Ask HN: Advice on starting a YouTube channel?

My experiences with YouTube and other platforms have been mixed.

I recently setup a personal channel for the occasional development video, for when the content doesn't quite fit into just a written blog post. On such a new channel with no prior uploads and without posting the videos anywhere else, I got close to 500 views on a video about the Godot game engine in less than a month. Yet, another video about Ansible had under 50 views in 3 months.

Meanwhile, a separate gaming YouTube channel where I upload Twitch VODs and such has over 100 videos, yet the most viewed video has under 200 views and there are like 20 subscribers on that channel, whereas on Twitch there are 50 followers.

I don't think it's quite possible to figure out the YouTube algorithm, but in general, I get the feeling (through my lack of success), that:

  - Picking good video topics (popular subjects) probably helps.
  - Picking a good video format (edited videos up to 30 minutes in length, instead of something like unedited stream VODs/super long videos) probably helps.
  - Good presentation (good visual quality and content, some effects/transitions and just competent video editing) probably helps.
  - Good audio (decent mic, audio not peaking but being balanced, some EQ, limited background noise, well articulated speech) probably helps.
  - Getting people who are actually interested in your content watching it and interacting with it (likes, comments, watch time) probably helps.
  - You probably need all of that, as well as a good amount of luck and the right kind of personality/manner of presenting to succeed.

In regards to that last point, I absolutely do not think that I have the right personality for this (for example, too passive/monotone voice), yet I don't mind using YouTube for those blog videos or stream VOD backups for the sake of it. Others probably experience 10x-100x the growth that my channels might. For example, a friend who streams on Twitch has gotten 50 followers in about 3 weeks on a new account, yet it took me over a year to get there.

I recently figured I'd try out YouTube Shorts on the gaming channel - I got about 400 views in a single day after bringing over some Twitch clips, but then views for those more or less fell off a cliff. Seems like they were presented to a wider audience at first to see if they're interested and when there wasn't too much interest, the content wasn't offered up that much.

The same applies to other types of videos - if you make development videos and nobody is interested in the topic or your video itself, then growth is unlikely. Thus, try to go for quality over quantity, yet upload somewhat regularly, if possible. Probably share the videos in any relevant communities as well.

I wouldn't call my experiences positive, but I guess that helps avoid survivorship bias ("Hey, I failed at the thing, here's some of the stuff I did and it didn't work."). At the end of the day, try to have fun doing it, listen to what the more successful people out there have to say, yet be prepared to put in more work than initially anticipated, sometimes with way less than optimal results. Good luck!

yeah, see, advice like "pay for translations" is unrelatable for anyone under 10m subs, which is where everyone here is.

Run your audio through Adobe podcast. It's insanly good.

All major creators are uploading YT shorts also now.

How’s the mechanics for it? Have you considered starting from there?

I started making YouTube videos for my classes back in 2015, currently have around 3.3k subs and just crossed 1 million total views. Link in my profile and MAKE SURE TO SMASH- I'm not doing that...

Others have said it, but audio is 100% critical. Some of my earlier videos were using a dinky Samson Meteorite omnidirectional mic, and the results were subpar imo. I finally picked up a Yeti over COVID remote learning and I am very happy with the results. Side tip, ALWAYS CHECK THAT YOUR MICROPHONE IS ON! It really sucks to knock out a 10 minute video and find out there's no sound...

Play around with your "style" of video - floating head, voice over, literally standing in front of a giant white board, or go wild and use the 3brown1blue animation style. I personally like the floating head with Khan Academy style drawing.

Things that matter if you choose to do the "floating head" style - lighting and deciding if you want to use a green screen.

- Lighting is hit and miss, mostly because the room I currently record in has massive windows so light is always pouring in. I do use a standing lamp with 3 bulbs to help balance out my shadows.

- Green screens are a pain in the ass in my opinion. One of my most viewed videos (on the Luhn Check algorithm, thanks CS0) has me fading in and out. I ended up donating it to the school's visual arts lab when I moved.

If you're going with the Khan Academy / Slides approach, get familiar with OBS and all the little annoyances like making sure menus appear in the recordings (I think its some Windows 7 compatiblity thing?). If you have your head on the screen, try to frame things so your head isn't hiding the material (some of mine have that but always minor stuff).

In regards to tempo, I sort of disagree with the comments on speaking slowly. I do a slightly faster tempo with casual but VERY articulate and excited speech. Many of my comments from my lectures and videos point this out as a positive. Rather than users needing to 1.5x the video, remember that its a recording and if they don't get something, they can rewind. Again, I try to be very articulate with my speech (with no accent) because I know not everyone is from my state. I try to follow an "assume they no nothing but want to learn it" approach, so nothing is hand waved away.

In regards to "what to record" - it helps to have a general plan. There's a lot of intro CS material out there, but that's not always a bad thing. Try to think about what students trying to learn are looking for. Some of my more popular videos are about Microsoft Access.

Growth and Revenue - I can't offer much help. I get some ad revenue like once a year at most. However, the videos are supplemental for my students in my classrooms. I'm not trying to make it a career, just trying to offer people not in my class a chance to see my explanations.