I was over on Reddit today answering questions and ended up writing an entire article.
Here's the question that I set out to answer... (my 2,059 word answer follows)
>>>But here's what happened: I took the advice of some other YouTubers back in the day to make videos that would bring in traffic. So I made some tips and reviews videos periodically, and one of them has over 100k views. But then people subscribed ONLY expecting me to make more of those videos and not actually giving a flying f**k about anything else on my channel.
Yep. Been there.
I have a video that gets 700+ views a day, is fast on it's way to 100k views, brings in 20 or so new subs a day...
...but that video is a unique one of a kind video, unlike anything else on my channel, so people who sub from it never watch any of my other videos.
And you know what the stupid thing was? It was just a test video I made to test out a software program I had. It wasn't even a proper video, is shit quality, has terrible editing... and it's top rated video?!?!?! Why? I don't know.
I'll be at 3k subs in a week or 2 and most of my videos get only 30 to 50 views each. That's only a 10% sub to view ratio. which I thought was bad, until I realized Jacksepticeye will cross 15 million subs in a few weeks and he struggles to get 200k views per video... ihm... I guess that means I'm doing good if I'm getting a 10% view to sub ratio then?
I don't know.
>>I wish I could just purge the ones that never watch anything but unfortunately I can't.
Yep. I actually contacted YouTube and suggested they give us the option to block subs that don't view any of our videos after 6 months. Never heard back from them of course, but, at least I put the thought out there to them. Maybe if enough of us request it they'll make a feature like that?
>>I could potentially delete the big view videos that are still bringing in dead subs even though those people should see I'm not making those videos anymore...but then again maybe views on those videos are still better than no views at all?
Yeah, I had considered that too, but then I realized my analytics says that one video is bringing in 54% of all my AdSense income...even more income then my websites are bringing in from AdSense.
So I left it up. It's getting views, it's gaining subs, it's making money... I just wish it's spread some of those things to the rest of my channel, you know?
I have found that changing tags helps, but I have 2,000+ videos so kind of not practical for me to go back and change all the old tags.
There is a way to turn off the subscription counter, so that no one sees your sub count and only sees you view count. Can't remember how to do it, but I know it's in the dashboard somewhere. That might help if you are worried people will think you sub4sub.
Actually AdSense does NOT pay based on views. It pays based on interaction with ads.
Also, AdSense pays based on the amount an advertiser bid for the keyword of your video. AdSense pays you .61c for every $1 of advertiser bid.
Meaning that if you use any of the following tags and got clicks today...
(today's actual bid prices for those keywords, according to AdWords Keyword Planner - log into your AdWords account using your AdSense account to watch the current bid prices - like the stock market they change every hour)
This is also why it is bad practice to use your username as a tag, because no one is bidding on your username.
When you see HOW MANY ad clicks it takes to earn $100, that's when it hits you hard, how few of your subs and viewers clicked on an ad and how many views it takes to actually earn a livable income on youtube.
Because of this, it is possible to get 100k views on a video and still earn only .50c if only 1 of those 100,000 people clicked to visit the advertiser's website. And at the same time it is possible to get only 10 views yet earn $100 if all 10 of them clicked ads. Which is why it is so difficult to guess how much any one video earns, because the bid prices change hourly, so the tag that earns you .10c a click today, could earn you $10 a click tomorrow depending on trend in advertisers bidding against each other for top placement of that keyword.
This is why you choose your tags wisely, because you are paid according to which tag brought your viewer to you. If you use higher bid tags, you get paid more per click.
According to Google' Help Center Blog,you can expect to receive between .20c to $4 per 1,000 page views, with most people earning $1 per 1,000. They also state that for every 1,000 page views, you can assume only 10 click the ads.
Interestingly that same article by Google also states to not expect to earn $100 per month off AdSense until you have at minimum 100,000 views per month.
(Also- I build websites and have been living off AdSense income since 2007, which is how I know this stuff.)
If you want to earn a living off YouTube you can do so even with very low amount of views. The secret is the tag your videos with the keywords that pay the highest price per click. Tags are NOT for drawing traffic to your video, tags are for AdSense to determine how much to pay you for each click. Use high performing tags (tags with minimum bids of $5) and you'll see faster income with fewer views.
when you consider the fact that only .04% (less than a half of 1%) of Pewdiepie, Jacksepticeye, or Markiplier's subs watch their videos, I'd say youtube doesn't care about sub to view ratios.
Each of them has multiple million subs (54m, 14m, and 18m) and each of them struggles to get even close to a million views per video (500k, 200k, 300k). That alone speaks volumes.
Jack will be crossing 15 million subs soon, and he can't even reach 300k views per video? What good are 14 million subs who never watch your videos? Think about it.
>>>EDIT: Had a quick look at your channel, and I'd guess some of it is game choice. Even with the new updates, my sense is NMS and Slime Rancher have both been around for long enough that, unless you've already got a large following with fans devoted for them, you'll have a hard time keeping viewers engaged. (I've put myself in the same boat with XCOM 2, to be fair, but dammit I just really want to play that game! So the low views are the price I'm wiling to pay.)
You do realize Slime Rancher's release date was only 2 weeks ago, right? Have we become so new game centric that a game released only 14 days ago is considered old now?
All those videos from last year were the people who got invited to play it in pre-alpha...the alpha was dramatically different from the beta, the beta even more different from the actual release. So far only 1 single, solitary YouTuber has actually played the release of Slime Rancher...JackSepticEye. Of course, it's only been out for 14 days so most people haven't even had time to buy it yet.
Like you everyone was saying "That's already been played"... you might want to look into the actual release edition of it, it's vastly different from anything you saw in last year's beta.
You know, there was a thread yesterday where I asked why someone thought a 2002 game was old... but this is talking about a 14 day old game? What next? Is 14 seconds after the release going to be an old game soon?
How in the heck is a 14 day old game to old to play?
Often a decrease in views is simply caused by a trend in search results.
For example, 5 years ago, when Twilight was big, vampire tags were getting millions of searches per day. Now, the vampire trend is over and vampire tags only get searched for 10 or 20 times per day. (This is according to SBI's Brainstomer-It keyword trend tracker program.) So 5 years ago, all you had to do was say vampire and you'd be raking in cash. Now vampires will not bring in traffic or money.
Search trends fluctuate daily, monthly, seasonally, yearly, depending on new movies, current events, the latest Trump tweet, the next holiday, whatever. This in turn affects us as YouTubers.
The only way to keep on top of search trends, is to be constantly changing tags in videos. So, while a video is getting views, don't touch it. But once views tank, change up it's tags to try to find on-topic tags that are still trending in searches.
Log into AdWords (use your AdSense log in) and then, in the top click "tools" and then "keyword planner"
Than take the titles of your games, and type them into the search box. Google will give you all the current advertiser bid prices for the keywords of your game.
Sort the results by highest bid, make a note of the top 10 (relevant) keywords. Then sort by the most searches (how many times that keyword was searched for in the last 28 days) and make a note of the top 10 keywords. (Note that off topic/non relevant keywords get in the mix - ignore those)
Look at both lists. Are there any keywords on both? If so, use those keywords as tags in all of your videos for that game.
Do this for each of your games, each of your videos.
Changing tags like this will both increase your chances in search results and increase your pay per click.
Older games can be difficult to rank high in, simply because they don't trend very high in search results. What you have to do, is find out which keywords ARE trending for that game and use them.
For example: Today's top 10 ad bids for Stardew Valley
Those are all VERY high paying keywords, as the average keyword pays $1 or less. Finding a keyword above $5 in incredibly difficult, and yet, Stardew Valley is ranking very high with several keywords over $10 today, meaning the game is trending VERY HIGH in search results right now, thus advertisers are bidding very high to get ad placement on those keywords.
but the game title alone is...
And the worst keywords to use right now are these...
So, you can see how big of an impact, just changing up your tags can be. The game itself is still trending in searches, but certain keyword phrases are trending while others are tanking and the trick is to figure out which ones are bringing you down, getting rid of them and replacing them with ones that will boost you up again.
Hope that helps, and good luck with your channel!