Last week’s LTM 2K performed better than the previous week, but it didn’t start out looking promising. After 9 hours, the video had only received 4 impressions. Impressions refer to when YouTube displays the video as a suggested thumbnail. This figure was considerably lower than the previous week.
So, on Friday night, I sat wondering what I could do to try and get the algorithm to recommend it to more people. I added some keyword tags and waited to see if they would help. By the time I went to bed on Friday night, the video had garnered nearly 1500 impressions and 19 views.
That stark difference made me realize that just posting the videos isn’t quite enough for YouTube to figure out who to show them to. I spent the week testing out a new thumbnail but did not see much of a difference from when the impressions plateaued. This led me to believe that YouTube temporarily boosts your impressions when it gets an idea of what the video is about to see if it will stick with an audience. I confirmed this belief later by watching an interview with the algorithm designers. For new creators, YouTube tries to figure out who the audience is, and if it can’t determine an audience, it stops showing your video.
With this in mind, I am trying to get ahead of the curve a bit this week. I am going to start out with a better thumbnail than I have been using on my videos instead of adding it mid-week. I’ve shortened the lead time into the story to get to the content more quickly so that people can hear the story in the auto-play preview. And I have changed the recommendation on the end screen to be the previous video, LTM 2K #3.
So, what I hope to see are more click-throughs on the videos from a better thumbnail, a longer watch time before the drop-off – the average is about 2 seconds right now – and increased views for LTM 2K #3 from the recommendation at the end.
Let’s see how these changes go. If they don’t work, then I guess I just don’t have an audience 🙁 Or, more accurately, YouTube hasn’t been able to figure out who an audience is for my videos. I will post my findings next Thursday before the new LTM 2K comes out, with possibly a bit of analysis to go with the results.
I am a scientist after all. What’s a little experimentation between you and me?