This makes me wonder what the watch times are. If people are dropping out after 4 minutes on average, then they’re stopping in for the intro, but not sticking around for the content. Which would raise other concerns. Does it speak to attention span? Would it help if we serve the podcast up in specific 4-5 minute chunks and link to the full interview after? Does it speak to the content? I don’t know. It seems the content/guests is/are interesting enough to be cited on other sites: the hackernoon podcast mention
That people are finding value in the intro enough to stop in and check it out on a regular basis is pretty cool.
My feeling is that listeners are a different audience than viewers; I have no data to back that up; it makes me wonder if there are any converts from listeners to viewers, and what those figures look like (seemingly impossible to discern).
I feel like watching is a much bigger time investment than listening. If I want to sit down and watch someone talk for 10+ minutes, that means I can’t really do anything else with my time while I watch. Listening, on the other hand, is something that requires less attention and can be done while I’m working on some unrelated task.
I wonder if YouTube users are actually listeners even though it is a “viewing” platform. The distribution might account for the impression bump.
I wonder what that says about the location of the listener when they’re watching. I typically won’t watch youtube unless I’m home, but I’m a watcher more than a listener. As a watcher, I’m also one who actually watches rather than listens (maybe it’s more of a hybrid watch/listen). There’s something in the interaction (facial expressions, body language, etc.) that I need to see for some reason. So much more data…