 In this video, I'm going to show you how you can get more views on YouTube by using TubeBuddy's A.B. split testing tool to drive more clicks through on your thumbnails, your titles, and descriptions, and so on. So this is really a great little tool that's pretty simple to use. Just need to follow these few steps that I'm going to outline in this video. So let's have a look at where you can find this if you come over into YouTube Studio. This is one of these tools that you can access it from YouTube Studio from up here. Although the actual tool itself lives on the web. So come down here, you've got the extension tools, and then you've got website tools. So down in this website tool section, we've got a whole series of tools that you've got in here. And it is this one here, A.B. tests. So this is where you're going to run your split testing, A.B. tests because you're comparing A to B. That's the idea behind it. So if I come over to my browser now, this is where we can start creating tests. And by the way, I should just mention, I do feel that all of this stuff that I talk about is technically easy. It's just a case of learning the steps. So I had a comment recently on someone said, if it's easy, I wouldn't be watching a tutorial. I disagree with that. I think that the whole point of tutorials is to simplify the process and show you the easy steps that you need to take. So hopefully there is nothing in here that's going to be overly complicated. It's just a case of you may not know how to use it. So just wanted to address that little comment though, because I think that everything can seem difficult if you don't have to do it. But once you're shown, hopefully it can become quite easy to do. So this is one of those things where they have simplified, to my mind, the process of actually carrying out split tests. And so if I come to the top here, we've gone into the A.B. testing area, just click on here to create A.B. test. And once you click on that, it will take you through to this page. And it will bring up your most recent uploads as a default. So you've got there your six most recent uploads. But equally, you can actually search for a specific video as well. So you can either search by name or you can drop the video ID in or whatever you want to do like that to actually find the specific video that you want to run a test on. For the purposes of this demonstration, I'm just going to literally click the first one there. Next, you've got the choice of these two options. And this is for a thumbnail CTR test. So checking the click-through rate of a specific thumbnail. And as it says, focus on improving click-through rates with a better thumbnail does not affect video ranking or discoverability. So that is basically just changing the thumbnail. But it's not going to change anything in terms of the search or anything like that. But this one here, meta data test, this is going to give you a few more different options basically. It allows you to change the thumbnail, but you can also change the title, description and tags as well. Now that is going to have an impact on the discoverability of the video. Because it's going to basically re-index. So it actually says here, there's a little warning sign. Changes to your title, description or tags will cause your video to temporarily drop out of search results while you usually re-index it and can be detrimental to high-performing videos. This does not happen with thumbnail tests. So that's just something to bear in mind. One of them, they're just switching out the thumbnail, but this one is going to change in terms of how it's been found in search and stuff like that. So just bear that in mind. If you are, you know, I've got a really great performing video and you're going to try and just tweak it that little bit. This will have some sort of effect. It can go both ways though. It can improve that discoverability as well. But that is working at the, you know, the YouTube metadata level, whereas this one is literally just testing, you know, people's, how people experience it and find it and whether they click on it. That will feed back into the YouTube algorithm though, eventually obviously because YouTube bases, you know, impressions based on how people react and click on it. So it just kind of feed in, but it's just a little caveat that is to bear in mind. I will just say though, even with this one though, if you are going to be thinking about changing up your title, description, tags and thumbnails, don't do them all at once. Just change one thing at a time because you may well change the thumbnail and have a negative impact, but you've changed your title and it's really had a positive impact, but they've canceled each other out. So you won't really know anything that actually happened with it. So definitely try and test one thing at a time. But I'm just going to click on this one for the time being, click on the thumbnail. You've got here the option to set the number of days that you want to run a test for. The default is 14 days. You can run it for as long as you want. It does depend a little bit from a sort of statistical point of view. You want to make sure you give it enough time to run. You know, don't run it for a day if you've got a video that averages two views a day because you're not going to have a statistical significance. But if you have got a lot of views on your channels or it's a high performing ad, then you could consider potentially reducing that if it was going to give you enough data. But that is the default anyway, 14 days, and that's just what I've left mine at. You can also, though, choose this option to run until click through rate is statistically significant. So that's basically giving you a way around that where you just actually let it run and then it will tell you when it's got enough data to draw some conclusion. So that is the other option there. Let's just assume we'll go for this one, though. And then if I click on continue, it's going to prompt me to then upload my new thumbnail. So this is the original thumbnail, and then I'm going to just add in my new thumbnail here and click on start test. And that is it. Once you've done that, the test is running. I'm not going to actually do this right now, but that is the test would be running. And here are some that I've previously run. So here is an example of a particular split test. If I click in here, then this will show you what you see as a result. Now, this one has already completed. So we've got this big green box saying the test is complete. This is what I mean by ease of use. It's just basically telling you that the test is complete. And by the way, those first four steps to upload it, although we're going to drill down into the stats here, let me just come back to the results page because here are some other tests that I've run. It does just tell you here, look, this is the original. The original was 8.8, sorry, 10.82% click through rate. The variation was 11.37. I've mentioned this on other videos, but just bear in mind that when comparing these things, don't look at the actual difference between these in terms of, okay, from 10.8 to 11.3 is only 0.5% of a percent. It's not really that much, is it? It's easy to think that, but actually what you want to look at is the relative change. So what is the change as a percentage of the first one? So in the actual fact, the variation click through rate was 5.1% higher. So that means for every 100 impressions, an extra five people are clicking on your video. So these do actually make a real tangible difference to the number of views that you're getting on your YouTube video. So definitely something to take note of and definitely come and have a look down here where it's going to tell you the actual change in the uptick basically or down tick in viewers. And I've had some of those, so I'm going to come on to that in a moment as well. But you can see how, yeah, it just tells you this is the name of the video, this is the test date when it was completed, and then also the results. So it will say it's complete and either the variation is the one that needs to be set or the original. If you just do click into one of these though, then you'll see the results. You got a few more results in here. So you can drill down into more detail if that's what you want. So you can see that the original, these are the click through rates that were quoted on that first page over there. But then also you can see the actual number of impressions and the physical number of clicks. And then here you've got the watch time as well. So interesting to note that you've got one thing which is how many people actually click on it, but then seeing the difference in the average watch time between the thumbnails is interesting too. I mean, that might give you an indication of if your thumbnail is giving a good indication of the content of the video, because if people are coming to your video and you're getting loads of click-throughs, but then they're just leaving the video really quickly because they haven't got what they were expecting, then that gives you some feedback as well. So there is space to sort of drill down into these a little bit further in terms of the detail. Now we've got the number of daily impressions and we've got the daily clicks as well just in graph form and the same here with the total views and duration. So you've got some different graphs there. So that is basically the summary you get as the test is going on. It'll take a couple of days before it starts giving you some results. And so it may just say, you know, not enough information when you first start running it. But after a couple of days, you'll start to actually see here and you can actually come in and have a look at how it's performing as the test goes on. And you can see here, based on the total number of impressions and clicks, we are 90% sure that your variation will perform better than the original. This is not yet considered statistically significant. 95% level is required. So this one, in fact, it should really have run for a little bit longer. So this would be a case where this one had a low number of views and so it'd be worth having it to run for a longer time to be statistically significant. If I just come out of this one though, you can have a little look at some of these others. And one thing I want to point out is, I mean, this was an example of where I was changing up the general style of my thumbnails. So this was kind of like the style that I'd had since I started my channel. And then I wanted to just refresh them a little bit. It's not a major change, obviously. Still my silly YouTube face there. But it was just changing the style a little bit. So it worked for that one. However, it doesn't mean to say that it's going to work for everyone. And I changed a few things here. I changed the style, but also the text. So that's something where, you know, changing a couple of things at one time. I did keep the same picture, but yeah, that is a case of the variation was actually performed worse than the original. Same with this one, same with this one, same with this one. So in this case, although I felt like the thumbnails looked more pretty, not me, but the actual style of the thumbnail looked more pretty and, you know, was a bit more well designed. I felt it didn't actually lead to a higher click through rate in a lot of these cases. In this one it did, in this one it did, in this one it didn't, and this one it didn't. So just because something works for one as well doesn't mean like test it on one or two and then just blanket it across, you know, all of your content. It's definitely worth going through this process of trying it out. And I mean, in fact, if you were going to just apply this to all of your thumbnails anyway, for example, or to a batch of them, still go through and create them, but just run them all as split tests. And then once the split test has finished, you can just, it will just tell you, you know, do you want to assign this new thumbnail in place of the old one, you click the button and TubeBuddy will do it for you. So it's a really valuable tool for being able to click through, sorry, improving the click through rate, I should say, and get people onto your videos. And one of the things that I found interesting about that was that point about the retention once people are on the videos, you know, to see actually, once they click through on your thumbnail, how long are they actually staying on your video for? And that is the other, you know, crucial part of sort of growing your channel and driving views is getting people onto them, getting people to stay on them, and then getting people, presumably, also hopefully to stay till the end so that they click through onto some other thing. But that thing about the retention analysis is another key part. And there's a retention analysis tool in TubeBuddy which I'll be talking about in the next video. So I'll see you in that video.