Most of you guys might not even realize that
there's a way in YouTube analytics that
you can do some AB testing
and comparing how one series of content
might be performing against another series of content,
so you can make a better decision
what series to invest into,
what's worth your time,
what's not worth your time
rather than just making a lot of videos
and just trying to do some guesswork.
Let me take you to my desktop here
and show you exactly how you can do this
in your analytics.
(techno music)
Hey, guys!
My name is Tim Schmoyer
and welcome to Video Creators.
This channel's all about helping you guys
grow your YouTube audience
so you can spread a message that reaches people
and changes their life
and if you want to reach people,
you need to make content that goes out and gets them.
Now, one thing that I've been trying to do
here on this channel
is experiment with a lot of different formats
from doing a blog,
doing podcast interviews,
and doing, like, motion graphics,
and doing vlog style thing,
just to compare, like, what do you guys respond to?
What doesn't, what's getting me more watch time,
subscribers, things like that.
And so, what you can do in your YouTube analytics,
they actually have a feature that
most creators don't even use,
it's called groups
and you'll find it right in here in the upper right corner
and you can either create a video group or a playlist group.
Now, I recommend just for our purposes here
starting off in making a video group
and what you do is
you can search for different videos on your channel,
give the group a name,
so let's just say for this,
for example, this is going to be,
you know, I've already made these groups,
so I'm just making this up right now but, like,
Podcast Videos, okay?
And then I go through here,
search for them,
find all of the podcast videos
or you can add them by ID if that's easier,
just copy and paste them in here, your video IDs.
And you just check the ones,
I'm just checking random ones right here,
and then you save and create that group.
And I've already created a few groups.
You can create groups from playlists and things as well,
but I'm gonna go browse all groups like this,
and so what I can do,
I have how-to interviews,
consultation recordings,
tutorials, live Q and As,
live news streams, podcasts, vlogs,
you see how many videos I have in each of them right now
and I can add these,
I can look at these side by side.
So, here I am going to look at how-to interviews.
Now I'm looking at the real time analytics
for these right now,
go to comparison,
maybe I'm gonna change this second one to
how-to tutorials.
I don't know, how-to interviews versus tutorials.
So, rather than real time,
let's come down here,
let's look at, like, watch time, for example,
and see how these two groups are comparing.
Okay so, I can see, whoa, okay,
how-to tutorials far out-performing the interviews.
So, that might be a good indication to me that
Tim, if you're gonna teach someone how to do something
on this channel,
do it as a tutorial, not as an interview.
Now, I also need to look and see, like,
what videos are in each of these groups
because it might be I have one
really high-performing tutorial
that's kinda throwing all the data off.
So, sometimes what I'll do is I'll make these groups,
I'll take the 10% top
and the 10% bottom off,
so the top 10% that have performed the best,
I won't include those videos,
and the lowest 10% of videos
that are performing the worst,
I won't include those either.
And then I'll make sure that
there's similar date ranges and stuff
but over here you can kinda just seeing
how valuable these groups get
when you can start seeing how they compare on watch time,
in views, and average view duration
on each of the videos that are in each of these groups here
and really start comparing these things side by side.
Now, I'm doing this as, like,
a side by side comparison of videos
based on the content, you know,
based on the style,
but you could do it based on,
let's say you have a branded intro that
you want to evaluate and you want to know
is this new branded intro going to increase subscriptions
or decrease subscriptions?
And watch time, or decrease watch time,
so, maybe you have one branded intro A
that you run for, like, two months,
and then branded intro B
that you run for two months on all your content,
and then you have a group of branded intro one,
branded intro two,
just put them in your analytics
and look at them side by side
to kinda see how they're doing.
Now, you could evaluate styles,
you could evaluate content elements,
like, call-to-actions, or branded intros.
You could do it based on your content goals,
such as, like, discoverable content
versus community content.
If you wanna measure, like, engagement
or things like,
or any of the combination of all of the above.
Coming back to this example,
you can see that one of the reasons why
how-to tutorials is so much bigger is that look,
I only have four videos in the how-to interviews,
the blue,
and I have 25 videos in that group.
So, it's not quite an apples to apples comparison.
Another thing that I might do with these is
only include videos in these groups that
I've published over the past two months,
or three months, or six months,
whatever date range you wanna set, so that,
so I'm making the how-to interviews and tutorials
side by side at the same time,
posting maybe one of those, each of those a week,
and then I'm only taking into consideration
the videos that were published around the same timeframe
so it's a better apples to apples comparison.
So, let's look at, like, live stream Q and As
versus, like, maybe live stream,
or news live streams.
Okay so, how do those two live things
compare to each other as far as performance is concerned?
And I'm gonna set the date range on these to, like,
a lifetime probably here
so I can get more data
since I haven't updated the live streaming groups
here recently.
Here we go.
Okay so, I can see that
there are seven videos in the news streams
and only four
in the Q and A.
So I can look at these and kinda see already, though,
the watch time minutes,
I mean, the top one I got in the Q and A is,
you know, it was only 26,000 watch time minutes
and the live news ones
are getting a lot more.
So, all of my live news streams
outperform my live Q and As,
which is good for me to know.
So, when I go live,
doing live Q and A is probably
just strictly a community play,
it's not a discoverability play.
Whereas when I live stream about news,
it's a good, probably better discoverability play
than the Q and As.
There is a lot more you can do in here,
like, you can edit groups,
make copies of groups, delete groups,
and everything,
but it's a really great tool to start
doing some things intentionally and strategically
on your channel
and putting them side by side to see
what's working, what's not working,
what to eliminate, what to invest more into,
and test a lot of your stuff
rather than just throwing it against the wall
and seeing what sticks.
I wanna hear from you guys in the comments
about how you plan to use groups in your YouTube analytics
and what data you hope to find from that.
And the rest of you guys,
if you want more ideas about this,
read the comments down there below,
you're gonna learn a ton from the other people here
in this community who are experimenting with this as well.
Now, if this is your first time here at Video Creators,
I would love to have you subscribe.
We're just all about helping you guys
grow your YouTube channels
so you can spread a message that reaches people
and changes their lives.
So, thank you for letting me be a small part of that
here today.
Subscribe,
and I'll see you guys again on Monday, hopefully,
for our live stream.
See you then, bye.
Không có nhận xét nào:
Đăng nhận xét