Heya friendos! No big lead-up this week, we’re just looking at a video I dropped on Thursday. We’ll dive into some analytics and how I’m going to try to trick the algorithm for the next video. Let’s go!

How Am I Doing?

I’m good! I still have pretty erratic energy spikes, I think because of the weather. I’m cutting down on coffee and switching to herbal teas instead. I also bought a climbing gym subscription, so I’ll aim to go two times a week instead of once. Hopefully, that can mitigate the winter depression a little bit.

What Am I Doing?

So I published another flagship video last week. It’s about procedural narrative and all the problems you can run into while using it. Give it a watch if you haven’t already! I’ve embedded the video into this post, but I don’t remember if my email service supports that, so here’s a link to the video, just in case.

Now, it hasn’t seen the same level of views as my other flagship video. I was a bit bummed about that at first, but now that I have some statistics, the programmer in me wants to start debugging the process. Let’s start that now!

Unraveling the Algorithm

I had expected this one to do just as well as my earlier video. Why? I have approximately 8 times the amount of subscribers than I had back then. I figured if all subscribed folks give it a watch, it’ll get boosted at least a little bit. Turns out, only 12% of subscribers gave it a view so far. That’s at least useful data that I hadn’t found anywhere before!

Then the other reason is, that the subject of procedural narrative is quite popular among game development students, and very much under-represented in educational materials. I think this will affect the popularity of this video in the long run, but apparently has no influence over being highlighted in the algorithm. Most views for my videos come from the Home page, not by being searched for a subject. Is SEO still important then?

So, what else differentiates the two videos:

• I published this one on Thursday in the afternoon, and the other one on Friday in the morning.
• It’s been a while since I’ve published anything, so I’m assuming activity has to do with it.
• I published a bunch of shorts leading up to my first flagship video. Even though they didn’t seem to get that much traction, the activity might lead to an algorithmic boost.

Future Tests

We’ve also been fiddling with A/B testing thumbnails but to no conclusive results yet. We’ll have to get an influx of new people to the channel to actually measure anything regarding that. Here are three examples of the first flagship video’s thumbnails:

I’m particularly interested in whether my face or a popular game attracts more viewers.

Now, I have another video in the works that I hope to publish this coming Friday. With that one, we can test if activity has anything to do with it, since I published a video last week. Publishing date and time as well. But I won’t be publishing any shorts, so we can measure if those other things have serious influence or if it’s really about being prolific throughout YouTube’s whole platform.

Why Am I Doing? (this)

I’m trying to find joy in the process of these videos. As of yet, I love writing the scripts but I’m not a huge fan of filming and editing them. With previous videos, the dopamine hits I got from comments telling me how helpful it is got me across the finish line for making another one. But that’s not a stable motivator: some videos might not do as well and people are not always so friendly.

I think this “debugging” process, of checking analytics and comparing videos’ subject matter and presentation reminds me of game development. That’s a process I enjoy well enough. I’m not sure if trying to trick my brain into enjoying something is the healthy way, but we’ll see where it leads me.

Keep a look out for this week’s video!