A Test - Does Promoting = Less Views?...

7:00 - 34%
23:40 - 25%
13:08 - 36%
17:26 - 28%
11:25 - 38%

I have cards, annotations, and use the "secret tag" method which nobody really believes me about lol shhhhh
 
My promotion primarily consists of adwords advertising. They are already immersed in the Youtube experience, going from video to video, and they see a relevant thumbnail/title (my ad) on the right at the top of all of the other suggestions. It is the same subject matter as those other suggestions. Once they actually click and get to my video it's no different than if they clicked any other of those suggestions. No?

Yeah, that should have similar results as to if it's a click from the suggested vids list unless people are in some way influenced by knowing it's an ad, but I doubt that. For me, the type of promotion that does more harm than good (in the long-term) is things like reddit, forums, social media, email,YT comments.



Coming back to video length etc, my theory is that the algorithm looks at the "relative retention rate", ie it compares a video's retention with the average benchmark figure for other videos with a similar length (and I also think it groups them by topic too) So basically, I believe our videos' retention rates are being compared to other videos' retention rates which have a similar length and similar topic. Those in the top x% get a chance of a spot in the suggested videos list.

Where long videos have an advantage is the CPM. It's the preroll ads which earn the most $$ and in particular, the non skippable ads, but nobody is going to sit through a 30 second ad to watch a 2 minute video. In fact I'm not sure an advertiser would want to place his ad on such a short video or even if he can if he wants to.
 
Yeah, that should have similar results as to if it's a click from the suggested vids list unless people are in some way influenced by knowing it's an ad, but I doubt that. For me, the type of promotion that does more harm than good (in the long-term) is things like reddit, forums, social media, email,YT comments.



Coming back to video length etc, my theory is that the algorithm looks at the "relative retention rate", ie it compares a video's retention with the average benchmark figure for other videos with a similar length (and I also think it groups them by topic too) So basically, I believe our videos' retention rates are being compared to other videos' retention rates which have a similar length and similar topic. Those in the top x% get a chance of a spot in the suggested videos list.

Where long videos have an advantage is the CPM. It's the preroll ads which earn the most $$ and in particular, the non skippable ads, but nobody is going to sit through a 30 second ad to watch a 2 minute video. In fact I'm not sure an advertiser would want to place his ad on such a short video or even if he can if he wants to.
I removed all my non skippable adds but would it be with trying them again on 1 half hour video
 
For me, the type of promotion that does more harm than good (in the long-term) is things like reddit, forums, social media, email,YT comments.
I agree, except for videos with potential viral content. For viral-type videos I think some of those other platforms are better choices. I've had a number of videos go viral (over 1 million views in a short period of time) with a jumpstart from Reddit posts. From my experience if you have a quality viral-type video and get it exposure anywhere and everywhere, those views will typically result in very quick suggested video placement. For a couple of them I actually paid a well established Reddit account $100 to post my video. They are very selective and turned down one of my videos but the other two they took my money and it ended up being a good investment as both videos surpassed 1 million views within 2 weeks (a couple hundred thousand from Reddit traffic and then suggested video traffic). I have another viral-potential video that I will be doing the same thing with shortly.
my theory is that the algorithm looks at the "relative retention rate", ie it compares a video's retention with the average benchmark figure for other videos with a similar length
Well, what we're told by Youtube is that whichever video leads to an overall longer viewing session will be ranked higher, regardless of length. That seems to make the most sense. Relative retention rate might also be one of the things the algorithm takes into account, but ultimately if a 1 minute video with 50% retention consistently leads to a longer viewing session than a 10 minute video with 80% retention, the 1 minute video will be ranked higher by the algorithm. That seems to make the most sense, since it completely solves the problem of whether to favor videos of a certain length. Of course, it's not anywhere near as simple as we are making it seem. In reality these algorithms are unbelievably complex and they tinker with them constantly. As of right now 'longer viewing session' is the single biggest factor in ranking (according to Youtube) but that could change by next week.
 
Great thread. It's one thing to guess or speculate, but hard data always sets things right.
Well, I think it must depend on a number of factors. No one else has reported anywhere near the increase in sub and cc views as OP has, percentage-wise (I'm speaking about those who already had subs enabled). I would like to know the geographic statistics of OP and some of the others to see if that also plays a part. Also type of video and whether the commentary is very important to the viewing experience.
 
Yep, those will factor into it. You will need data from many different channels to come to a conclusive conclusion, but this is a good start.

Well, I think it must depend on a number of factors. No one else has reported anywhere near the increase in sub and cc views as OP has, percentage-wise (I'm speaking about those who already had subs enabled). I would like to know the geographic statistics of OP and some of the others to see if that also plays a part. Also type of video and whether the commentary is very important to the viewing experience.
 
Just revisiting this thread, it's been a while since I started it. The video that was promoted has passed 100K views where the one that wasn't is at around 15,000... I still won't be promoting it to see how things go.

I know there are many variables that can be compared as expressed earlier but these 2 videos were very similar. Many other videos I promote pass the big 100 however, I feel the audience retention from some sites it's promoted is pretty poor so may start being a bit more selective.

Will update again in a few months or if the non promoted vid decides to do something crazy!

SegdSHw.png
 
Yeah I've done this... my channel I used to promote alot more and I get alot fewer views when I don't bother to promote it...

But for the last few months I have been doing almost nothing in the way of promoting...
 
I suppose it could do! My new strategy which I need to test is searching common keywords that are used on google via google adwords and making a video about it to test whether that one video ranks higher in the views and equal more views!
 
Back
Top