I Make Thumbnails for YouTube

Over on the 2nd channel, I posted a video detailing my process for making thumbnails for my YouTube channel:

My videos target a more technical audience—and often deal with subjects crowds like those on HN might find interesting.

Technical audiences don't often enjoy the feeling of being 'bamboozled', especially in subtle ways that feels like marketing or deception.

Therefore YouTube thumbnails are often a contentious topic. In the early days, thumbnails were just a still frame from the video, meaning it was harder (but not impossible) to 'game the system' and introduce clickbait.

But many years ago, YouTube (and now most online video platforms) offered custom thumbnails, and once that was available, it was a race to the bottom to see who could tweak and hone their thumbnail artistry to 'get the click'.

With that backdrop, I don't hold it against anyone who decides to use a tool like DeArrow to use community-submitted video titles and still frames from the video instead.

However, for the vast majority of YouTube's users, they will see the thumbnails and titles provided by content creators, therefore for anyone who wants to get more than a few hundred views on a video, the title + thumbnail combination must be considered.

The video I embedded above goes through my thought processes, but I thought I'd call out a few specific points in this post, as I think many of my fellow programmers / engineer-types might not realize their importance:

• Vanishingly few people use subscriptions to browse YouTube content anymore. I do, but I also mostly read news via RSS feeds, another antiquated practice only weirdos who were around the Internet before it was 'Web 2.0' do. Because the algorithm is 'god', basically, you won't even have people who subscribe to your channel see your videos unless you can hit some minimum metrics for CTR (Clickthrough Rate), watch time, and 'engagement' rates (which is amorphous but probably linked to likes, comments, and watch time).
• Marketing is not entirely evil. Something I've learned over the course of my career. I never trusted sales and marketing, but there is some good there. Like getting anyone to ever notice (and especially pay for) the work engineers do.
• Clickbait to deceive is bad. But I think mild amounts of clickbait (just grabbing attention, really) is good, if it's not deceptive, and still conveys the central themes / ideas in the video.
• I always use real photography—my goal with most thumbnails is to have the entire thumbnail be 'straight out of camera' (I don't count tweaking exposure and white balance as not being straight out of camera...). Many creators are relying on AI more heavily these days, I only will use it if I need to do something like fill in part of a desktop surface without spending 30 minutes with the clone tool and smoothing brushes. But I don't begrudge creators who don't take actual frames from their videos, or photos of the equipment/setups their using. Most people don't have a small studio with good lighting like I do, so they find other ways to achieve the goal of a thumbnail: to get a click.
• I don't have problems with arrows, circles, etc. — I use them sparingly, and you can definitely go overboard... but they can be helpful tools to get someone to focus on part of an image, especially if I'm working with a thumbnail I didn't really plan in advance.

Most of the top YouTube channels now have thumbnail designers, thumbnail artists, concept artists, and they even have policies of not beginning work on a video until a certain number of catchy thumbnails has been designed.

For my channel, it's just me, and I like having a good video and storyline. Late in the process (usually the last step before upload), I'll hammer out a thumbnail.

That's not the 'right' way to do it. And probably not even in my channel's best interest, because "getting the click" is one of the two golden rules on YouTube (the other being "keep people watching as long as possible").

My goal is always to share things I've learned, whether that's about some new product that caught my eye, or a bit of tech like radio, GNSS, clustering, HPC, IoT, or whatever. Trying to get a wide audience to enjoy the videos is a bonus, but as long as I can earn a living (which I do, currently), I'm content.