Transcription
If nobody's clicking on your videos, at some point you probably wondered, "YouTube, are you actually showing these to the right people?" So, I actually asked. I used YouTube's own AI app studio to audit my channel's data, and the answer was yes. YouTube was showing my videos to the completely wrong audience. So, now I am showing you those exact reports so you can audit your own channel, plus the three-step fix to finally point the algorithm to the right audience.
Now, when you hit publish, YouTube has no idea who wants to watch your videos. So, for the first 48 to 72 hours it is seeding out your videos to various audience groups to basically just see who bites. So, if you look in your impressions graph, you'll see a bunch of spikes. That's seeding. It's basically YouTube acting like a detective, but with no clues. So, our job is to give the detective better clues so it can find the right rooms, the right audiences. But, how do we even know that we're giving it the right clues?
Now, this is the first report you are going to do. I want you to pick a video that didn't perform well, and then we're going to go into its analytics. So, I'm going to do this one here. You go into analytics, reach, and you're going to scroll down until the content suggesting this video section, and then click see more. I want to show you who YouTube what other videos YouTube was recommending my content next to. Why the declining birth rate is none of Gen Z's business. Clearly related to content creation, 0% click-through rate. Not surprised there. Nobody's pretty in person anymore. Oh, I got two views. I got a 1.2% click-through rate there. For the guy who accidentally abandoned his a woman to die in the Alps, I got three views from that uh with a average view duration of 31 seconds. So, clearly the people interested in that were not interested in my content. The Love Island Slop has taken over TikTok. I got a one view from there, and they stayed for 23 seconds. Almost instantly, I can see YouTube had no idea who this video was meant for. It was just pairing it beside literally the most random assortment of videos. But, if they watch video A and it's on just marrying your son already, people aren't going to click on video B. They see my video B being recommended, and they're like, "Oh, that's not for me. That's not for me. I prefer other things." So, if three out of this top five suggested videos have nothing to do with your niche, you have not given the detective enough clues yet. So, the system reads it as nobody wants to watch your video. But, really it's just the Love Island people don't want to watch my video.
This is our next report. So, pick another video and go into its analytics. So, I'm going to do this one here. And I want you to go into reach tab. And first, we'll pause here for a second cuz I want to I want to explain something. This click-through rate number that you see is not actually a single number. It is a combined click-through rate of the various traffic sources because every traffic source on YouTube has its own separate click-through rate. And the mismatch and why you're not getting a good click-through rate often lies in the gap between those sources.
Now, I'm going to scroll down into the how viewers find this video section and click see more. Now, this is where we're going to find the data that we're looking for. If we look right here, we'll see that our my browse click-through rate is 8.1%. That's really good. That means that my title and my thumbnail were working for people who knew me, who knew the type of content. But, if you look at the suggested videos click-through rate, 1.9% That's not good, right? Like that's a huge difference. So, what does that mean? Suggested is YouTube trying to find me new people, and at 1.9% well, I don't think the people it's finding me are the right people cuz they're clearly not clicking on my videos. So, a high browse and a low suggestion means that existing fans are loving it, but YouTube's struggling trying to find new fans. I am not giving YouTube enough clues on this video to help it find new audiences. And there you can clearly see how the click-through rate is the combined average of all these sources, and suggested is at the very bottom.
The last report I want you to look at. Good. Pick a video, go into the analytics, click on reach, and then click on the impressions click-through rate tab. And I want you to look at this graph here or chart. Is it a chart? It's a line. What you're looking at is as your impressions are getting higher, your average click-through rate is going to start dropping because that means it is seeding it to more and more newer viewers, more and more different audiences. So, if we look at this chart from the click-through rate, it's all over the place. So, some audiences are really high, some are really low. And when you get these like zigzags that are kind of all over the place, you know that YouTube's having a really hard time trying to figure out who your audience is. Some click-through rates are 6.8%, some are 2.2, there's a 1.8, a 5.1. A more steady line means that YouTube is having more success finding your audience. A sleep steep slide down or this zigzaggy means it's it's it's struggling and it's getting it wrong. It's getting it right and getting it wrong.
So, I asked YouTube's AI the obvious next question, "What do I actually do to change this? How can I improve my suggested video click-through rate?" And it came back with three fixes. And I want you to notice that not one of them has anything to do with the thumbnail. They were all related to the title. Just telling YouTube exactly who the video was for.
So, the first fix it recommended was to cut vague words and just name the actual thing that we're talking about. So, what does this mean? No tips, tricks, secrets, hacks, growth. Like, what does growth even mean? Growth could mean like growing a garden, growth in stocks, growth in hair, personal growth. It's so vague. Like, it names the action but not the actual thing that we're growing. So, a title like skin-care tips and tricks becomes the double cleanse routine for oily skin. More specific, right?
So, our second fix is to stop making the title only about you and name who it's for. And we want to be careful here because the trap isn't in the words like my, me, I. I know I use those a lot in my titles. The fact that there is no signal for who else would be interested in this. There are two ways that we can do this. The first is you can name it outright. Now, my video I'm tired of seeing small channels fail literally says small channels, right? Like, that's the label. It's It's pretty obvious. The second way that you do this is by placing them by their situation. My first 30 days on YouTube. I never say this is for new creators in the title, but if you've only been on YouTube for 30 days, then that would mean that I'm a new creator. This is for new creators. So, it it kind of like situationally applies it there. Both of these count. And what flops is a title that does neither of them. Like, my workout routine tells me nothing. Who are you? Why do I care about your workout routine? My go-to workout for total beginners. Ah, okay. This is a total beginners workout routine. I'm a total beginner. I'm looking for a workout.
Fix number three, dropping the broad category killer words. These are things like money, success, video, recipe. Single words that are too big for a small channel to win. If I did a video on like how to make money on YouTube, now suddenly I that video is getting placed next to Ali Abdaal, Think Media, Mr. Beast, right? Like all of these massive channels. There's no way that my little video is going to be able to compete next to that. It's going to be so much harder. Here's another example. So, if you were cooking creator, easy dinner recipes. You are now fighting all of food YouTube on that one. But, 30-minute sheet pan dinners for busy parents, that is a super specific video targeting a super specific person. YouTube's going to have a lot easier time finding a better fit for you, for your audience. And then you will get a higher suggestive video click-through rate. So, more people will watch.
And the simplest way to think about this, what AI was doing with all three of these fixes was just naming a who, so who the video is for, and a what, what it's actually about. So, your first 40 title characters and your first description line will do the heavy lifting on the who and the what of the video. If I did a my secret YouTube tips video, no, that's no good. What's the who? What's the what? The metadata audit for small channels under 1K. Maybe not the best title, but we'll use it for the example because you can clearly see the who, which is small channels under 1K, and you can see the what, which is the metadata audit.
Now, if you have been digging at Studio as much as I have and you want to see the exact prompts that I've been using to help break down my channel's data and really help me analyze what's going on, I put together a full-on like prompt packet. It's copy and paste, so it's super easy, and the links will be down in the description if you want to check that out.
So, after learning this, I actually went back and I redid all the titles for my lowest-performing videos to make sure that I had a who and a what in them. Because also, YouTube never gives up on your videos, truly. On my old channel, I had videos that were years old that were flops. Like, they were complete flops. And I was like, "You know what? Let's go back. Let's change up the packaging. New title, new thumbnail." They started growing again, and they became un-flops. You also need to be patient and give YouTube time, because it takes about 48 hours to process metadata changes. So, sit back, relax, don't go crazy. I actually tracked all my numbers from the first 30 days of YouTube, and some of it actually really surprised me. So, if you want the honest, unfiltered version of what starting at zero really looks like, you're going to want to watch that right here.