Mike talks about YouTube's algorithm and how the analytical impressions act unfairly for small car channels compared to big ones. This is "proven" by comparing successful videos to those that did not perform as well. He takes a stab at solutions, like addressing thumbnail design, reworking titles, and even discussing alternative outlets. Clickbait is addressed.
00:00 Intro
00:55 Depression
01:50 Comparing
03:31 Thumbnails
04:40 Impressions
05:16 Improving Thumbs
05:41 Clickbait
06:20 Title Game Show
07:35 Alternatives
08:52 Reaction Videos
09:27 Conclusion
10:29 Bloopers
Mike the Backyard Mechanic (not a real mechanic) works on his cars in his garage (not actually his backyard) and shows how to do repairs and mods, often not correctly, for your entertainment. Mike is not a REAL mechanic, he has common knowledge. Any injury, damage, or loss that may result from improper use of these tools, equipment, or from the information contained in this video is the sole responsibility of the user and not Mike.
[Music] Hola! I'm Mike the backyard mechanic,
interrupting my usual Mustang related videos to talk about how... the
algorithm is SCREWING small car channels! How do I know? Because it's screwing me! I've been getting very small view counts,
which correlates to the analytical impressions. It could be my type of channel. It's a challenge!
I'm talking about cars AND being funny. Well... I'm talking about cars. I guess that's not so much of a challenge. Note that I do know I have
several hundre
d loyal fans, and you guys are MUCH appreciated and loved... that is until I hit the YouTube
jackpot and you guys all ditch me because I'm no longer an indy
darling and I become a sell out... doing commercials for Keeps. Recently I tried releasing a video a week for 6 weeks. Nothing changed. YouTube does
not reward that like they used to. Now, I haven't made a video in a couple
weeks. The lack of views has me feeling off. So I decided I needed some psychotherapy. My therapy started as all p
roper therapy does, with a website. This mental health self
assessment quiz is what I used. I took all of the quizzes, and I did
quite well. It appears I have them all. The postpartum depression surprised me, since
I'm a man... and my kids are in their 20s. Maybe I overdid it, but the last
two quizzes had me feeling unstable. The words were looking like gibberish. I haven't done my final evaluation yet, because
I had to stop and learn what schizophrenia is… WHO'S THERE? It's just a wall. An
yway... I think YouTube is out to get me. It's not like the channel on a whole is
bombing, there is one successful video: "Fix your Catalytic Converter
for $10?" With 2 million views. Do I expect this performance for every video? YEAH. That's 2 million people who haven't
come back! It has 30,000 likes. Why do I think YouTube has it our for
me? All the evidence is there. I've put it the work! I went from this.. Hey guys, Mike the Mechanic here. Hey guys, Mike the Backyard Mechanic here. Today
I'm gonna show you how
to test for warped rotors. Hola. To this... I'm installing headers! [Music] I'm fixing my ECU! Do your ideas suck? Hola I was really low energy in those early videos. Now... I'm slightly less low energy. Comparing a recent video with a modest 5,500
views, to the big video over the last 28 days... The big video has a click through rate of 3.6%
and 3.7 million impressions The modest video has a click through rate of
8.1% but only 5,300 impressions Engagement (view durat
ion) is
even the same, so that's not it. In the YouTube FAQ it says: "Even if a video is performing well
compared with other videos on your channel, it may be that videos from other
channels are performing even better." There's your proof right there. YouTube is basically saying you're
a small channel and you suck. No YouTube, you suck! I got some decent tips from TubeRanker,
which suggests the video might have the problem that it is using two key
words, "cross threaded" and "galled". Look
ing at the title again, "Galled or Cross
Threaded Bolt Hole" might be a clever way to jam multiple concepts in few words, but it also
might be too complicated for a keyword algorithm. I changed the title to 5
Ways to Fix Galled Threads, but now I have a redundant
keyword in the thumbnail. Changed that to "Seized" which
I think describes the thumbnail better and addresses topics in the video. TubeRanker also suggests that
small channels stick with long tail keywords. So it's possible that
"galled
threads" doesn't hold up to the big boys. But that's search. Not impressions. Meanwhile a "$10 catalytic
converter fix" is more specific. But ultimately, I think that video
blew up because I tapped into a lot of people's frustrations with their
catalytic converters and costs. I know I got lucky with that video. The goal is
to figure out how to make more successful videos. I CRAVE attention. Views correlates to impressions, and
impressions correlate to click through rate. Impressions
are how many times
YouTube serves up your thumbnail, in the home page, side bar, or end screen. Click through rate is how many times it is
clicked, as a percentage of impressions. The click-through is completely in your control, based on how clickable your
thumbnails and titles are. Looking at my latest videos, I do see a pattern.
Everything with my face gets less views. According to this analysis, 72% of
popular thumbnails contain a face. I have to figure out why
people won't click on my
face. Let's see if I can fix this thumbnail for this
video on Cabin Filters that WAY underperformed. First thing is to change the ugly text
and simplify it. That's the easy part. Let's see if I can fix my face... enhance...
auto smart fix. Not sure what that does. Oh! That should work. We'll see if people will
click on my enhanced face. My titles are part of the click through
rate, and they don't seem to be working. After scanning some high
performing channels in my category, I think I ca
n word smith titles that the
algorithm might like, such as this one: Ok, that's a joke. I don't plan to do clickbait
titles. At least not black hat clickbait. YouTube's rhetoric condemns clickbait, while
at the same time the algorithm rewards it. Well, it does and it doesn't. The difference
between white hat and black hat clickbait is how long the audience views the video,
combined with subscribes, likes or unlikes. The algorithm supposedly takes this into account. More simply put, if the
content didn't match the title, and your audience feels cheated, it's black hat. Let's play the Internet's favorite game
show "Is it white hat or Is it black hat?" And rank thumbnail and title ideas
from 0 for white hat to 10 for black hat. Now for our first contestant! "The INSANE story of a broken stud" It's a bit of a stretch to say the broken stud
was a story. But it seems more clickable. 2/10 "If you don't do THIS, you're STUPID!" I'm gonna stand by this! It got terrible
views, and I kn
ow this was a good PSA. 0/10 "The DEADLY truth about ChatGPT" This was too off topic for my channel, so
I don't think it matters. But ChatGPT is only going to kill you if you
are incredibly stupid. 5/10 "If You See This, Call the POLICE!!" I've seen a few of these videos
around YouTube. I don't believe it's a real thing. I don't plan
to do this, but if I did... 8/10 "The government is controlling
your mind with cars!" Hey... I might actually do
that one... if I can think of a good justific
ation for it. Otherwise... 10/10 Maybe I should consider
alternative outlets to YouTube. One alternative is Rumble. Of course, due to the nature of the existing
content, I may have to change my style... Today I'm going to show you how to put sporty
decals on your hood to dress up your car. And you know why Bigfoot is never found? Because the world is flat. Just before he's captured, he goes over the edge, and hides underneath the Earth.
With Jeffery Epstein... and Disney's frozen head. [Scra
tch] Today I'm going to show you how to black
out your chrome emblems using Plasti Dip. But first we need to talk about the lizard
people that are drinking baby blood, to give them powers to fake
the moon landing using 5G! Open your eyes sheeple! Another possibility is TikTok. Here's how to change a flat tire. Pay close
attention or you could hurt yourself badly. Now we're ready for the beach! Damn it's cold... All suggestions for increasing
views says to focus on content. I guess it's poss
ible that my content is not good enough.... NAHHHH! Reaction videos are a really popular
format for many channels on YouTube. Maybe I could augment my content and
do reaction videos about… my videos! HAHAHAHA!! I was really sweating that day. Hola! Heh heh... I knew I was going to say that! The bloopers are my favorite part! Heh heh... Oh! That looked like that hurt! Mmmph mmmpph... I'm eating
the same thing I'm eating now! How did I do that? Oh my god!!!! I've been trying to do all the thing
s correctly. Good looking thumbnails with a little bit of text that adds to the image but
doesn't repeat the title. Titles that are short, simple, descriptive,
and hopefully create curiosity. Maybe it IS the content. The next plan for the Mustang was to upgrade it with Detroit Rocker cams and a
Whipple Intercooled Supercharger. However, looking over what's popular on my
channel... the stuff on headers got no views... The next video will be about changing
the rotors on the Honda Pilot! Oh ye
ah, that's some sweet sweet content. So what can you do? Well, every YouTuber says "Hey everybody don't forget
to like and subscribe!" and it sounds fake, but I am sincere, when I say, "Hey everybody don't forget
to like and subscribe!" Maybe someday I will look
into retention. Although, according to the analytics,
only 35% of you know that. C'mon! What other channel is
going to give you analytics humor? [Music] Heh... you know I'm recording this and
you;'re going to be in the bloopers? Yo
Comments