topics or channels have they watched previously? If you
subscribe to the Food Network’s YouTube channel, the algorithm will
likely show you more cooking content.
 What videos are typically watched together? If you watch “How to
change a monster truck tire,” and most people who watch that also watch
“Monster truck repair 101,” YouTube might recommend that as followup
viewing.
That’s why a Millennial comedy-fan design-writer mom has a home page that
looks like this:
Of course, YouTube wants to recommend relevant, quality videos to each of its
precious users. It’s not exactly a positive experience to follow a suggestion to
watch “The World’s 36 Most Stylish Cats” and find it boring, low-quality or
weirdly racist.

So how does YouTube evaluate if a video is worthy of recommendation?

First: it’s not about the content. The actual content of your video is not evaluated
by the YouTube algorithm at all. Videos about how great YouTube is aren’t more
likely to go viral than a video about how to knit a beret for your hamster.

“Our algorithm doesn’t pay attention to videos; it pays attention to viewers. So,
rather than trying to make videos that’ll make an algorithm happy, focus on
making videos that make your viewers happy,” says YouTube.

Instead, YouTube looks at the following metrics for its recommendation algorithm:

 Do people actually watch it? When a video is recommended, do people

actually watch it, ignore it, or click “not interested”?
 How long do people watch it? The YouTube algorithm looks at both the
view duration and the average percentage viewed to inform the ranking.
 Did viewers like it? Likes and dislikes are evaluated, as are engagement
rates and post-watch survey results.
 What is your regional context? The time of day and the language you
speak also influence the YouTube algorithm.
How YouTube determines the algorithm
More than 500 hours of content are uploaded to YouTube every single minute. The
search system would be absolute chaos without the YouTube algorithm trying to
help you find the most relevant content.

That’s right: the goal isn’t to bring you the most popular or the most recent video
on your search term. It’s to bring you the video that you specifically will find the
most useful.

That’s why two different users searching for the same term may see a totally
different list of results.

YouTube’s search algorithm prioritizes the following elements:

 Relevance: The YouTube algorithm tries to match factors like title, tags,
content, and description to your search query.
 Engagement: Signals include watch time and watch percentage, as well
as likes, comments, and shares.
 Quality: To evaluate quality, the algorithm looks at signals to determine
the channel’s authority and trustworthiness on a given topic.
 User search and watch history: What have you enjoyed or viewed in
the past? This will impact which search results the YouTube algorithm
will assume will be helpful.
That being said: these factors are combined in slightly different ways, depending
on where on YouTube you are receiving recommendations.

YouTube recommends videos in three different places on the platform.

Home

This is what you see when you open up the YouTube app or visit the YouTube
website. It’s personalized to each viewer. The recommendation engine selects
videos for the Home screen based on…

 Performance of the video

 Watch and search history of the user
Suggested Videos
These are the videos recommended alongside the video you’re already watching:
the list of vids that appear under ‘Up Next.” The algorithm suggests videos here
based on…

 The topic of the current video

 The viewer’s watch history

The keyword obviously plays a role here. But each user’s search results will be
slightly different thanks to the personalized signals the algorithm takes into
account. These signals include:

 The relevance of the title, description, and video content to the search
term
 Performance and engagement of video
What is the YouTube shorts algorithm?
YouTube Shorts are short, vertical videos created using a smartphone and
uploaded directly to YouTube from the YouTube app, like Stories. Millions of
YouTube viewers are watching Shorts daily… so don’t sleep on this new format.

YouTube’s VP of Product Management described Shorts as “a new short-form

video experience for creators and artists who want to shoot short, catchy videos
using nothing but their mobile phones,” and goes on to say, “Shorts is a new way
to express yourself in 15 seconds or less”.

But how do you get your Shorts discovered?

YouTube users can find and watch Shorts on the YouTube homepage or via the
Shorts tab on the website or app. Shorts are subject to the same recommendation
signals as “long” YouTube videos. (YouTube Classic?)

Here’s what the YouTube Shorts algorithm takes into account:

 Relevance: Do the title, tags, content, and description match the search
term?
 Engagement: Do other people like and comment on this video?
 User watch history: What have you enjoyed or viewed in the past?
 Similar content: What other Shorts do similar audiences like to watch?
 Watch time: Less important than for classic videos. But if someone can’t
even sit through a 15-second video, that’s probably not a good sign.
Hot tip: you can schedule your YouTube, uh, Longs via the Hootsuite Dashboard
so you have time to focus on more spontaneous YouTube Shorts on the go.

Paige Cooper is the Hootsuite Inbound YouTube Lead. She runs Hootsuite Labs,
our YouTube channel and she sees Shorts as an opportunity ripe for the taking.

“The rise of vertical video hasn’t changed the main algorithm per se, but YouTube
Shorts are creating a big new opportunity for creators,” she says. “If you’re already
running an Instagram Reels or TikTok strategy, publishing on YouTube Shorts
seems to be an easy win.”