Great thumbnail! Thank you for mentioning that it takes two weeks for testing. I am testing a video now when you suggested it a week ago and it’s still running and I almost started to give up on it. Nothing lets you know how long it takes. I like how the channel you’re speaking of never loves or responds to any comments ever. That’s the MAIN reason I stopped watching their content. If you read this to end, respond with Go for it! How’s THAT for a test! 🤪
Actually, a test can be complete sooner. It depends on the results and how fast those results come in. Low CTR can result in requiring the full 2 weeks. Not enough clicks or inconclusive results, then you won't get results at all in that time. Click on "test results" on your selected video, then the 1st question mark for details on what Iam saying. Also, click the 2nd question mark for more information. Additionally, from there, you can click "learn more." VidIQ does heart and reply. They've explained in more than one of their videos (as they are suggesting you do the same) that they respond starting within the first several hours of uploads. They spend many hours reading, replying, and liking. They add that because of sheer volume of comments they receive, and the huge amount of time it takes, beyond 24 hours they may not be able do those things, especially respond. Think about how many comments you'd get if you had 2M subs. That would be difficult to keep up with, right? It's still a good resource and tool, but there are other good resources, and tools, but I'm not sure there are better tools. Smaller channels will likely be able to provide greater attention to subscribers like you are wanting. Best to you!
As a TH-cam viewer, the one and only reason I don't like a video is if I'm left feeling the Title & Thumbnail were clickbait. That's it. As long as the title delivers what I expect, I rarely care about the video quality. Clickbait is the number 1 reason I ever feel dissatisfied and stop watching that creator. I believe that TH-cam knows this. It's quite simple and doesn't need a deep dive into thought or experiment IMO. This is a good video Brian, BTW.😀
Vid iq is terrible and I will be cancelling my subscription with them. I actually asked a question once on their live stream, and Rob was extremely rude in his response. Unfortunately, they are not genuinely trying to support the community.
I agree with you Brian on making each thumbnail vastly different. There's no point in using it if you're gonna make slight changes. Also, I think TH-cam chose Watch time CTR because they prioritize watch time. It would make no sense to make a thumbnail that gets high CTR but then your video still ends up getting low AVD. I like Vid IQ but they need to be sure there info is backed up by data from TH-cam and not just their own theories.
Agreed on the vastly different thumbnails. I actually am using the same thumbnail on this video just so I can solely test out the text as I mention often in this video. It’ll be interesting to see the results. Thanks for your feedback. Voice dictated.
It's fine to have varying opinions as Brian has here, but given that TH-cam doesn't openly share the methodology of how and when it tests the thumbnails, everyone is left to theorize a bit and I don't think that's bad. Open discourse on anything that isn't completely obvious and transparent is a good thing. As I stated in my own comment here, gaming the system through CTR is a perfectly valid reason behind TH-cam's choice, but there are so many variables aside from the thumbnail and title that could influence a video's watch time. For example, we could be seeing a preferred thumbnail that has nothing to do with the thumbnail and everything to do with the time of day that the majority of your viewers organically watch your video. I In fact channel makers did a test where they released the same thumbnail in all 3 slots and got vastly different percentages. So if having an objective discourse leads TH-cam to make the tool more useful that's much better than "Trust me bro"
Closing crew here. To me, VidIQ is looking for a different audience than I am. They have become so full of silliness. My audience is basically very mature, (old) women like myself. I think they prefer a large font more than what the font says. 😂🤣😂🤣😂
I wish the A/B testing allowed for a creator to know how well each is doing over a shorter period than 2 weeks and if we want to keep the dual or trio thumbnail, we can.
I agree we need more data on A/B testing of thumbnails. Number of impressions for each thumbnail during test would be helpful. Good point, Brian, about the thumbnail text variations.
I know I'm not good at thumbnails either way. I definitely feel like it's a barrier for me. I've got one video currently at 75k views, but I can't recreate that momentum. I'm trying to get better. This was great content.
I watched your 75K more video. I wanted to see why it grew fast. I think it was because you hit a nerve among that group. Out of that GenZ group how many new subs did you get? I think you have a good site. A suggestion: Do another video addressing Gen Z. Like See if it also gets more than avg views. Also put over 40 in your later titles. It is trending. I subscribed to you because you seem to be real. I am starting a new channel next year which talks to new TH-camrs , (I Am keeping my current TH-cam too). 🌴🌴Aloha and Good luck
@@ADAtrips Thank you for taking time to watch and for the feedback. I appreciate that. I did a follow up video on Gen Z but it didn't get traction. The main difference is how the video was promoted by TH-cam. It's the only one I have that has been a "browse feature." That seems to be the big driver for this video. I have noticed that over 40 trend. That's great advice. From this video alone, I have grown about 300 subscribers. I haven't been doing TH-cam long so I'm still pretty new. That one video helped me tremendously. I took a look at your channel also. I'll have to spend some time on your videos, Hawaii is on our desired destination list. I didn't think it would be cold in the swimming pool in Hawaii as you mentioned That's interesting. Thanks good and good luck to you as well.
Thank you Brian. I went and watched VidIQ's video first and couldn't work out why I felt annoyed for most of it. You've hit the nail on the head multiple times here. Plus, you've clarified something for me (make the text really different) that I realise I need to address in my A/B testing. I love that tool, it is really helping me look at thumbnails in a more informed manner. And I love that you turbo drive than informed manner.
Hi, Brian, I got a question: after I done title, description, made my video and uploaded it, should I wait 24 hours before I post it on facebook and other platforms like discord groups etc. or should I be doing it immediatly? some say I need to let the algorithm do its thing, some say that's a myth, I'm not sure which one is it?
i'm going to chime in here. VidIQ is taking a conventional approach with subtle changes to see what actually effects user behavior in this case thumbnail clicks. For example a simple color. This is how you would split test landing pages, emails subjects etc because if you change too much at once you wont know what you changed that made a difference. Their mistake is for this to work they have to keep repeating the test tweaking slowly over time. You re taking a bit more innovative and i would say unconventional or daring approach and you can see differences faster based on your experience and the results pay off in your next videos. Great example of when to break conventional rules :) as for me, I use it to see text vs no text or text vs text in the same thumbnail. I like what you are doing man and maybe i will also give it a try :)
A very interesting discussion. I'm a data nerd and have been collecting my stats since the day I first published. You raise an interesting point. Does a higher CTR equal more watch time? What I want to explore is the relationship between CTR and Watch Time as my videos mature. If we assume one is NOT just putting up BS click bait titles and thumbnails - not always a safe assumption - it would seem logical that a better CTR would improve watch time. But being a data nerd I need to see the numbers before I make a decision. Thanks for giving me a fun brain teaser to mess around with.
Without question, CTR is the multiplier of views and watch time. If you get 100 impressions and a click your rate of 5% percent you get five views. However, if you get a click through rate of 10% you get 10 views Twice as many views, multiply that by the time and multiply your watch time by two
For my last video, I started A/B testing right after I published it. The video had generated almost no impressions and therefore hardly any views. i think it was not delivered by TH-cam. After 3 days I stopped the test and suddenly the views and impressions increased rapidly. Have you experienced this with your tests?
no my friend, i have not had any issues. I wonder how the metrics were to that video ? You got some impressions and views, how were those metrics overall?
@@BriangjohnsonTv Disastrous data: on the first day only 260 impressions and 15 views, in the next 2 days only 60 impressions were added. But no views! So I stopped Test&Compare. The next day the numbers tripled but I think the video is permanently damaged, because at the moment there are only about 15 impressions per day again. By comparison, the last video had 5200 impressions in the same period, the best 49,000 with 2200 views.
good thoughts on it but we have to realize that this tool is 50% complete, you see, the thumbnail is not the main factor for a click and stay, but the title, the title should also be able to use ABC test so for now yes it work but 50% only its potential.
Ive done this thumbnail test on all my videos after watching the very first time Brian posted a video about this, must be weeks ago, not sure if because I am a small channel, but there was no improvement, not a slight improvement after the test completed, it only selected the thumbnail but there was zero improvement or movement in all of my 80 videos.
I think the best way to see results is to test for an extended period, my friend. I’ve been using this strategy for quite a while and it’s starting to show results.
Agreed. I wasn't surprised about watch time share either. If youtube is going to give me more impressions on more watch time because the thumbnail resonates then that's the thumbnail I want. Not higher ctr with lower views. It would be cool to see how many impressions and what ctr each thumbnail gets though and what kind of audiences they got shared to. Also agree on the small changes. When I originally saw the video I was also thinking to myself "of course small changes will yield barely different watch time share".Great video
Yeah, it is true. A good CTR and a good AVD must work together. If one of them performs badly, it will not only destroy the video but also it will destroy a good thumbnail.
I never found a benefit from the A/B testing on my other channel (17.5k subs) . I think greater attention to the thumbnail in the first place. For those bashing VidIQ in the comments. I upgraded to Boost last month, worked through the courses, watched videos on the new niche and launched. Much greater uptake than on any channel before. Ranked number 1 on titles which was a first.
Seems like by using watch time, YT is comparing the sales proposition of the thumbnail relative to the content, not irrelevant eye catching that would happen if they were just measuring clicks. It'd be interesting to get some comparison between TubeBuddy's A/B testing and YT's. I wonder how close the numbers are.
I love your videos and I love your content, but I have to disagree on this. The testing compare should absolutely go on click through rate your thumbnail is there to get the click not the watch time the watch time is dictated by how good the video is so in that vidIQ video all their results are varied by two or 3%. How do you know which one is really beneficial without seeing how many people clicked on it compared to how many people seen it
Sometimes I think it may be worth thinking about split testing 2 things at a time rather than 3 so you can have a better understanding of what the difference maker was. The lines get a bit blurred when you add that 3rd variable into the equation.
I remember talking about this topic a few weeks ago with you and I mentioned it in one of your videos. I have heard about the test not being accurate on "channel makers" as well.
Thank you Brian for finally speak the truth. Vidiq, Nate Black, Think Media, Thubuddy and other so called "youtube gurus" I stopped listen to for a long time ago. They hold back information and do not speak the truth, I they did they would be out of business. In short, I listened to them at first when I was new, but then I realized what they were all about. So I did the opposite, and that's how I started to grow.
i’m gonna respectfully disagree my friend. I think all the channels you mentioned strive to bring helpful information to TH-camrs. Often times we may disagree on slight points but the fundamentals are almost always in alignment as far as what we believe works.Sean at ThinkMedia, the teams at TubeBuddy and VIDIQ and Nate … these are individuals in my humble opinion that strive to put out goodness into the world. I simply wanted to share my disagreement with one point that Dan made from VIDIQ. Regardless I’m sending you peace and power.
The thumbnail testing may not have significant impact on big channels, but it's worth it for small channels like mine. This further proved that all the privileges that TH-cam reserved for big channels are actually much more needed by small channels to grow. Although my channel is small, it's been around long enough to earn TH-cam's trust and some privileges, the thumbnail testing being one of them. Some creators have suggested Title Testing will have more impact than the thumbnail testing; hope TH-cam will give it a try.
Yes it is extremely confusing! I ended up unsubbing from the "yt growth" channels except for your channel Brian. I love how you break it down and give legit analytics and thoughts.
Short on time, but wanted to say Ive definitely seen value from the tool. More on the psychological aspect of the viewers mindset when clicking. The hookability if the thumbnail to get the click. Its definitely helped me make a few decisions that seem to be driving more views. ❤🎉
Very interesting video Brian. I just had a video 'take off' and do reasonably well in comparison to my other videos and I believe I struck against the criteria the youtube algorithm is looking too. This is too small a place to go over it all, so I'm going to stay at the 10,000 ft level. The two most important matrix I found were ctr and watch time. The final unmentioned matrix is time. The first 24 hours after a video is published. Why would TH-cam leave crumbs in the analytics page like letting you analysis your video in the first 24 hours? Hmmm. Anyhow, I found when I have a video obtain a ctr of at least 3.8% and a watch time hovering into 7 mins, in that magic 24 hour period, the video gets a lot of syndication. And it makes sense, when you consider what yt wants. They want viewers to consume more content, more often. If you provide videos that have a higher ctr and longer watch times, you are coming in line with their wants. Love to hear what others think on this vain.
Excellent video Brian and another opinion but thumbnails may not impact your watch time because you may have a good thumbnail but not get a long watch time because the video quality is more likely to either decrease or increase watch time. Thanks 👍 for sharing your video and thoughts on the vidiq video. Love 🥰 the video and love ❤️ Brian.
I think the ONLY thing that A/B testing can test and maybe one other. 1. Is CTR the only thing. 2. Is number of people that are NEW/OLD - Subscribers. I seldom watch VID/IQ anymore. I think they are only about $. Thought this was an interesting story. Aloha
VidIQ (the best 3rd party TH-cam tool), but their recent experiments are not scientifical. Rob once did a video about 3 minutes Shorts and conducted only one test. The whole video was about a single try. I was shaking my head through out the video.
For the most part, I'm interested in what you have to say in your videos, but I am REALLY irritated by that second camera, side-on view of you speaking that you've introduced. I've never seen that done on mainstream TV where information is being presented (documentaries, news reports etc); I guess there has to be a reason for that. Sorry for the negative feedback but it has stopped me hitting that subscribe button.
Hey, thanks for the feedback. Channel might not be for you then sadly. I love it and I need to stay true to myself. Most people don’t seem to mind and I’ve had a lot of nice comments. It’s a bummer that these small things can distract from a video, but I understand. Anyway, good luck to you.
I've seen some of vidIQ's videos and some are ok. I just like Yours Nick's and Nate's videos the best it may be the simplicity or ya'lls personalities. I long for the information and the help to how to grow your channels problem is there are lot of channels that make videos on how to make videos ( Mostly smaller channels) don't know what they are talking about
at least 90% of my videos flatline after the second hour lol, some do hit and really hit, and a lot of those flatlining quick, will get views for years due to search.
How can the A/B algorithm actually be any kind of tool with accuracy unless it sends your 2 thumbnails simultaneously to 2 groups of equal type…? And even then both groups of test subjects could be vastly different. It’s not without skepticism for me. Skill combined with a chance at going viral, just by chance, lends to a lot of dopamine excitement and expectation. The cream rises to the top and change is important when it comes to success. However, content creation is just that… data… content… and in the end we just wanna share for purpose of teaching or fame and or/make money. I don’t trust any of the tools, especially after I learned the C.anva connection. 😮
Totally agree on your stance about the content. I focused on that and preached that for so many years I guess now with this new tool I’m interested in testing. Not perfect by any stretch.
@@BriangjohnsonTvSometimes I put no thumbnail, no title and things go better than I expect. 🤷🏻♀️ I am not convinced there is a magic formula or that any testing or tools truly help. I am new though and I am humbled by your experience, Brian.
I left this comment on the VIDIQ video and I'll leave it here to get your perspective Gaming the system is an undeniable downside to the CTR approach, but let's keep brainstorming rather than picking a metric that may have nothing to do with the thumbnail. Once a user has clicked the thumbnail, watch time then has everything to do with the quality of your video and potentially nothing to do with the thing that got them there in the first place. A variable as simple as what time of day a thumbnail is tested could VASTLY affect the thumbnails success rate.
I’m not sure about time of day as far as impacting a video performance. I don’t really look at that myself. I tend to focus solely on the metrics. I talk about often. CTR and average duration and subscriber gain.
@@BriangjohnsonTv Sure obviously the core metrics are sound and I'm not saying that it's wrong in it's current iteration, only that there are other variables that could skew the "better thumbnail" that may not be factored into the winning one. Maybe they are? But since TH-cam doesn't share the entire picture of the testing process it is hard to know.
Tengo videos con ctr 3% y retencion 16% y traen cientos de visitas diarias. En cambio tengo videos con Retención sobre el 79% y ctr del 7% y estan muertos, hablan se lo mismo
Vid IQ is awful. When I ask for video suggestions, it's always something that's not even in my niche. Their thumbnail creator is so horrible as well... Chat Gbt makes them better.
It is a bit more complex than that because there are more factors involved. There is a reason for clickbait and there is a reason that TH-cam does not want clickbait. Let's say you made a video of 3 hours. You made two thumbnails: the first thumbnail tells the viewer exactly what he can expect, and the second thumbnail is more leaning towards clickbait. The honest thumbnail might not get many clicks but the people who do click are your loyal subscribers, who will watch the video to the end. The clickbait thumbnail gets a lot of attention from new people but only 10% of those viewers might actually like your video and watch until the end. The other 90% of the viewers click away and go do something else with their time. I am pretty sure that TH-cam wants you to use the honest thumbnail because they don't want 90% of their viewers leaving the platform. You as the creator on the other hand would love to get many new viewers on your channel. You don't care about the 90% that don't like your video. You care about the 10% that do like your video. If that is 10% of a big number and mostly new viewers that just discovered your channel, then that is great for you. By using the A-B testing, you give TH-cam the opportunity to pick the thumbnails that is the interest of TH-cam. If you only use one thumbnail you can lean towards clickbait to get more eyes on your channel.
Do you what works for you and not for others. No one medical formula that they work for everybody on TH-cam. That’s the truth and it’s also been scientifically proven. You know that saying you can’t please everybody it’s true so don’t bother. Just be yourself do what makes you happy and the people really wanna watch your videos will watch your videos. Don’t worry about subscriber count don’t worry about the views. Don’t worry about the algorithm or even the analytics. Your driver so crazy and you’ll be brought out And give up on TH-cam altogether. Truth bomb
Great thumbnail! Thank you for mentioning that it takes two weeks for testing. I am testing a video now when you suggested it a week ago and it’s still running and I almost started to give up on it. Nothing lets you know how long it takes. I like how the channel you’re speaking of never loves or responds to any comments ever. That’s the MAIN reason I stopped watching their content. If you read this to end, respond with Go for it! How’s THAT for a test! 🤪
Go For It! 🎉🎉🎉 ummm, guess you didn't mean me.
😂🤣😂🤣
@@JulieTorrens That was awesome Julie, thanks! 🤪😂😎 I did mean Brian, but that was super cool! 👍😎
Actually, a test can be complete sooner. It depends on the results and how fast those results come in. Low CTR can result in requiring the full 2 weeks. Not enough clicks or inconclusive results, then you won't get results at all in that time.
Click on "test results" on your selected video, then the 1st question mark for details on what Iam saying. Also, click the 2nd question mark for more information. Additionally, from there, you can click "learn more."
VidIQ does heart and reply. They've explained in more than one of their videos (as they are suggesting you do the same) that they respond starting within the first several hours of uploads. They spend many hours reading, replying, and liking. They add that because of sheer volume of comments they receive, and the huge amount of time it takes, beyond 24 hours they may not be able do those things, especially respond. Think about how many comments you'd get if you had 2M subs. That would be difficult to keep up with, right?
It's still a good resource and tool, but there are other good resources, and tools, but I'm not sure there are better tools.
Smaller channels will likely be able to provide greater attention to subscribers like you are wanting.
Best to you!
@@TeaTreeApothecary Great information thanks! 👍😎 I really appreciate it.
@4KVideoTours You're welcome. There are so many pages in analytics, it takes a while to discover them all.
As a TH-cam viewer, the one and only reason I don't like a video is if I'm left feeling the Title & Thumbnail were clickbait. That's it. As long as the title delivers what I expect, I rarely care about the video quality. Clickbait is the number 1 reason I ever feel dissatisfied and stop watching that creator. I believe that TH-cam knows this. It's quite simple and doesn't need a deep dive into thought or experiment IMO. This is a good video Brian, BTW.😀
Thanks for the feedback, appreciate that.
Vid iq is terrible and I will be cancelling my subscription with them. I actually asked a question once on their live stream, and Rob was extremely rude in his response. Unfortunately, they are not genuinely trying to support the community.
Sorry, you think that way my friend.
Agreed!
They are the best tool out there though. But know what you are doing and be making money before paying.
Goodluck, after i cancelled the subscribion the ad for it pops up over everything within yt
😮😮😮
I agree with you Brian on making each thumbnail vastly different. There's no point in using it if you're gonna make slight changes. Also, I think TH-cam chose Watch time CTR because they prioritize watch time. It would make no sense to make a thumbnail that gets high CTR but then your video still ends up getting low AVD. I like Vid IQ but they need to be sure there info is backed up by data from TH-cam and not just their own theories.
Agreed on the vastly different thumbnails. I actually am using the same thumbnail on this video just so I can solely test out the text as I mention often in this video. It’ll be interesting to see the results. Thanks for your feedback. Voice dictated.
It's fine to have varying opinions as Brian has here, but given that TH-cam doesn't openly share the methodology of how and when it tests the thumbnails, everyone is left to theorize a bit and I don't think that's bad. Open discourse on anything that isn't completely obvious and transparent is a good thing.
As I stated in my own comment here, gaming the system through CTR is a perfectly valid reason behind TH-cam's choice, but there are so many variables aside from the thumbnail and title that could influence a video's watch time. For example, we could be seeing a preferred thumbnail that has nothing to do with the thumbnail and everything to do with the time of day that the majority of your viewers organically watch your video. I In fact channel makers did a test where they released the same thumbnail in all 3 slots and got vastly different percentages. So if having an objective discourse leads TH-cam to make the tool more useful that's much better than "Trust me bro"
@chancethetechie good points
Closing crew here.
To me, VidIQ is looking for a different audience than I am. They have become so full of silliness. My audience is basically very mature, (old) women like myself. I think they prefer a large font more than what the font says. 😂🤣😂🤣😂
Exactly
I wish the A/B testing allowed for a creator to know how well each is doing over a shorter period than 2 weeks and if we want to keep the dual or trio thumbnail, we can.
I think the two-week window is so we don't overreact to early data.
I agree we need more data on A/B testing of thumbnails. Number of impressions for each thumbnail during test would be helpful. Good point, Brian, about the thumbnail text variations.
thanks Jackie.
I know I'm not good at thumbnails either way. I definitely feel like it's a barrier for me. I've got one video currently at 75k views, but I can't recreate that momentum. I'm trying to get better.
This was great content.
You can do it!
I watched your 75K more video. I wanted to see why it grew fast. I think it was because you hit a nerve among that group. Out of that GenZ group how many new subs did you get? I think you have a good site. A suggestion: Do another video addressing Gen Z. Like See if it also gets more than avg views. Also put over 40 in your later titles. It is trending. I subscribed to you because you seem to be real. I am starting a new channel next year which talks to new TH-camrs , (I Am keeping my current TH-cam too). 🌴🌴Aloha and Good luck
@@ADAtrips Thank you for taking time to watch and for the feedback. I appreciate that. I did a follow up video on Gen Z but it didn't get traction. The main difference is how the video was promoted by TH-cam. It's the only one I have that has been a "browse feature." That seems to be the big driver for this video.
I have noticed that over 40 trend. That's great advice. From this video alone, I have grown about 300 subscribers. I haven't been doing TH-cam long so I'm still pretty new. That one video helped me tremendously.
I took a look at your channel also. I'll have to spend some time on your videos, Hawaii is on our desired destination list. I didn't think it would be cold in the swimming pool in Hawaii as you mentioned That's interesting.
Thanks good and good luck to you as well.
I wonder if by using A/B testing youtube pushes your video more to get a result?
Wil never knows but I would get guest probably not myself
Thank you Brian. I went and watched VidIQ's video first and couldn't work out why I felt annoyed for most of it. You've hit the nail on the head multiple times here. Plus, you've clarified something for me (make the text really different) that I realise I need to address in my A/B testing. I love that tool, it is really helping me look at thumbnails in a more informed manner. And I love that you turbo drive than informed manner.
Glad I could help you understand this update, and I am really happy to hear the tool is helping you!
Such great insight into this,I actually stopped testing, and just try to be better at the thumbnail
I have been doing the test a lot and it has been giving me more viewers on my small channel. So will keep using it!
Awesome! I'm glad it's working for you.
Hi, Brian, I got a question:
after I done title, description, made my video and uploaded it, should I wait 24 hours before I post it on facebook and other platforms like discord groups etc. or should I be doing it immediatly? some say I need to let the algorithm do its thing, some say that's a myth, I'm not sure which one is it?
I don’t think it would make that much difference, my friend, honestly.
@@BriangjohnsonTv Thank you for the reply! 🔥
i'm going to chime in here. VidIQ is taking a conventional approach with subtle changes to see what actually effects user behavior in this case thumbnail clicks. For example a simple color. This is how you would split test landing pages, emails subjects etc because if you change too much at once you wont know what you changed that made a difference. Their mistake is for this to work they have to keep repeating the test tweaking slowly over time. You re taking a bit more innovative and i would say unconventional or daring approach and you can see differences faster based on your experience and the results pay off in your next videos. Great example of when to break conventional rules :) as for me, I use it to see text vs no text or text vs text in the same thumbnail. I like what you are doing man and maybe i will also give it a try :)
thanks for your thoughts!
A very interesting discussion. I'm a data nerd and have been collecting my stats since the day I first published. You raise an interesting point. Does a higher CTR equal more watch time? What I want to explore is the relationship between CTR and Watch Time as my videos mature. If we assume one is NOT just putting up BS click bait titles and thumbnails - not always a safe assumption - it would seem logical that a better CTR would improve watch time. But being a data nerd I need to see the numbers before I make a decision. Thanks for giving me a fun brain teaser to mess around with.
Without question, CTR is the multiplier of views and watch time. If you get 100 impressions and a click your rate of 5% percent you get five views.
However, if you get a click through rate of 10% you get 10 views
Twice as many views, multiply that by the time and multiply your watch time by two
For my last video, I started A/B testing right after I published it. The video had generated almost no impressions and therefore hardly any views. i think it was not delivered by TH-cam. After 3 days I stopped the test and suddenly the views and impressions increased rapidly. Have you experienced this with your tests?
no my friend, i have not had any issues. I wonder how the metrics were to that video ? You got some impressions and views, how were those metrics overall?
@@BriangjohnsonTv Disastrous data: on the first day only 260 impressions and 15 views, in the next 2 days only 60 impressions were added. But no views!
So I stopped Test&Compare.
The next day the numbers tripled but I think the video is permanently damaged, because at the moment there are only about 15 impressions per day again.
By comparison, the last video had 5200 impressions in the same period, the best 49,000 with 2200 views.
good thoughts on it but we have to realize that this tool is 50% complete, you see, the thumbnail is not the main factor for a click and stay, but the title, the title should also be able to use ABC test so for now yes it work but 50% only its potential.
Good point my friend, perhaps in the future that feature will be released.
Ive done this thumbnail test on all my videos after watching the very first time Brian posted a video about this, must be weeks ago, not sure if because I am a small channel, but there was no improvement, not a slight improvement after the test completed, it only selected the thumbnail but there was zero improvement or movement in all of my 80 videos.
I think the best way to see results is to test for an extended period, my friend. I’ve been using this strategy for quite a while and it’s starting to show results.
Agreed. I wasn't surprised about watch time share either. If youtube is going to give me more impressions on more watch time because the thumbnail resonates then that's the thumbnail I want. Not higher ctr with lower views.
It would be cool to see how many impressions and what ctr each thumbnail gets though and what kind of audiences they got shared to.
Also agree on the small changes. When I originally saw the video I was also thinking to myself "of course small changes will yield barely different watch time share".Great video
BINGO MY FRIEND.
Great work Brian! Always appreciate your insights 🙏
I appreciate that!
I'm 17 seconds in and I'm leaving! That's right....
because I've already watched this whole video and liked it recently.
LOL
Yeah, it is true. A good CTR and a good AVD must work together. If one of them performs badly, it will not only destroy the video but also it will destroy a good thumbnail.
Can't have watch time without CTR.
bingo.
I wish the would add different titles with the thumbnail to test
that would be great for sure.
I never found a benefit from the A/B testing on my other channel (17.5k subs) . I think greater attention to the thumbnail in the first place.
For those bashing VidIQ in the comments. I upgraded to Boost last month, worked through the courses, watched videos on the new niche and launched. Much greater uptake than on any channel before. Ranked number 1 on titles which was a first.
Great point!
Seems like by using watch time, YT is comparing the sales proposition of the thumbnail relative to the content, not irrelevant eye catching that would happen if they were just measuring clicks.
It'd be interesting to get some comparison between TubeBuddy's A/B testing and YT's. I wonder how close the numbers are.
Totally. I'm not a big believer in any tool telling me what to do as far as my creative content. I'm just using these to gather data.
I love your videos and I love your content, but I have to disagree on this. The testing compare should absolutely go on click through rate your thumbnail is there to get the click not the watch time the watch time is dictated by how good the video is so in that vidIQ video all their results are varied by two or 3%. How do you know which one is really beneficial without seeing how many people clicked on it compared to how many people seen it
Appreciate your positive feedback. Thank you for that. It’s nice when people can disagree in a respectful manner.
Sometimes I think it may be worth thinking about split testing 2 things at a time rather than 3 so you can have a better understanding of what the difference maker was. The lines get a bit blurred when you add that 3rd variable into the equation.
That is a good point and makes sense, my friend.
I remember talking about this topic a few weeks ago with you and I mentioned it in one of your videos. I have heard about the test not being accurate on "channel makers" as well.
Indeed
Thank you Brian for finally speak the truth. Vidiq, Nate Black, Think Media, Thubuddy and other so called "youtube gurus" I stopped listen to for a long time ago. They hold back information and do not speak the truth, I they did they would be out of business.
In short, I listened to them at first when I was new, but then I realized what they were all about. So I did the opposite, and that's how I started to grow.
i’m gonna respectfully disagree my friend. I think all the channels you mentioned strive to bring helpful information to TH-camrs. Often times we may disagree on slight points but the fundamentals are almost always in alignment as far as what we believe works.Sean at ThinkMedia, the teams at TubeBuddy and VIDIQ and Nate … these are individuals in my humble opinion that strive to put out goodness into the world. I simply wanted to share my disagreement with one point that Dan made from VIDIQ. Regardless I’m sending you peace and power.
Don't know if it is the thumbnail or not, but I get less and less vieuws instead of growing.
The thumbnail testing may not have significant impact on big channels, but it's worth it for small channels like mine. This further proved that all the privileges that TH-cam reserved for big channels are actually much more needed by small channels to grow. Although my channel is small, it's been around long enough to earn TH-cam's trust and some privileges, the thumbnail testing being one of them. Some creators have suggested Title Testing will have more impact than the thumbnail testing; hope TH-cam will give it a try.
good points for sure! thanks for sharing.
Yet TH-cam is now awash in clickbait.
Hello brian wat tool u use?
canva, lightroom, topaz gigapixel
Great job Brian breaking it down. I would agree they did not go deep enough in their testing!
Yes it is extremely confusing! I ended up unsubbing from the "yt growth" channels except for your channel Brian. I love how you break it down and give legit analytics and thoughts.
Appreciate you, my friend, that means a lot!
Short on time, but wanted to say Ive definitely seen value from the tool. More on the psychological aspect of the viewers mindset when clicking. The hookability if the thumbnail to get the click. Its definitely helped me make a few decisions that seem to be driving more views. ❤🎉
Glad it's helping you, thanks for sharing!
Thank you Brian. I always learn so much from you 💯
glad to hear my friend.
Got paid last month. Couldn't have done it without you bro!
Dude, that’s amazing! Nice work
I have used it one various videos and all thumbnails were close to 33% click through, seems I'm good without it lol
that is amazing!! wow!
@@BriangjohnsonTv hehe true, but I'd rather have it show me at least something I could improve 😁
Very interesting video Brian. I just had a video 'take off' and do reasonably well in comparison to my other videos and I believe I struck against the criteria the youtube algorithm is looking too. This is too small a place to go over it all, so I'm going to stay at the 10,000 ft level. The two most important matrix I found were ctr and watch time. The final unmentioned matrix is time. The first 24 hours after a video is published. Why would TH-cam leave crumbs in the analytics page like letting you analysis your video in the first 24 hours? Hmmm. Anyhow, I found when I have a video obtain a ctr of at least 3.8% and a watch time hovering into 7 mins, in that magic 24 hour period, the video gets a lot of syndication. And it makes sense, when you consider what yt wants. They want viewers to consume more content, more often. If you provide videos that have a higher ctr and longer watch times, you are coming in line with their wants. Love to hear what others think on this vain.
Interesting point about the importance of the first 24 hours, I’ve never thought about it that way!
Your lighting is phenomenal!
Thank you for that. It’s one of my guilty pleasures.
This was actually very helpful. I will do a little testing of all options. Thank you 😊
Glad it helped
When it come to thumbnail, is there anyone as good as Brian G? Only one or three words sometimes for over 5 years now.
Last update is weird! Great scores, great numbers 1st hour then TH-cam fails to suggest. 90%+ in first hour then nothing. Weird
I totally agree 👍 well done Brian
Thanks for that Suzy I appreciate you.
"It's freakin CTR" 😅😅😅 that was hysterical!
LOL
Excellent video Brian and another opinion but thumbnails may not impact your watch time because you may have a good thumbnail but not get a long watch time because the video quality is more likely to either decrease or increase watch time. Thanks 👍 for sharing your video and thoughts on the vidiq video. Love 🥰 the video and love ❤️ Brian.
Absolutely my friend. Both average view duration, and the CTR impact overall watch time. Thanks for the valuable feedback Friend.
Brian G at his best, thanks for the info Brian
You bet
I think the ONLY thing that A/B testing can test and maybe one other. 1. Is CTR the only thing. 2. Is number of people that are NEW/OLD - Subscribers. I seldom watch VID/IQ anymore. I think they are only about $. Thought this was an interesting story. Aloha
You're right, the CTR is a key factor. Aloha!
I e tried the a/b testing and each time it didn't work. "CLOSING CREW"
did you use the methods mentioned in this video? Did you create very different looking thumbnails and or ‘compelling text’ …
@ I did and it came back inclusive
Yes closing crew and thanks for the straight forward information
Appreciate it!
Look at you with a multiple camera set up... loving it.
more views, more fun!!
VidIQ (the best 3rd party TH-cam tool), but their recent experiments are not scientifical.
Rob once did a video about 3 minutes Shorts and conducted only one test. The whole video was about a single try. I was shaking my head through out the video.
13:12 that reaction
Yeah, I totally agree that their testing methodology needs to be way more robust.
Vid IQ to me isnt as good as they were. I feel like theres been a change
sorry to hear that …
Great and very interesting video. Thanks for sharing.
Glad you enjoyed it
I mean 37% compared to 32% is actually huge and will make a difference over time! His logic makes no sense!
yes!
Think only good thing they got at the moment is thumbnail feature they have but I could get better .
For the most part, I'm interested in what you have to say in your videos, but I am REALLY irritated by that second camera, side-on view of you speaking that you've introduced. I've never seen that done on mainstream TV where information is being presented (documentaries, news reports etc); I guess there has to be a reason for that. Sorry for the negative feedback but it has stopped me hitting that subscribe button.
Hey, thanks for the feedback. Channel might not be for you then sadly. I love it and I need to stay true to myself. Most people don’t seem to mind and I’ve had a lot of nice comments. It’s a bummer that these small things can distract from a video, but I understand. Anyway, good luck to you.
Yep unsubbed to VidIQ a long time ago.
sorry to hear that my friend …
uT’oH… I JUST subscribed to VidIq, & I had high expectations for the tool.. Definitely tapping into this video 🤣
🤞
I hope this video helps!
I've seen some of vidIQ's videos and some are ok. I just like Yours Nick's and Nate's videos the best it may be the simplicity or ya'lls personalities. I long for the information and the help to how to grow your channels problem is there are lot of channels that make videos on how to make videos ( Mostly smaller channels) don't know what they are talking about
Agree. So many channels are just trying to get views.
at least 90% of my videos flatline after the second hour lol, some do hit and really hit, and a lot of those flatlining quick, will get views for years due to search.
That’s why I’m doubling down on search myself my friend. I’ve got a very specific strategy and it’s starting to work and it’s pretty exciting.
How can the A/B algorithm actually be any kind of tool with accuracy unless it sends your 2 thumbnails simultaneously to 2 groups of equal type…? And even then both groups of test subjects could be vastly different. It’s not without skepticism for me. Skill combined with a chance at going viral, just by chance, lends to a lot of dopamine excitement and expectation. The cream rises to the top and change is important when it comes to success. However, content creation is just that… data… content… and in the end we just wanna share for purpose of teaching or fame and or/make money. I don’t trust any of the tools, especially after I learned the C.anva connection. 😮
Totally agree on your stance about the content. I focused on that and preached that for so many years I guess now with this new tool I’m interested in testing. Not perfect by any stretch.
@@BriangjohnsonTvSometimes I put no thumbnail, no title and things go better than I expect. 🤷🏻♀️ I am not convinced there is a magic formula or that any testing or tools truly help. I am new though and I am humbled by your experience, Brian.
There is certainly no magic formula. Only skills we can learn that can help improve things so we can reach a bigger and broader audience.
I left this comment on the VIDIQ video and I'll leave it here to get your perspective
Gaming the system is an undeniable downside to the CTR approach, but let's keep brainstorming rather than picking a metric that may have nothing to do with the thumbnail. Once a user has clicked the thumbnail, watch time then has everything to do with the quality of your video and potentially nothing to do with the thing that got them there in the first place. A variable as simple as what time of day a thumbnail is tested could VASTLY affect the thumbnails success rate.
I’m not sure about time of day as far as impacting a video performance. I don’t really look at that myself. I tend to focus solely on the metrics. I talk about often. CTR and average duration and subscriber gain.
@@BriangjohnsonTv Sure obviously the core metrics are sound and I'm not saying that it's wrong in it's current iteration, only that there are other variables that could skew the "better thumbnail" that may not be factored into the winning one. Maybe they are? But since TH-cam doesn't share the entire picture of the testing process it is hard to know.
AB testing could hurt views if one of the thumbnails is bad, but I don't think it is a major thing, just remove that thumbnail and keep testing.
Yeah, I totally agree. It makes sense for me to have a thumbnail test always.
Thanks for the information.
Any time!
And we can easily "over analyse" too. I just think VidIQ just like making very long videos to strech out the time.
I’m not so sure about that myself. But I’m glad you shared your thoughts and I appreciate your feedback.
I just had to watch this again
awesome!
Yep, I really do not like click bait and when this happens too much I unsubscribe 10:09 😊
thanks for the feedback. not sure why you think this is clickbait ... but, good luck to you.
I believe you are so right Brian, His 2 million subs is too far out from my measly 78. LOL
small steps add up over time …
Tengo videos con ctr 3% y retencion 16% y traen cientos de visitas diarias. En cambio tengo videos con Retención sobre el 79% y ctr del 7% y estan muertos, hablan se lo mismo
Vid IQ is awful. When I ask for video suggestions, it's always something that's not even in my niche. Their thumbnail creator is so horrible as well... Chat Gbt makes them better.
Don’t think VIDIQ is awful, simply did not agree with there stance on the testing tool.
It is a bit more complex than that because there are more factors involved. There is a reason for clickbait and there is a reason that TH-cam does not want clickbait. Let's say you made a video of 3 hours. You made two thumbnails: the first thumbnail tells the viewer exactly what he can expect, and the second thumbnail is more leaning towards clickbait. The honest thumbnail might not get many clicks but the people who do click are your loyal subscribers, who will watch the video to the end. The clickbait thumbnail gets a lot of attention from new people but only 10% of those viewers might actually like your video and watch until the end. The other 90% of the viewers click away and go do something else with their time. I am pretty sure that TH-cam wants you to use the honest thumbnail because they don't want 90% of their viewers leaving the platform. You as the creator on the other hand would love to get many new viewers on your channel. You don't care about the 90% that don't like your video. You care about the 10% that do like your video. If that is 10% of a big number and mostly new viewers that just discovered your channel, then that is great for you. By using the A-B testing, you give TH-cam the opportunity to pick the thumbnails that is the interest of TH-cam. If you only use one thumbnail you can lean towards clickbait to get more eyes on your channel.
thanks for sharing!
Do you what works for you and not for others. No one medical formula that they work for everybody on TH-cam. That’s the truth and it’s also been scientifically proven. You know that saying you can’t please everybody it’s true so don’t bother. Just be yourself do what makes you happy and the people really wanna watch your videos will watch your videos. Don’t worry about subscriber count don’t worry about the views. Don’t worry about the algorithm or even the analytics. Your driver so crazy and you’ll be brought out And give up on TH-cam altogether. Truth bomb
Thanks for sharing your thoughts
@ You’re welcome
I'm coming back every week brother
Appreciate you, my friend!
if you dont have tausend of views - this tool is useless
You don’t understand TH-cam like you think you do my friend. Regardless I appreciate your feedback.
@@BriangjohnsonTv look i tried this abc thumbnail testing multiple times. one video had
Vid iq is no good to me as i do everything on the phone. No laptop
I use VidIQ on my mobile but also you can put your phone into desktop mode as I do for the YT studio to look at my analytics
I am new here! Just subscribed
Thanks for subbing!
first
Hi Brian. Very nice sharing. I am new to content creation in You Tube. I subscribe to your channel👍
Thanks for subscribing and welcome to the channel!
I mean 37% compared to 32% is actually huge and will make a difference over time! His logic makes no sense!
bingo!