Two years of data on which games get viewed, and which don't

Hello all. I have been on the TigSource forums for a few years now. I spend most of my time in the Playtesting forum. I record my playtests and my voice and I upload them to youtube.

I'm not into youtube for the views or the money. It's just a convenient place to store videos. But over the years I have some data for what videos get views and what videos don't get views.

Youtube is similar to Steam or Epic in a lot of ways. There is a huge amount of content and users usually only take a few seconds to decide to engage with something. I wanted to share what I have discovered with you here today.

A quick side note: if you see your game posted here as a negative example, please don't take offense. Your game might be great and I just used a bad screenshot. Also, I'm only using views as a quality of success. Not the quality of the game.

You need to have good screenshots

Here are some videos that have done very well:

Notice how good these screenshots look. Having good art is a big part of it, but doesn't seem to be super essential. It seems to be more important that the game is read-able. Can your potential buyers figure out what your game is all about based on the screenshots?

Here are some videos that did poorly:

Notice these screenshots don't look great. It's hard to read how these games are played. Some of the shots are of the "Game Over" screen or of another menu. How much time is the player in a menu rather than playing the game?

You need to have a good title

Here are some titles that did very well:

   Ants Cutter
   Devoul, Curse of the Soulless. A gothic themed metroidvania
   Hyper Ultra Astronautics
   Lunar Cross. A retro JRPG
   Tiny Toy Tanks, a tactical tank combat game

Notice how you can get the gist of the game based on the title. Later on in my playtesting I starting adding a one sentence descriptor. Notice that interesting descriptors can improve views.

Here are some titles that did poorly:

   Armortale
   Doomies. A fast paced shooter
   Fate
   Fruitball
   Last Joy
   Other Cell
   Space Boi
   Turbo Island Purger
   Tempest Might, first person elemental combat

Sometimes strange titles will do well:

   Fruit Man Adventures
   I Don't Know Which Name I Should Give This Game, a wacky platformer
   Super Punchy Face. A brawling adventure game

But this doesn't always work. The game "Christmas Adventure IV- The Eternal Fruitcake of Light" didn't get many views.

Sometimes you get lucky

Sometimes the algorithm will blow up a video:

   Hands of the Killer

Junon.io

These games are good, but the view count was abnormally higher. My videos average around 40 views. These got more than 1,000.

I have no idea what happened here

These videos, in my opinion, have great titles and great screenshots. But the views are low.

Airborne Motocross, a motocross racer - 16 views

GunsBlazing, an old school first person shooter - 8 views

Hijack Overdrive, a side scrolling shmup where you steal vehicles - 9 views

MechCube - 18 views

Shark Bait, an arcade balancing game - 5 views

Soccer Journey, a tricky kicking simulator - 5 views

It's a shame too. Some of these games were quite fun.

These videos, in my opinion, had poor titles or screenshots. But views were high.

Born of Fire - 52 views

Fatal Proxy - 141 views

SafeGuard - 159 views

If you have any idea about these please post in the comments. I would love to know what happened to these.

If all else fails..... sex sells

This is hands down my highest viewed video, it's also the only game I've played with sexual content. It's no surprise why it did well.

I hope you found this helpful. Please let me know if you have any insights from this.