The difference between the two videos may be in the written copy that accompanies the ad, the wording of the call to action, or other small variables. Test isolated variables so you know which element of the video is a determining factor. By running two versions of the same ad side by side you are able to measure their effectiveness against one another. See which one works best then try to create a version of the better performing ad to really optimize the campaign.
I do not think this is how most people use YouTube. YouTube videos are more like blog posts, and fit more effectively into the niche of content marketing. Sure, people will comment—but they do so in a manner similar to how they comment on blog posts. They come to view and digest videos, not necessarily share their thoughts about the day. Because of this, you should approach YouTube as content marketing instead of social media marketing.