Posts
As I've shown in earlier blog posts, I have a low subscriber count but a very high video view count.
I'm actually unsure what the problem is. Currently the jfreedan channel directs people to subscribe to my TheRPGFanatic channel, so one might think the subscriptions would go there but that's not been the case. People still subscribe to the jfreedan channel despite flat out telling people no new videos will be uploaded to this channel.
So the one thing I cannot tell you is how to get lots of subscribers on Youtube.
What I can tell you is how to get more views on Youtube, and why placing all your hopes and dreams on the Related Videos feeds will probably end in disappointment.
I feel I need to burst people's bubble concerning related videos because there are many Partners whose entire Youtube strategy has been focused on getting their videos into Related Video feeds, to the exclusion of all other marketing efforts aside from AdWords (another fruitless effort if you don't have a big pile of money, as I'll talk about in a future blog).
However, to fully grasp the complexity of the situation I'm going to talk about a few of my videos.
#1: Evil Ronald. This was the first video I made that went viral. I don't have it on the channel anymore, it got flagged by trolls with false privacy violation reports. Someone claimed their likeness had been used without permission after the video had been up for three years and I had no way to prove I had permission from all parties filmed in the video (fortunately when they tried that on my Cosplayers documentary I was not alarmed because when making that project I had everyone sign release forms, which I have also done for every other video involving other people in it).
When I say it was easy to get views back then, I'm not kidding. Most of these videos no longer exist; when I became a Partner I removed the majority of my gameplay videos because they had no commentary and keeping them on the channel would violate the agreements.
I will also admit the video was attracting negative attention and this contributed to my removal of it. The content of the video involved me going into a McDonalds with an evil Ronald McDonald costume, and accidentally scaring the crap out of a six year old girl, whose father then proceeded to try to fight me inside the restaurant. A lot of people thought it was funny but some people tried to use my prank gone awry as proof I was a horrible person. I received death threats on a weekly basis, and it gave ammo to trolls who sought to get my Partnership revoked.
Some people were also unable to take my videogame journalism work seriously after seeing me accidentally scare a child while wearing a crappy Halloween costume. As popular as it was, the video wasn't helping my current goals.
The video had half a million views when I deleted it. I did reupload it to my channel but it has remained private.
Anyway, the video did not result in a lot of subscribers. This is despite making more comedy-type videos, some of which were successful and some that were ignored. I made a sequel to the Evil Ronald video titled 'Evil Ronald McDonald Birthday Party' that received 16,700 views by the time I made it private (again, to protect my channel against the trolls). I have considered making this one public but the general consensus is that people don't find it as funny as the first one.
So why am I talking about it? The reason is because, if I remember correctly, this video went viral because it was pimped on the Something Awful forums a lot. From there it got shared on Facebook, Myspace and people emailing it around. I have none of the data from this video because I deleted it (I can see how many views it had, but not where the views came from) but the video went viral because people shared it, not because "Evil Ronald McDonald" was a trending topic in 2007.
#2: Ultraman review. During the course of the RPG Fanatic Show I decided to try my hand at a review of tokusatsu shows. The only video like this which I created was a review for the first season of Ultraman.
It actually did surprisingly well and I think the reason is because I submitted it as a video response to another Ultraman related video that had a lot of traffic. That act raised the awareness of the video considerably.
I've included this screenshot of the page so you can fully understand what I'm trying to convey here; just because you get a video to have lots of traffic doesn't automatically mean they will watch your other videos or subscribe to your channel.
The Ultraman review has 93,711 views as of this writing. Three of the other videos you see to the right of it are also my own videos and only one of them has anywhere near the same number of views (and it's not because of this video. I'll explain why in #3).
(Note: From this point on you might see some screenshots that have extra browser bars on the top of them. This is because I made these screenshots using the Print Screen command and pasting into MS Paint, which apparently likes to drop down a few spaces when doing so. I made this blog while sleepy so I didn't notice it until now, and really don't care that much about this to go back and make new screenshots. The information I talk about can be clearly seen. That's all that matters.)
You might think, "Well it has a lot of views but that doesn't mean people are watching more than a few seconds of it. That is why they don't watch any of your other videos."
If you thought that, you would be mistaken.
Compared to all other videos of similar duration on Youtube, it has about an average retention rate. For a five minute long video, that's not too bad.
For whatever reason, my video that people seem to like a lot and share has poor conversion rates. Just because someone enjoyed one of your videos doesn't automatically mean they want to watch your other videos.
#3 Unlosing Ranger Vs Darkdeath Evilman Review. This video got a lot of views because I released it several days before it was actually available for sale in North America-- I received an advance copy of the game for review from Original-Gamer.com, whom I often contribute reviews to. Because this was one of the first Youtube videos about this game it gained a lot of momentum soon after uploading.
As of this writing the video has 52,112 views. It gained the majority of its views within the first month after I released it (somewhere around 35k I believe) and now it tends to get about 1,200 views every month.
 |
| Note: Earnings data has been removed since publicly displaying that information would violate my AdSense agreements. |
Now if we look at the entire lifespan of the video (10/17/2010 to 5/5/2012) we can see where all the traffic has been coming from.
A large portion of the views are coming from searches inside of Youtube, but almost as many are coming from embedded players -- that is, it is receiving a ton of traffic from external websites like Facebook and member only internet forums which Google can't data-mine. Only around 5,000 views have actually came from the "suggested videos" feed, which is also known as the Related videos feed.
These are the videos where my Ultraman review appeared in the Related Videos feed and people actually decided to click on it.
As you can see almost all of the Related Videos traffic came from my own videos. The videos which don't belong to me are called "Official Z.H.P. Unlosing Ranger Vs Darkdeath Evilma.." , "Blazing Souls Accelate Gameplay 2", and "[JAP] ZHP -All Dengeki Characters- List in Description, READ IT". All of the other videos can be found on my own Youtube channel.
Basically, I received barely any traffic from anyone else's videos. My video is basically isolated from the rest of the Unlosing Ranger related videos on Youtube, and if people don't find it in the search results or see it embedded on another website they basically cannot find it.
This is also worth pointing out: a whopping 146 views came from my own subscription feed. That's right, a little over one hundred of the two thousand subscribers I had back then actually saw my video from their subscription feed. This is not unusual and matters have been made worse given Youtube's recent changes to the subscription feeds that, by default, hide a large number of videos and only show users what Youtube thinks will be "relevant" to them. Youtube calls this content "highlights" but it's really just a fancy name for, "we're hiding videos that don't get a lot of views in a short amount of time".
#4: My Human Torch Special Effect Test. This is one of my most popular videos right now. It is currently at 97,179 views and I guess will eventually hit 100k. I say eventually because since the recent changes to Youtube's algorithms for search and Related Videos feeds, it has lost a dramatic amount of traffic.
As with the Ultraman and Unlosing Ranger videos, you'll notice my other videos appear in the Related Videos feeds but don't have a lot of views either.
( You will notice this video has no annotations. This has been my test video to see if including annotations on videos would increase traffic or not; this video was used as the control, but I ultimately didn't even need it since the annotated videos performed so poorly. If you're curious what I think about Youtube video annotations you can read a prior post of mine. )
Anyway, let's look at some of the Analytics for this video,
In the last 30 days the video received 25 views. I don't even have to bother removing how much money it earned, because it didn't earn jack.
Let's look at how well the video was doing last year during 2011, before the re-design of Youtube's layouts and removal of several subcategories for videos, as well as the overhaul of Youtube search and related videos algorithms,
The video received 10,432 views. That is approximately 869 views per month. Not spectacular but certainly a lot better than 25.
That said, there is a chance the dramatic drop in views is not entirely due to the algorithm changes but because of other factors.
This is the number of views for my Human Torch video during 2010.
25,312 views, for an average of 2,109 views per month.
This is the number of views for the same video during 2009. I didn't have a Youtube Partnership back then so I wasn't able to capitalize on any of these views, and I did absolutely nothing to promote the video either.
21,078 views, for an average of 1,756 views per month.
Let's look at 2008.
32,700 views were earned during this time period. That is roughly 2,725 views per month.
Finally let us look at the very first year the video was uploaded, which was on May 23, 2007.
4,174 views for that first year. That's quite a difference from what it received the next few years, isn't it? So what happened. Let's try to figure this out using Analytics.
2007-2008: I have no Analytic traffic sources data for these years because back then, apparently, Youtube wasn't tracking where video views came from. I guess nobody thought that was important? Anyway that is highly unfortunate because I would liked to have known what factors led to my little known video starting to skyrocket in views.
2009: As you can see in the below image, almost all of the views to this video came from related videos feeds on other user's videos.
I have no idea what "other features" is supposed to be. It just says "Other" and "Homepage" when I click on it, as well as the front page of other user's channels. I guess Youtube was tracking all those locations together back then. Anyway I had 729 views from the Youtube homepage during 2009, so I guess I can say at least one of my videos made it to the front of Youtube?
These are the videos that led traffic to my own video,
Only one of these videos belonged to me, the speed punch special effect test. All other videos belonged to other users.
This is the traffic sources from 2010,
Once again the related videos feeds are the primary source of traffic. However the video was also marked as a 'featured video', receiving 7,133 views from that service.
A 'Featured video' is either one of two things; a video someone has paid to promote using Adwords, or a video that has been enabled for monetization so Youtube is promoting it alongside the paid ads so that Youtube makes more money from the video.
If I remember correctly Youtube invited me to take part in the "Individual Video Monetization" program, or whatever the hell it is / was called. This was a program that allowed people to run ads on videos that were getting lots of traffic but without Youtube allowing the person to be a full Partner and monetize all their videos. I believe Youtube used the program to find new potential Partners, since I eventually became a full Partner due to how well my individual monetized videos were doing at the time.
Anyway you will notice I got very little traffic from Youtube search, Google search or external websites. All of my traffic came from Related videos and having my video featured in search results.
Onward to 2011,
Almost identical results to the year before. Related Video and Featured Video feeds are doing all of the heavy lifting for me. Nobody is sharing the video and I'm most certainly not doing anything to promote it.
And now we come to the present year, 2012
As you can clearly see after Youtube made their overhaul of the algorithms for Related videos the traffic was pretty much killed off. My theory is that other larger viewcount videos were also being benefited by the Related Video feeds and I was piggybacking on their Related Video feeds. As a guess, based on the traffic I received I would say my video was about two or three degree separation from hugely popular 1 million+ videos.
Now that Youtube changed their algorithms to heavily favor videos that are getting lots of search and external traffic, my video is dead in the water.
I can't say I'm entirely surprised. It's not that amazing of a video.
Now there are two other factors we should be considering but I am unable to do so. I lack the data.
The first factor is how many other videos similar to my 'Human Torch' were uploaded and used similar titles, description and keywords. Those videos would be competitor videos in the search rankings and Related Videos feeds, and the more competitors there are the more difficult it is for people to find my video.
The other factor is how often did users search Youtube for keywords that would bring up the videos that ultimately led users to my video? There is no tool I can use to obtain that data. Only someone at Google would know.
I won't further bore you with detailed analysis of all my videos. But if I did, you would clearly see that, for videos made after 2010, the videos that got a lot of traffic did so because they were virally shared on services like Facebook, Twitter and embedded onto forums. The Related Videos algorithms occasionally snatched them up and started pimping them on other high traffic videos, but that isn't where most of the traffic came from. I know this because I have the Analytics data to back it up.
As for my videos from the early days of Youtube, they got snatched up by the Related Videos algorithms and piggybacked off the popularity of other videos. This is no longer as easy to do because videos now need to have their own traffic from external websites (like Facebook, Twitter, blogs and forums) to benefit from those services.
So in conclusion there are a couple things we can learn from this:
#1 Just because a person watches a video doesn't necessarily mean they subscribe to a Youtube channel. Thinking about my own Youtube behavior, that sounds about right; I tend to only subscribe to a channel which produces content similar to my own show because I want to stay aware of what others in my field are doing, and for several years I didn't really subscribe to anyone because I wasn't watching videos while logged in.
#2 Viral Videos Don't Always = Lots of Subscribers. My Evil Ronald video had half a million views but I don't have even a fraction of that many subscribers. I'm not the only channel that is like this. Consider "I am your Grandma"; the video has a million views but the channel itself has only 2,649 subscribers.
Even I'm not subscribed to her, or at least I wasn't until I realized I should be.
#3: The Related Video feeds should not be your priority. I know of at least five big-time Youtube Partners who, from the periods between 2008 and 2011, got hundreds of thousands of views to their videos because of the Related Video feed algorithms. As soon as Youtube made changes to these algorithms, those Partners claimed to have lost anywhere from 60% to 80% of their daily view traffic.
This is despite spending hours upon hours using the Google trends and AdWords keyword tools to find new keywords, as well as studying the title, description and keyword tags used in other videos to try to "match" those videos. (It doesn't appear to be that simple anymore. Even if I copy and paste keywords from The Angry Video Game Nerd videos into my own videos, my videos aren't going to appear in his Related Video feeds, at least that I can see. I've tried it out before.)
When a video is heavily promoted by the Related Video feeds on high traffic videos it is largely due to sheer dumb luck. It is not something you can control and it has little to do with the quality of the video. High traffic videos are more likely to display other videos from the person who made the video than they are to show someone else's videos. This has been my experience with videogame related videos and I imagine it is the same for other kinds of content, too.
The lesson to learn here is to promote the hell out of your videos OUTSIDE of Youtube!
It is vitally important that you do this. The videos I post onto reddit and my Facebook Fan page may only receive a few hundred views from these third party sources, but those videos get a lot more traffic from Related Videos feeds than the videos I don't do anything with. That said, trying to get your video into the Related Video feeds of very popular videos is like shooting at a dime in the dark; there is no guaranteed way to get traffic from them.
If you don't have a legion of fans and simply upload videos to Youtube then you are uploading into the aethers. Nobody will ever see your videos, I promise you that. Anyone who tells you differently is lying to you.
Where you might promote:
I primarily make videogame related content so my knowledge of where to submit videos is largely related to that field. For example, if you make Let's Play videos you should submit your content to the Let's Play Archive (which I mentioned on a previous blog about Project Wonderful) .
Some of the usual suspects are places like Facebook, Twitter, reddit, digg and so forth. The only drawback with these is that, just like on Youtube, it can be hard for people to find your content because there is so much content being shared all the time.
You can also try making your own "video sharing" website focused on the type of content you make; this is my latest and greatest plan for getting my content seen. We re-designed my roleplaying videogame fansite to be a Web 2.0 "submit your own content" style database for RPG related videos and blogs. I came up with the idea to do this primarily because (aside from the Let's Play Archive and ScrewAttack) the majority of gaming websites don't like it when people submit videos about videogames. You tend to get banned from forums for doing so.
Anyway, viral youtube videos can be "one hit wonders" and while they can raise awareness of your work it is an entirely different matter to translate that awareness into loyal subscribers-- and I say loyal because just because people subscribe to your channel doesn't mean they will watch all your videos.
I have absolutely no advice to give you on how to translate viewers into subscribers, so if you were reading this hoping for some magic solution then I apologize. I promise I'll tell you when I find one.
I hope you've still learned something valuable from this article.
Follow Carey on Google+
No comments:
Post a Comment