Hey everyone, we all try different ways when looking for content on YouTube. Some prefer using the algorithm, while others rely on community recommendations. Which method do you think is more effective? Or how much does the video you watch shape the next recommendations? Share your experiences on which approach actually works!
What are the best methods for recommending content on YouTube?
👁️ 10 views💬 2 replies❤️ 0 likes
2 Replies
Yesterday, after watching a tech video, YouTube started recommending completely irrelevant "10-year-old laptop advice" videos to me. From the next day on, I cleared my watch history and started following only relevant channels, and now I'm getting much more consistent recommendations.
The algorithm’s suggestions are definitely worth trying, but I always start by typing keywords into the search bar and manually filtering the results. For example, when I searched for "microservices with Golang," most of the videos that came up were current and high-quality, whereas one of the top three videos the algorithm directly recommended hadn’t been updated in three years. So the algorithm is useful, but you can’t blindly trust it—you absolutely need to vet things yourself.
After watching a video, I end up liking over 70% of the algorithm’s follow-up recommendations, especially long-form follow-ups and in-depth analyses. For instance, after starting a series on Spring Boot in Java, the system almost instantly surfaced videos like "Migrating to Spring Boot 3.x" or "Deep Dive into Spring Security," which blew me away. Community suggestions, on the other hand, feel more hit-or-miss; sometimes the content that pops up in Discord servers or Reddit threads is better than what the algorithm offers, but it takes way more time to dig up.