This took a couple hours. I've been reflecting and analyzing a lot of videos lately, and I've tried to distill some principles, that should be the bedrock of content success. Those principles are pretty meta, and for this reason they lack detail. Disclaimer, I haven't had any success with YouTube, so this is theory. But it could provide some useful insights, and I'd be curious to see the opinion of some experts, I think we'd all benefit. So, here goes: -------- The Algorithm (what we all know) -------- The YouTube algorithm uses metrics to push videos: - Click-Through Rate (CTR) - Audience Retention There are other metrics obviously, but those are the ones we want to maximize. -------- Click-Through Rate -------- You can manipulate two things to increase this metric. A great CTR is 5%. - Thumbnail - Title The thumbnail is what will catch the eye. The eye sees thumbnail → title The Mental Process 1. Capture Attention - Viewer’s attention is captured by the thumbnail 2. Curiosity - Viewer reads thumbnail and title. I define curiosity as a desire to know 3. Hooked! - Viewer is curious and opens the video Capturing Attention The thumbnail is what captures attention. Our brain is much, much quicker at working with images. To capture attention, the thumbnail should stand out in the sea of videos in the YouTube home page. This can be done with a big face, with colors, with arrows, with text, or with a simple image. The important thing is that the eye is drawn there, and the thumbnail truly stands out. There are millions of ways to do this successfully. Instead of trying to explain all of them, ask yourself: Does this stand out? A thumbnail that stands out matters more than one that creates curiosity!! Look at the mental process! Curiosity Without step 1, people don’t even know about your video’s existence. But without a reason to open it, the viewer will keep scrolling. Curiosity is achieved through a combination of thumbnail and title. There are millions of ways to do this successfully, but there are a few main categories: