it's really hard either way for small channels, sometimes i really can't understand RUclips. It's really frustrating. Some really stupid and bad quality youtube videos are getting millions of views. It's so unpredictable.
It's not as unpredictable as it seems. It's really all about what the audience wants. The audience unfortunately these days are more about quick hits like a dog falling off the stairs over something that might hold more meaning for them but they just don't know it yet. Don't get me wrong, I rather not make clickbaity videos and would be happy getting that million views over 20 years of hard work rather than create a quick one shot viral video that goes to a million and have 20 years of single digit views because I didn't take the time to learn about content creation
In my opinion, the biggest issue with this feature is that it only considers watch time. I conducted a test where I evaluated thumbnails simultaneously through RUclips and a third-party tool. In one case, RUclips's data showed no significant difference in watch time between my two thumbnails. However, one of the thumbnails had a 20% higher CTR. Naturally, RUclips should consider CTR as a determining factor when it finds no significant relationship between the thumbnails in terms of watch time. Unfortunately, it currently doesn’t do that.
I think creators get WAY too hung-up on thumbnails. By far the most important aspect of a video is simply the topic, if your video topic is especially interesting/intriguing for your audience then the quality of your thumbnail doesn't really matter all that much. Your video won't magically get millions of views just because of a great thumbnail - if the topic sucks then no one will click it anyway.
But the thumbnail plays a role when viewers are browsing because it's larger and easier to take in. Colors are easily seen apart from views and timeliness
I Agree but also disagree with your statement. If you have a video that's worth a million views but a thumbnail no one will click on then it gets 0 views. If you have a great thumbnail but a bad video then it will get a few views then die out quickly. Note that if you have a good following you will still get views even when the thumbnail is bad. I Sit at around a 7-10% CTR on avg and my videos get between 10-40k views but the videos with a bad thumbnail that fall below the avg cause my views to dip below my avg view rate also. Now the videos with a below avg ctr still have a 3-5 Min avg watch time so it isn't the video but the thumbnail. Now You are right on the part where the topic and video needs to be good. I have a video where the thumbnail had a 2% ctr but a 70% avg watch causing the video to receive 300k views but there were also other factors included where the video was short enough to be shareable and I was on a topic that was extremely high in trending. If I had anything to say to new creators its that work on a video idea and make it as good as possible and have friends watch it and give their opinions on the video. Once you have a great video then make 3 thumbnails that you like personally then go to 5 people and ask which one they like the most. If there is a clear winner then use that one and if there isn't then remake a thumbnail until its something people notice out of the other videos on their recommended. Also make sure the title is good to because if your title doesn't cause intrigue or curiosity then people may avoid it also.
I'm with you man. Seeing the my data and the data shown in this video, the thumbnails are close to identical in terms of the percent of people who click on it. That leads me to believe that any of the thumbnails that were posted would be sufficient for the video and topic given out by the creator. So maybe, just maybe all this hype about thumbnails is overblown. As long as you have a decent thumbnail, you will get clicks if the topic is relevant to the person seeing your thumbnail.
If anyone reading this, i hope you reach your goal really soon! Don't get demotivated! I always support small channels as i am also running a small channel! 😊🐨
The one video I used it on did horrible, youtube only showed it to a fraction of the people my other videos usually get shown to. So haven’t used it since. Still grinding on the road to 1k subscribers! 🕊️
It would probably be unwise to assume that was the only factor in its performance. Especially when you only tested it once. Once is not a good bellwether.
I got to 100 subscribers on this blog, wow that’s crazy!!! I want to thank each and every one of you. As I have said before, your support and love means the world to me! 🎉🎉🎉🎉🎉🎉🎉🎉🎉
The only thing about this is that I have a viral evergreen video (almost 450k views now) that has low average watch time because youtube themselves said that they were sharing this across a larger audience. Hence, not everyone will watch it fully, but some will. It is by this very metric that my video exploded because a lot of ppl were "satisfied". But with this same logic, YT is also saying that low avg watch time is a marker of ppl not being satisfied......confusing!
At the 3 min marker rn and I have to say that at first I thought the idea of such a tool would be cool to let you optimize what thumbnail gets the most clicks if youre starting out. However, I am really enthused by the reality that they decided to optimize what thumbnails align with videos in accurately illustrate and foreshadow the content you can expect to see. Because from the consumers perspective that is really annoying for most of the videos you see to be not the content that the thumbnail entices you into to click in the first place. Ive noticed much more prominent channels doing this and it has become a strategy which can be annoying. I myself wanted to structure my channel in this way where I didnt have to clickbait and instead just made interesting content that I am passionate about. So if implemented correctly Id be happy to see this happen personally. Although some people might not...
I use thumbnail testing for testing thumbnails with different languages to see which countries watches contents; for example, one with English, one with Spanish and one with no text at all.
To everyone starting their journey, trust the process, this stuff works and just takes time, I’m celebrating my first 100 subs!!! Hang in there everyone!! We will get there!!!
Seriously thank you for posting this. I have not yet had time as a new creator to even consider A/B testing as I have pokers elsewhere, but it does make sense that watch time is a primo metric for RUclips, and anything that increases that should be considered.
It makes perfect sense actually. CTR is great but if the promise of the thumbnail isn't met that may be the moment someone closes the app. It's all about getting viewers to see more ads.
I still use this on all my videos.. It's been pretty telling for me. I learned that me in the thumbnail performs worlds better than leaving me out, and text of just a few words does better than no text at all. pretty big margin differences on those for me.
I haven't tried the thumbnail testing, but I have noticed that I saw a video with a good thumbnail and a good title that I wanted to watch but I watch another one first. Then when I go back looking for that thumbnail its different and not as attractive to me but same title.
For starters, It has a limit factor for only channels that have the "features being legit" through out giving their valid ID of some sort/video verification. What a shame for RUclips.
I always get way less impressions compared to when I upload just one, hence I decided to not use this tool at all and just upload a new thumbnail every couple of hours 🙄
I agree with this video extremely. I've seen the exact same issue where I cannot tell which thumbnail is better and my thumbnails are definitely different sometimes enough so that it can't be that they get the same amount of interest other than if the tool is not giving them the same amount of impressions
I was excited to use it. Used it on several videos. Decided to stop because my overall CTR dropped significantly. And as you said, there's minimal difference in comparison results.
If its going off watch time and retention and not the whole CTR as well of it for a thumbnail - how can you get an A/B thumbnail test using non thumbnail data of the video....
I gave up on this feature. I did a testing on one of my videos and after the testing was done I chose the thumbnail that worked the best and RUclips didn't share my choice, it picked a different one 😢
Yeeees, been saying this since the feature dropped. Huge disappointed that was designed to benefit RUclips as a company on a whole, not invidivual creators.
I wish more people complained about the new square that blocks your view on the app if you are literally watching anything... i miss the old mini player and wish everyone would complain so they change it back or at least give an option
I tried it on a couple videos and basically got nothing out of it. So, it took more time for me since I had to dev out multiple thumbnails and the difference wasn't there so far as I could tell. Meh. I'll just keep doing what I'm doing. At the rate I'm going, I'll hit 100k subs in 25 years or so.
The A/B testing never worked for me. I would test combinations of my normal thumbnail, thumbnail without words and thumbnails with words. None were click bait, just tweaks to see if things like no words were better than words etc. My test would end with "The test finished without a conclusive result" My views for the tests were 1/4 to a 1/2 of what I normally would see. What I found. I just keep using the same few thumbnails style I have been using for years. The data provided by YT was useless and the main take away factor it crushed my views on the videos. Every video this low views was one that I tried the A/B testing every video with higher views is just uploading a video with no testing.
How can you say youtube is focused on viewer satisfaction following the total borking of the non-premium viewer experience via endless mid-roll adds. I have found myself clicking off videos and even unsubscribing from channels where a 16 minute video has ads before, an internal promotion and 3 or 4 ad breaks with 2 adds a time and adds after. And I have practically stopped sharing cool content with my friends. I don’t need a reputation as someone who shares bad experiences. 😢 What makes it worse - this is creator controlled. They - the people I like and trust and have followed and even sometimes done patron for - for ages are deliberately choosing to make my viewing experience SO much worse!!! And before any one gives me stick about RUclipsrs spending hours on making videos and needing paying - i agree i’d like to get some money for my work but like most RUclipsrs (90%) of them i don’t get a single solitary penny for the ads surrounding my videos - it ALL goes to RUclips - indeed on my channels I make animations - sometimes even hand drawn ones. Don’t come at me with how long it takes to make a video I know ALL about it. I have a video on my channel that took over 400 hours. 😂
I feel the A/B testing is for YT itself, not creators or viewers' satisfaction. Not long after it rolled out they started pushing AI production harder - particularly thumbnails. So yes, it seems to be about clickbait, but only in as much as using creators to crowd test how significant thumbs are to viewers generally. There is now a glut of ridiculously inaccurate AI thumbs on practical how to and tutorial videos. I'm guessing that's because YT is looking for any way to boost hours watched across the platform - that's what the advertisers are interested in, not *why* someone is watching. As far as I can tell, it's all about getting creators to bang out more, more, more, so YT keeps throwing out 'tools' to hook smaller channels into producing, so YT has more data in as short of a time as possible.
I am gonna play devil's advocate here RUclips is not the really easy to test out it is better that they release something to everyone then take in feedback and improve it than have it be in perpetual testing
@@joshuacarre06 I do actually agree but the UI changes mixed with new features is what is bugging me though I guess for the average viewer new creator tools like 3 minute shorts and A-B thumbnail testing don't matter a lot, but for me all those things together are a bit overwhelming
Imo, thumbnail testing might not be important for certain channels. But I must say, after studying the RUclips analytics for some time, a good ctr and a good avd must work together. If a good ctr but bad avd, or a good avd but bad ctr, it will just destroy a video, even how perfect the thumbnail were.
I miss the good old days where you would just upload any random video with no production at all no script. No thumbnail. No keywords, nothing. And you would get tons of views, likes and comments 😂
Before I came across this video, I would ask my own audience what style of thumbnail for a specific form of content, would work best via image poles. And how I would make these poles is I will take one video type, and re-use the thumbnail on 4 of those videos, and simply have those thumbnail styles be voted on
I tried it on a video I thought would do very well no matter what...during the testing period it did miserably but after the testing period it shot up quickly. It didn't help engagement at all and seemed to actually hurt it. As a super small channel I can't afford to go through that "test" again.
I tried it for a little while and was not at all impressed. I even deliberately created rubbish thumbnails and I found through my testing, which ever one was uploaded first always came out as the winner, even if it was deliberately rubbish. One thing I did notice while using the tester was a dramatic drop in Impressions, CTR & Watch Time. Sadly I feel my channel is also now struggling to recover from it as my watch time and Impressions still is not upto the mark it was even if the CTR, Viewer Duration & interactions such as Likes & Comments is higher than it used to be.
I have been testing this tool on my channel for a month. And I had a several times contravesial results. When I was comparing a thumbnail A vs B. And during the 1st test - A won. After I ran another test in 2 days. And then B won. But each time, after I was seeing a thumbnail after running the test results - the CTR and watching time were growing. So, I believe it works.
It's definitely not a good idea to start the test&compare feature right after uploading a new video. Unfortunately, that's what I did with my last video. It was almost not visible for 3 days, only when I stopped the test run did it start.
I just revanced mid video upload settings to see if content holds true. I'm not buying personality/style that's a copyscrub or worse, outright pulling thumbnail and content from other creators.
I always believed A/B thumbnails was going to be the option to have changing thumbnails on your videos. That would be a much better feature. Make 3 different thumbnails that rotate and keep things fresh.
I use A/B Thumbnail testing a lot on my newer videos, with little to no changes on some of them. Thank you for the tips on how I could possibly get even more views than what I normally get.
I just thought of another thing that's been bugging me. I've noticed RUclips recommending videos to me that I've already watched, and often there's a different thumbnail (particularly on VidIQ videos, and I assume you're testing). If I click on something thinking it's a new video, but I've already seen it, I'm clicking off. Which would affect the watch time of that particular thumbnail in a test. That's potentially throwing off the results. And why is RUclips recommending I watch something I've already seen, but with a different thumbnail? Doesn't make sense.
I support the idea of RUclips not encouraging clickbaits, but the algorithm can still use the watch time to decide which videos to push to more people. The thumbnail test thing was only supposed to help us creators to know how to visually improve our thumbnails. Bc people can only know if a video is good or bad after watching a little of it, so if a video is good but have very boring thumbnail and no one clicks it this might sabotage its potential.
Glory!!! After so much struggles I now own a new house with an income of $330,000.00 every month God has kept to his words, my family is happy again everything is finally falling into place.
Actually these all made perfect sense, I keep an eye on watch time and CTR, probably keep the A/B test rather than A/B/C (unless I have three killer ideas!) recently I had a video skyrocketing on impressions but hardly had any likes or views and I was wondering how the hell is RUclips showing this video lol
I’m so glad you said this because that’s the question I was going to ask can we go back and check a/b thumbnail testing?? Will they delete or just have to press edit on RUclips studio and will go from there…
I want A/B testing for thumnbails AND titles that will solely be based on CTR. Just help me find the option that get the most clicks. It is up to me to satisfy or fail the viewers when they do.
Nobody knows what they're doing over there. I still can't grasp why they would completely ignore Like/View ratio and Sub/View ratio. Those are INCREDIBLE metrics that provide the most precise data on whether my content hits the mark, yet youtube somehow disagrees. Lol, okay youtube. Keep NOT giving impressions to my videos with excellent ratios. Such nonsense. My vids with good ratios also hit youtube's preferred metrics of retention, ctr, and engagement. It makes no sense to ignore these metrics. Good luck, everyone!
So for a small channel like myself (just under 100 subscribers), is A/B testing okay to use or should I really try and focus on just 1 great thumbnail? I feel like thumbnails are my weakest skill and having the option to try 3 has given me more confidence that at least ONE of them will land, but if I'm overall hurting my channel by using the A/B testing, I'd like to wind that down.
Thanks for the details. It makes SENSE on why they did it on watch time and not just CTR. Ive used A/B testing on a few and only for sublet changes in the Thumbs. Cheers!
Interesting and helpful. Only used it once and did not get expected results. In a re-uploaded video after a few updates, I experimented by using same thumbnail as first video and a new thumbnail which I personally liked better. While the old thumbnail had the higher percentage, RUclips declared the newer one the winner and is the one still being used on the re-uploaded video.
I understand their point ofthem wanting to show how long people watched the vid after clicking but high watchtime doesnt always mean high average view time because if like a million people watch for a second the watchtime will be bigger than if 1000 people watch for a minute
I'm honestly surprised they don't have A/B title testing
Facts. A/B title testing would be incredible. I feel a title is just as important as a thumbnail
My exact comment pretty much, we got a/b thumbnail testing, but as we all know, It's a package, it should be A/B thumbnail and Title test
i think we would all love that
So much of this!
it's really hard either way for small channels, sometimes i really can't understand RUclips. It's really frustrating. Some really stupid and bad quality youtube videos are getting millions of views. It's so unpredictable.
It's not as unpredictable as it seems. It's really all about what the audience wants. The audience unfortunately these days are more about quick hits like a dog falling off the stairs over something that might hold more meaning for them but they just don't know it yet. Don't get me wrong, I rather not make clickbaity videos and would be happy getting that million views over 20 years of hard work rather than create a quick one shot viral video that goes to a million and have 20 years of single digit views because I didn't take the time to learn about content creation
Yes I don't understand either
In my opinion, the biggest issue with this feature is that it only considers watch time. I conducted a test where I evaluated thumbnails simultaneously through RUclips and a third-party tool. In one case, RUclips's data showed no significant difference in watch time between my two thumbnails. However, one of the thumbnails had a 20% higher CTR. Naturally, RUclips should consider CTR as a determining factor when it finds no significant relationship between the thumbnails in terms of watch time. Unfortunately, it currently doesn’t do that.
I've used it a few times, and my gut feel is that it throttles views, somehow. Those videos all got lower views than I expected.
Small channels let’s unite together, we got this ❤
Done and waiting 🎉
I think creators get WAY too hung-up on thumbnails. By far the most important aspect of a video is simply the topic, if your video topic is especially interesting/intriguing for your audience then the quality of your thumbnail doesn't really matter all that much. Your video won't magically get millions of views just because of a great thumbnail - if the topic sucks then no one will click it anyway.
But the thumbnail plays a role when viewers are browsing because it's larger and easier to take in. Colors are easily seen apart from views and timeliness
@@BradleyHallGuitar I would say thumbnails are actually super important of your channel is small but gets less important as a channel gets bigger
People perceive the thumbnail before they read the title. If it doesn't convey the topic, fewer people will know what it even is.
I Agree but also disagree with your statement. If you have a video that's worth a million views but a thumbnail no one will click on then it gets 0 views. If you have a great thumbnail but a bad video then it will get a few views then die out quickly. Note that if you have a good following you will still get views even when the thumbnail is bad.
I Sit at around a 7-10% CTR on avg and my videos get between 10-40k views but the videos with a bad thumbnail that fall below the avg cause my views to dip below my avg view rate also. Now the videos with a below avg ctr still have a 3-5 Min avg watch time so it isn't the video but the thumbnail.
Now You are right on the part where the topic and video needs to be good. I have a video where the thumbnail had a 2% ctr but a 70% avg watch causing the video to receive 300k views but there were also other factors included where the video was short enough to be shareable and I was on a topic that was extremely high in trending.
If I had anything to say to new creators its that work on a video idea and make it as good as possible and have friends watch it and give their opinions on the video. Once you have a great video then make 3 thumbnails that you like personally then go to 5 people and ask which one they like the most. If there is a clear winner then use that one and if there isn't then remake a thumbnail until its something people notice out of the other videos on their recommended. Also make sure the title is good to because if your title doesn't cause intrigue or curiosity then people may avoid it also.
I'm with you man. Seeing the my data and the data shown in this video, the thumbnails are close to identical in terms of the percent of people who click on it. That leads me to believe that any of the thumbnails that were posted would be sufficient for the video and topic given out by the creator. So maybe, just maybe all this hype about thumbnails is overblown. As long as you have a decent thumbnail, you will get clicks if the topic is relevant to the person seeing your thumbnail.
Subscribing to small channels and watching them grow feels good.
🎉🎉🎉
@TheFlixGlimpse 🎉🎉🎉🎉
thanks
True!
🎉🎉🎉
If anyone reading this, i hope you reach your goal really soon! Don't get demotivated! I always support small channels as i am also running a small channel! 😊🐨
Thanks! I subscribed.
Well, you're doing better than I am!😂
Good luck!
Thanks bro hope we reach there together
Best of luck. But everytime I do this "small channel gather" thing, I get a few subs...who immediately unsub within 24 hours. No point.
Thanks for 60 subscribers ❤😊
Happy Monday everyone, let’s make this last month count!
The thumbnail testing tool is either ahead of its time or complete nonsense leaving me just as confused as I was without it lol.
Same 😂😂
True 😂
It’s for big channels with a lot of return viewers. They give you the test data, then the winner goes out to new viewers
@@thehobbster I understand what it does, however the data is not near as useful as it could be! You should be able to dictate the test metric.
The one video I used it on did horrible, youtube only showed it to a fraction of the people my other videos usually get shown to. So haven’t used it since. Still grinding on the road to 1k subscribers! 🕊️
Hi😊
It would probably be unwise to assume that was the only factor in its performance. Especially when you only tested it once. Once is not a good bellwether.
It should have been based on CTR, rather than watch time 😿
I got to 100 subscribers on this blog, wow that’s crazy!!! I want to thank each and every one of you. As I have said before, your support and love means the world to me! 🎉🎉🎉🎉🎉🎉🎉🎉🎉
The only thing about this is that I have a viral evergreen video (almost 450k views now) that has low average watch time because youtube themselves said that they were sharing this across a larger audience. Hence, not everyone will watch it fully, but some will. It is by this very metric that my video exploded because a lot of ppl were "satisfied". But with this same logic, YT is also saying that low avg watch time is a marker of ppl not being satisfied......confusing!
At the 3 min marker rn and I have to say that at first I thought the idea of such a tool would be cool to let you optimize what thumbnail gets the most clicks if youre starting out. However, I am really enthused by the reality that they decided to optimize what thumbnails align with videos in accurately illustrate and foreshadow the content you can expect to see. Because from the consumers perspective that is really annoying for most of the videos you see to be not the content that the thumbnail entices you into to click in the first place. Ive noticed much more prominent channels doing this and it has become a strategy which can be annoying. I myself wanted to structure my channel in this way where I didnt have to clickbait and instead just made interesting content that I am passionate about. So if implemented correctly Id be happy to see this happen personally. Although some people might not...
I didn’t even know about this testing in the first place 😂
I had NO clue this was how it worked
I use thumbnail testing for testing thumbnails with different languages to see which countries watches contents; for example, one with English, one with Spanish and one with no text at all.
small youtubers,i hope your can reach your goals.I wish you the best of luck.💝
also good luck ❤
194 subbed
@@namalehusna ❤
Good luck to all small channels you can do this❤🎉
Thanks man.
Thank you!
This would be good for shorts I reckon. Thanks for your support all 656 subscribers! 🎉
To everyone starting their journey, trust the process, this stuff works and just takes time, I’m celebrating my first 100 subs!!! Hang in there everyone!! We will get there!!!
Let's hope our channels do well during the holidays as well.
The thumbnails are really important and looks like RUclips finally answering us 😅
Good luck everyone with their channel!
Maybe not perfect but I feel there is value in A/B thumbnail testing. I even wish a title A/B testing!
All the best to all in this RUclips journey 🙌🙌🙌
I was excited about thumbnail testing feature because I thought it will be helpful but I don't think it's that helpful now.
Those last # points were super helpful. Thank you!
Seriously thank you for posting this. I have not yet had time as a new creator to even consider A/B testing as I have pokers elsewhere, but it does make sense that watch time is a primo metric for RUclips, and anything that increases that should be considered.
It makes perfect sense actually. CTR is great but if the promise of the thumbnail isn't met that may be the moment someone closes the app. It's all about getting viewers to see more ads.
Will they add a thumbnail comparison feature to Shorts?
ps: thx for 289 subs!!
I still use this on all my videos.. It's been pretty telling for me. I learned that me in the thumbnail performs worlds better than leaving me out, and text of just a few words does better than no text at all. pretty big margin differences on those for me.
I still think they should bring back Direct Messaging. I remember when we were able to send messages back and forth...
Good luck everyone!
We small channels can do it - hang in there!
I haven't tried the thumbnail testing, but I have noticed that I saw a video with a good thumbnail and a good title that I wanted to watch but I watch another one first. Then when I go back looking for that thumbnail its different and not as attractive to me but same title.
For starters, It has a limit factor for only channels that have the "features being legit" through out giving their valid ID of some sort/video verification. What a shame for RUclips.
Thank you for sharing. ❤
I always get way less impressions compared to when I upload just one, hence I decided to not use this tool at all and just upload a new thumbnail every couple of hours 🙄
How's everyone's reach been last couple of weeks?
I have massive impressions, but my CTR is 0.2%. I don’t know what I’m doing wrong.
I agree with this video extremely. I've seen the exact same issue where I cannot tell which thumbnail is better and my thumbnails are definitely different sometimes enough so that it can't be that they get the same amount of interest other than if the tool is not giving them the same amount of impressions
I was excited to use it. Used it on several videos. Decided to stop because my overall CTR dropped significantly. And as you said, there's minimal difference in comparison results.
If its going off watch time and retention and not the whole CTR as well of it for a thumbnail - how can you get an A/B thumbnail test using non thumbnail data of the video....
I gave up on this feature. I did a testing on one of my videos and after the testing was done I chose the thumbnail that worked the best and RUclips didn't share my choice, it picked a different one 😢
Yeeees, been saying this since the feature dropped. Huge disappointed that was designed to benefit RUclips as a company on a whole, not invidivual creators.
Fewer click bait videos does serve us as a community, though.
Not to mention that if you are a small channel it just does not work. "You get the there is not enough data message".
I used it for almost every video when it first launched but haven’t used it for months now.
I wish more people complained about the new square that blocks your view on the app if you are literally watching anything... i miss the old mini player and wish everyone would complain so they change it back or at least give an option
I tried it on a couple videos and basically got nothing out of it. So, it took more time for me since I had to dev out multiple thumbnails and the difference wasn't there so far as I could tell. Meh. I'll just keep doing what I'm doing. At the rate I'm going, I'll hit 100k subs in 25 years or so.
The A/B testing never worked for me. I would test combinations of my normal thumbnail, thumbnail without words and thumbnails with words. None were click bait, just tweaks to see if things like no words were better than words etc. My test would end with "The test finished without a conclusive result" My views for the tests were 1/4 to a 1/2 of what I normally would see. What I found. I just keep using the same few thumbnails style I have been using for years. The data provided by YT was useless and the main take away factor it crushed my views on the videos. Every video this low views was one that I tried the A/B testing every video with higher views is just uploading a video with no testing.
How can you say youtube is focused on viewer satisfaction following the total borking of the non-premium viewer experience via endless mid-roll adds. I have found myself clicking off videos and even unsubscribing from channels where a 16 minute video has ads before, an internal promotion and 3 or 4 ad breaks with 2 adds a time and adds after. And I have practically stopped sharing cool content with my friends. I don’t need a reputation as someone who shares bad experiences. 😢
What makes it worse - this is creator controlled. They - the people I like and trust and have followed and even sometimes done patron for - for ages are deliberately choosing to make my viewing experience SO much worse!!!
And before any one gives me stick about RUclipsrs spending hours on making videos and needing paying - i agree i’d like to get some money for my work but like most RUclipsrs (90%) of them i don’t get a single solitary penny for the ads surrounding my videos - it ALL goes to RUclips - indeed on my channels I make animations - sometimes even hand drawn ones. Don’t come at me with how long it takes to make a video I know ALL about it. I have a video on my channel that took over 400 hours. 😂
I feel the A/B testing is for YT itself, not creators or viewers' satisfaction. Not long after it rolled out they started pushing AI production harder - particularly thumbnails. So yes, it seems to be about clickbait, but only in as much as using creators to crowd test how significant thumbs are to viewers generally. There is now a glut of ridiculously inaccurate AI thumbs on practical how to and tutorial videos. I'm guessing that's because YT is looking for any way to boost hours watched across the platform - that's what the advertisers are interested in, not *why* someone is watching. As far as I can tell, it's all about getting creators to bang out more, more, more, so YT keeps throwing out 'tools' to hook smaller channels into producing, so YT has more data in as short of a time as possible.
We are waiting for an update on how to gain more subscribers
We’d like that update too lol
@ChrysnScythe we definitely would like it
RUclips seems to love publishing unfinished features or badly thought out features
I am gonna play devil's advocate here
RUclips is not the really easy to test out it is better that they release something to everyone then take in feedback and improve it than have it be in perpetual testing
@@joshuacarre06 I do actually agree but the UI changes mixed with new features is what is bugging me though I guess for the average viewer new creator tools like 3 minute shorts and A-B thumbnail testing don't matter a lot, but for me all those things together are a bit overwhelming
Imo, thumbnail testing might not be important for certain channels. But I must say, after studying the RUclips analytics for some time, a good ctr and a good avd must work together. If a good ctr but bad avd, or a good avd but bad ctr, it will just destroy a video, even how perfect the thumbnail were.
Thanks for the video!
I miss the good old days where you would just upload any random video with no production at all no script. No thumbnail. No keywords, nothing. And you would get tons of views, likes and comments 😂
Before I came across this video, I would ask my own audience what style of thumbnail for a specific form of content, would work best via image poles.
And how I would make these poles is I will take one video type, and re-use the thumbnail on 4 of those videos, and simply have those thumbnail styles be voted on
I tried it on a video I thought would do very well no matter what...during the testing period it did miserably but after the testing period it shot up quickly. It didn't help engagement at all and seemed to actually hurt it. As a super small channel I can't afford to go through that "test" again.
I tried it for a little while and was not at all impressed. I even deliberately created rubbish thumbnails and I found through my testing, which ever one was uploaded first always came out as the winner, even if it was deliberately rubbish. One thing I did notice while using the tester was a dramatic drop in Impressions, CTR & Watch Time. Sadly I feel my channel is also now struggling to recover from it as my watch time and Impressions still is not upto the mark it was even if the CTR, Viewer Duration & interactions such as Likes & Comments is higher than it used to be.
I have been testing this tool on my channel for a month. And I had a several times contravesial results. When I was comparing a thumbnail A vs B. And during the 1st test - A won. After I ran another test in 2 days. And then B won. But each time, after I was seeing a thumbnail after running the test results - the CTR and watching time were growing. So, I believe it works.
It's definitely not a good idea to start the test&compare feature right after uploading a new video. Unfortunately, that's what I did with my last video. It was almost not visible for 3 days, only when I stopped the test run did it start.
I just revanced mid video upload settings to see if content holds true. I'm not buying personality/style that's a copyscrub or worse, outright pulling thumbnail and content from other creators.
I wish more of these features would be available on the mobile version of studio, even though as this video suggests it may not even Matter
When will we get more info and access to the foreign language dubbing
I always believed A/B thumbnails was going to be the option to have changing thumbnails on your videos. That would be a much better feature. Make 3 different thumbnails that rotate and keep things fresh.
Thats why i never use AB testing, i will change the thumbnail manually, the good old way is better.
Very strange decision on YT part. I Wondered why videos in which I used this didn’t do as well.
Small Channels gather here ❤
Me here 🌟
Here
Thankyou for the 261 Subscribers
I use A/B Thumbnail testing a lot on my newer videos, with little to no changes on some of them. Thank you for the tips on how I could possibly get even more views than what I normally get.
Idk why but many of my meme shorts get no boost at all, 0 views is all I see now
YT should have measured CTR and Watch Time. But thumbnails are there to elicit a click so... yeah pretty obvious should have measured CTR.
I just thought of another thing that's been bugging me. I've noticed RUclips recommending videos to me that I've already watched, and often there's a different thumbnail (particularly on VidIQ videos, and I assume you're testing). If I click on something thinking it's a new video, but I've already seen it, I'm clicking off. Which would affect the watch time of that particular thumbnail in a test.
That's potentially throwing off the results. And why is RUclips recommending I watch something I've already seen, but with a different thumbnail? Doesn't make sense.
best update: banning sniperwolf
@@ke__mot yes
Ong
U mean SSSniperWolf
@ omg mrbeast yeah i mean
RUclips runs on clickbait, encourages clickbait, and rewards clickbait.
I support the idea of RUclips not encouraging clickbaits, but the algorithm can still use the watch time to decide which videos to push to more people. The thumbnail test thing was only supposed to help us creators to know how to visually improve our thumbnails. Bc people can only know if a video is good or bad after watching a little of it, so if a video is good but have very boring thumbnail and no one clicks it this might sabotage its potential.
Glory!!! After so much struggles I now own a new house with an income of $330,000.00 every month God has kept to his words, my family is happy again everything is finally falling into place.
That's awesome!!! I know nothing about investment and I'm kind on getting started. What are your strategies?
Thanks to mrs Melissa Jonas Richard.
I'm surprised that this same name is being mentioned here, I stumbled upon one of his clients testimonies on CNBC news last week.
I know Melissa Jonas Richard, and I have also had success...
Absolutely! I have heard stories of people who started with little or no knowledge but managed to emerge victorious thanks to Melissa Jonas Richard
Actually these all made perfect sense, I keep an eye on watch time and CTR, probably keep the A/B test rather than A/B/C (unless I have three killer ideas!) recently I had a video skyrocketing on impressions but hardly had any likes or views and I was wondering how the hell is RUclips showing this video lol
So, as always, nobody has any answers, including RUclips themselves. Can we admit that this is a big roulette wheel?
Ive used it a couple of times, but testing for 2 weeks seems like a long time. most vids are gonna pop or not before then
I will be on it today and working on it today a/b thumbnail testing
Does RUclips actually push older videos if you change the thumbnail?
Thank you for your important notes
I’m so glad you said this because that’s the question I was going to ask can we go back and check a/b thumbnail testing?? Will they delete or just have to press edit on RUclips studio and will go from there…
I want A/B testing for thumnbails AND titles that will solely be based on CTR. Just help me find the option that get the most clicks. It is up to me to satisfy or fail the viewers when they do.
Nobody knows what they're doing over there. I still can't grasp why they would completely ignore Like/View ratio and Sub/View ratio.
Those are INCREDIBLE metrics that provide the most precise data on whether my content hits the mark, yet youtube somehow disagrees. Lol, okay youtube. Keep NOT giving impressions to my videos with excellent ratios. Such nonsense.
My vids with good ratios also hit youtube's preferred metrics of retention, ctr, and engagement. It makes no sense to ignore these metrics. Good luck, everyone!
So for a small channel like myself (just under 100 subscribers), is A/B testing okay to use or should I really try and focus on just 1 great thumbnail? I feel like thumbnails are my weakest skill and having the option to try 3 has given me more confidence that at least ONE of them will land, but if I'm overall hurting my channel by using the A/B testing, I'd like to wind that down.
Thanks for the details. It makes SENSE on why they did it on watch time and not just CTR. Ive used A/B testing on a few and only for sublet changes in the Thumbs. Cheers!
after 2-3 uses I stopped using it
Interesting and helpful. Only used it once and did not get expected results. In a re-uploaded video after a few updates, I experimented by using same thumbnail as first video and a new thumbnail which I personally liked better. While the old thumbnail had the higher percentage, RUclips declared the newer one the winner and is the one still being used on the re-uploaded video.
I understand their point ofthem wanting to show how long people watched the vid after clicking but high watchtime doesnt always mean high average view time because if like a million people watch for a second the watchtime will be bigger than if 1000 people watch for a minute
We are waiting for an update on how to gain more subscribers 😮
Kind of weird, it might confuse us
god this tool is such a pain in the butt
I havent used that future yet. I have switched them manually if i did... What I'm gonna try is changing thumbnail on my old videos...
As a Business Analyst A/B WORKS but then over time becomes something no one has focus or vision of
We felt it hurt our video, so we stopped using it
Wishing all small RUclipsrs good luck on your journey...