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

[FartingOnThings]

Interesting results here. I'm wondering, when it comes to retention, does the % viewed or watch time matter more? My videos are all really short 30 seconds max, so as a result, my average retention is 80%. On some videos the retention is over 100%. But, since the videos are so short, it takes me like 5 views to get 1 minute of watch time. Am I putting myself at a disadvantage by having such short videos, or do I have the upper hand since my videos can get high retention quite easily?

I noticed that promoting my videos has no effect on my retention at all.

 

[just Moto]

Interesting results here. I'm wondering, when it comes to retention, does the % viewed or watch time matter more? My videos are all really short 30 seconds max, so as a result, my average retention is 80%. On some videos the retention is over 100%. But, since the videos are so short, it takes me like 5 views to get 1 minute of watch time. Am I putting myself at a disadvantage by having such short videos, or do I have the upper hand since my videos can get high retention quite easily?

I noticed that promoting my videos has no effect on my retention at all.

Those videos with the higher retention than 100% are from people skipping back and watching twice without the video ending.

I was under the impression that YouTube favoured videos that gained more volume of watch time rather than the actual completion of watch time. There's only so many videos they can 'reward', therefore a half hour video with a retention of 33% will get better placement than a 30 second clip with 100% retention. Yes I have videos with these statistics and the longer vids with shorter retention do the best.

 

[FartingOnThings]

Those videos with the higher retention than 100% are from people skipping back and watching twice without the video ending.

I was under the impression that YouTube favoured videos that gained more volume of watch time rather than the actual completion of watch time. There's only so many videos they can 'reward', therefore a half hour video with a retention of 33% will get better placement than a 30 second clip with 100% retention. Yes I have videos with these statistics and the longer vids with shorter retention do the best.

Yea, I figured that's how it got over 100%.

Hm, so I guess watch time is more important.

 

[just Moto]

Yea, I figured that's how it got over 100%.

Hm, so I guess watch time is more important.

The longer someone is on youtube, the more likely they are to click on an add

 

[wasabiroots]

Those videos with the higher retention than 100% are from people skipping back and watching twice without the video ending.

Like these~ lol





My subscriber(I think it came from 1 subscriber) is very enthusiastic, might have used my videos to sleep or something. lol

 

[WilliamRayWalters]

therefore a half hour video with a retention of 33% will get better placement than a 30 second clip with 100% retention.

On the flip side, if overall session length is the primary factor (as Youtube tells us it is), then a video that is 5 minutes long with 50% retention and that consistently leads to an overall session length average of 15 minutes would theoretically be ranked higher by the algorithm than a video 10 minutes long that has an 80% retention rate but leads to an average session length of 13 minutes.

The longer someone is on youtube, the more likely they are to click on an add

That makes sense, of course, but you would think that shorter videos would be favored by the algorithm, the thinking being that the more videos a person watches, the more revenue they will generate by watching/clicking on ads. You would also think that shorter videos would be more likely to keep the viewer in that 'youtube' mindset of skipping from video to video. I mean, after I watch a long video on Youtube I'm ready to take a nap lol. watching 10 2-minute videos, I believe, would yield more ad revenue than watching 2 10-minute videos.

Someone will probably bring up something that I've failed to consider in that argument, though. I feel like I might not be taking something into account.

 

[FartingOnThings]

On the flip side, if overall session length is the primary factor (as Youtube tells us it is), then a video that is 5 minutes long with 50% retention and that consistently leads to an overall session length average of 15 minutes would theoretically be ranked higher by the algorithm than a video 10 minutes long that has an 80% retention rate but leads to an average session length of 13 minutes.

That makes sense, of course, but you would think that shorter videos would be favored by the algorithm, the thinking being that the more videos a person watches, the more revenue they will generate by watching/clicking on ads. You would also think that shorter videos would be more likely to keep the viewer in that 'youtube' mindset of skipping from video to video. I mean, after I watch a long video on Youtube I'm ready to take a nap lol. watching 10 2-minute videos, I believe, would yield more ad revenue than watching 2 10-minute videos.

Someone will probably bring up something that I've failed to consider in that argument, though. I feel like I might not be taking something into account.

See that's what I had in mind as well. I would think chains consisting of short videos would be far better than chains of equal watch time consisting of long videos. More video views = more ad views.

 

[WilliamRayWalters]

See that's what I had in mind as well. I would think chains consisting of short videos would be far better than chains of equal watch time consisting of long videos. More video views = more ad views.

I agree, but the thing is, Youtube gets roughly 6 billion hours of watch time per month so they have all the data they need to come up with the right answer. If they're focusing on overall session length rather than how many video views in a session a video leads to, then there must be a reason for it. That being said, it's likely that there is a sweet spot when it comes to video length that the algorithm favors that isn't too long and isn't too short. The algorithm isn't just one thing, it's a ton of things. And according to a couple guys I've spoken to in the past that would know, they tinker with it a lot more often than you'd think, so it's hard to really get a handle on what exactly what it favors as far as video length is concerned. I still like to keep my videos relatively short though (under 6 minutes or so if I can help it and depending on the channel/niche).

 

[just Moto]

my top 5 vids (i think I have around 400) are 7:00, 23:40, 13:08, 17:26, and 11:25.

In that case then, I guess it doesnt matter if they only have low % retention because of the total minutes of watch time they generate from being a long video.

 

[WilliamRayWalters]

In that case then, I guess it doesnt matter if they only have low % retention because of the total minutes of watch time they generate from being a long video.

Or that they lead your viewers to have an overall longer viewing session. How low is the retention on those videos anyway? And how are you attempting to steer them to other videos if they decide to bail 1/4 way through? Annotations? Cards? Last year I started using my retention numbers to help me place my annotations/cards and it has helped increase the views I get from those calls to action a considerable amount.