YouTube Experiment

In 2021, I have started a YouTube experiment. I am trying to upload 50 videos this year. Learn more about why here.

With each video, I am trying to learn at least one thing about making videos or improving my presenting skills. The following is every video, in order, along with some stats and some lesson(s) learned.

1. Take small steps towards your goals (Jan 18)

This video took me an hour to write the script, 30 minutes to record, and 3 hours to edit. There were so many issues that I didn't want to spend too long on the edit (plus I am trying to make this habit stick). Currently at 33 subscribers. Lessons learned:

  • Turn out the room light to help separate the background
  • Annunciate better
  • Deliver larger chunks, don't restart in the middle of a section
  • When starting over with a section, give a 3 second break to make an easier cut point
  • When there was a good take - do three claps to easily find it during editing (or stop recording)
  • Don't move back and forth in the chair, make sure each take is in consistent position or intentionally different (e.g. left or right)
  • Pick a better shirt (make sure there isn't cat hair on it)
  • Put 8% more energy into it
  • Do audio adjustments before cutting clips

2. Powerbeats Pro Review (Jan 25)

I spent 3h 50m on this video. This week I learned a lot of the Final Cut Pro (FCP) shortcuts which really helped with the rough cut. I also learned about keyframing so I was able to improve the visuals/broll. I used a lavalier mic this time which helped with the audio. Three claps did not help. Things for next time:

  • Add some background music
  • Remember to smile at the camera for a few seconds to grab a thumbnail
  • Learn to get comfortable with talking to the camera (spend more time practicing delivering lines without recording)
  • Always double check frame rate and white balance (basically all of your settings)

3. Reviewing PARA (Feb 4)

This video took me 3h 30m. I shot it pretty quickly since I spent some of my recording time working on another video. I am now at 34 subscribers! I attempted to add music but I didn't have enough time to find the right mood with it so I cut it. This video is also going up a couple days late (still within the week!) since I needed to do a little more work on my real job. Oh, and I started using TubeBuddy, description defaults, and added an end card. Lessons learned:

  • Read through notes and practice once before recording
  • Attempt to record two or three good takes on camera (get multiple options)
  • Use zebras on camera to check exposure
  • Manual focus doesn't stay as sharp
  • Watch out for too many umms
  • Leave more time for the end card

This was a fun video, attempting to do a Daniel Schiffer style commercial. It was a lot of work to get my cheap lights set up and to go through multiple shots. And at the end of the day I still didn't have enough shots. I forgot to record how much time I spent setting things up and recording, but I estimate I spent a total of 7h on this video. Lessons learned

  • Should have used 4K in order to fix alignment issues
  • Should have used slow motion for some of the movement shots to better control overall flow
  • When doing a montage like this, you need a lot more variations of shots
  • When doing a talking head and reviewing something (like cookies), having more close ups (of me and the cookies) for b-roll would have been helpful

5. Chocolate Mousse (Feb 16)

This was a last minute video that I put together for valentines. I think the shooting itself took an hour but the edit took closer to three hours. Without any planned shots/story it didn't turn out that great as a video, but it gave me a lot of practice with speed ramping (so much speed ramping) and slide edits. I didn't have enough time or resources for finding a good soundtrack. I ended up using YT's free music library (as I have for everything so far).

  • Plan out a video before hand
  • Figure out the story and what narrative should go with it
  • When doing locked off videos, the transitions have to be based on movement
  • Music takes a long time to pick out

5.5 Cookies & Cream (Feb 18)

I had seen some videos recommending creating shorts. I attempted to quickly do that with this video. I didn't track the time it took to make, but I think it was less than an hour. The video is 0:59:23 (one frame short of 1 minute) but Google marked it as 1:01 and it got 0 views. Also around this time, I lost a subscriber. Back at 33 subscribers.

6. Google Search Tips (Mar 2)

I ended up recording this twice. I had attempted to record it in my living room and the audio didn't turn out well so I wrote a script and shot it in my corner office. The second time around I think I spent around 7 hours on it, 4 hours editing.

  • The length of the script was good (shooting for > 10 minutes).
  • While I didn't use the script verbatium, having more thoughts written out did make it easier to cover all the material I wanted to.
  • I need to remember to take dedicated photos for the thumbnail.
  • I also did intentionally try to say an entire paragraph worth in one take which made the video flow better (but meant a lot of extra video I had to shoot).
  • This was my first time using J and L cuts in my videos and I think it helped.
  • I also included a table of contents for the video for the first time.

7. LastPass changes from a product manager's POV (Mar 9)

This video took so long that I wasn't able to track all of my time. I think it was something like:

  • 1:00 - writing script (including a lot of research)
  • 1:30 - recording
  • 6:30 - editing

For a total of 9 hours on this video. My rough cut came in at 22 minutes and the final video ended up being 19:25. My longest video so far. Much longer than I want to go for. I had over an hour of video to go through and cut down.

This was the first video where I felt like more of my actual personality was coming across. Awkwardness and all. The few sections when I went off script are some of my favorite parts. But it does shows where I need to improve my communication skills (which is why I am doing this!).

This is the first video I included a reference to a previous video. +1 for having a backlog! I had skipped a week previously so trying to make sure I got this uploaded by Tuesday, I stayed up late. :/

Lessons learned:

  • J and L cutes are my new best friend.
  • The benefit of deliverying as many lines together as possible is that your voice is consistent, your hand motions are tied together, and your posture is the same.
  • I need to stop slouching and sit up stright the entire time.
  • Recording longer deliveries takes a bit to get into it and is harder when I have facts (numbers) I am tryign to memorize/deliver. Try to avoid that. :)
  • Color grading at the beginning is good before you start cutting everything up (I put it all into a composite clip).
  • Doing audio adjustments up front makes it better to listen to during the edit and has the added benefit that it is easeir to see talking waveforms instead of background static. Easier to identify gaps in audio and make better transitions.
  • I tend to extend the last word in my sentances or as I pause to think. This makes for odd sounding sentances when I remove the gaps.
  • There are a couple segements where I go off script and I think it comes across more nautrally, but the thoughts aren't always super clear and my grammer sometimes falls apart. This is the big thing I am trying to learn through all of this: present my ideas clearly and with structure.
  • Get a shot of the background without you in it makes will make it easier to create a thumbnail if you need to move an overlay of yourself around.

8. Wrapbook's business model (Mar 23)

I attempted to just riff with this one. Minimal bullet points and just talk. That was a mistake. I ended up partially starting over half way through (which I forgot until I started to edit) and I ended up recording for an hour. All of that made the edit a lot longer. I ended up spending just shy of 5 hours on this one. It is also out a week later than I wanted, but I am trying to keep in mind the long game. I want this to be sustainable and I don't want to burn out within the first few videos. I know I need to improve the time it takes to edit, so I am going to work on that. Oh, and I got some new subscribers! I am up to 37 now.

Lessons learned:

  • Practice the entire video a couple times before recording. Having an hour of video to go through adds a lot of time to the edit.
  • Time spent planning upfront always pays off (does in programming too).
  • Mark each section with chapter markers so you can more easily rearrange and figure out what needs to get cut.
  • When doing a screen recording, decide if it should be flipped to the entire time or just pulling screenshots from it. It will change the way you want to edit (edit together vs just finding the snippets needed).
  • Unless you know the video is short and you aren't cutting parts out, avoid saying, "like we talked about" since that section might get cut. :)
  • I should invest some time in creating standard titles/fonts to create a standard style. Right now I have just been defaulting to use Baskerville. Hopefully I can speed up editing and get to adding some background music eventually.

9. Cricut and pricing changes (Apr 12)

With this video I attempted to do a dry run through before recording to help cut down on that part of the process. Problem was that it took another two days before I got a chance to record so I don't think the run through helped very much. I want to try that again, but I also want to learn to speak more off the cuff.

I've had another bump in subscribers, I am up to 40 subscribers now.

I'd also like to spend more time on the edit and adding flourishes. But more importantly, I'd like to be uploading every week instead of 3 weeks later. Here is how my 9.5 hrs for this video broke down:

  • Research: 30 min
  • Writing: 2h 20m
  • Practicing: 30m
  • Recording: 1h
  • Editing: 4h 15m
  • Thumbnail & Upload: 1h

The biggest improvements look like it would come from writing and editing. Most of the editing is going into cutting the recording down. So if I had a cleaner takes, that would go faster. I've thought about stopping the recording every time I have a clean take. That might make it go faster.

Main takeaways:

  • Again, try to repeat entire sentences, not just phrases.
  • Give longer pauses before starting a new section.
  • When stuttering or struggling to finish a sentence, instead of starting over, go ahead and finish the sentence. Build the mental groove for the entire phrase instead of just the first part.
  • Prep-work didn't seem to save too much time. Options: (A) try to do more prep work and closer to when recording will actually happen. (B) try to cut down on the amount of script writing and go ahead and riff for an hour on video.

10. AF Motor Sounds (Nov 24)

This was more of a quick test of the auto-focus sounds coming from my R6. It had been awhile since I posted (moved and had a kid since the last post) and I wanted to get something small out as a jump start. Plus I wanted to test out my new MacBook Pro with the M1 Max chip (tl;dr it is awesome).

As such, I didn't track my timings around this one.

I currently have 56 subscribers.

11. CLog on MacBook Pro M1 Max (Dec 9)

This was another quick test. I figured I should start small and build up my process in this new place. I'm not a fan of the (lack of) background, but I'm more worried about content that I'm going to focus on.

Right now the most popular video is the PowerBeats one with over 4K views (the LastPass one is #2 with 249 views).