If you wanted to figure out the most popular episodes of your podcast; how would you go about doing that?
I think there are a number of untapped opportunities for podcasters—and podcasting—if we had ready access to this information.
Put Your Best Of Foot Forward
Podcasting is far too reliant on a reverse chronological order listing for the episodes of our podcasts. Too reliant not only on our websites but also in our listening apps. In effect, we’re just sticking the most recent episode in new people’s faces as the first thing they’ll hear. Perhaps the only thing they’ll hear, because the most recent episode you published may not be your best episode.
Depending on where you are in your season or what breaking news you felt you needed to cover, that most recent episode might not even closely resemble your “normal” episodes. And if you’re engaging in feed drops or other cross-promotional activities, the most recent episode in your feed might not even be an episode of your podcast.
But if you knew which episodes were the most popular among your listeners, you'd have a nice “Start here” collection you could first offer to brand new listeners, ensuring they’re hearing the very best you have to offer first.
Regardless of how you (or I) feel about podcast awards, it’s pretty crummy that often only a single episode of a podcast is reviewed by the judges or committees. Many awards encourage nominees to create a custom “clip” episode, highlighting segments from various episodes. But come on. That’s not giving a true representation of what it’s like to actually listen to the podcast. That’s just a sampling, and it’s probably out of context.
However, if each show submitted a collection of episodes—three, five, ten, or pick a number—that represented the best of the show, now the judges and committee members have much more to go on. A much more true representation of how the show actually sounds to listeners. That should let them make a much much more informed decision on whether nominated podcasts—the entirety of those podcasts—are worthy of the award.
Sparking Innovation Among Podcast Service Providers
This will get me in trouble with music aficionados, but there’s a reason “greatest hits” collections from recording artists are quite popular. Scan your music streaming app of choice for your favorite band. Chances are, one or more list like that has been compiled.
Podcast service providers could present that previously mentioned “Start here” collection to new people just discovering a podcast. And here’s the cool part: When the new listener consumes all of those “greatest hits”, the app automatically subscribes them to the show’s main feed, and the podcast now has a fully onboarded and invested new listener consuming the show’s content.
Do We Already Have The Data?
But back to my question; how would you go about determining which episodes are your most popular?
Would you base it on downloads, assuming that the most-downloaded episode was the most popular? Well… I suppose that's a start. I think we put way too much emphasis on downloads. And we know that a goodly number of downloads are automated and never listened to. So I'm skeptical of downloads as the correct answer here. But still, it's a start.
I recently read a Twitter thread where some of the more forward-thinking podcast service providers were batting this idea around. It’s very early in discussions right now, but they seem to be rallying on it.
The idea is for the hosting company—with the express permission of the podcaster, I’d wager—to use download data to calculate some sort of relative popularity of a podcast episode and then pass that episode-level data into a new dynamic RSS feed tag. In turn, aggregators like directories or podcast listening apps could then do something special for those marked episodes, perhaps having them appear early for un-subscribed listeners or highlighting them in some other way.
But again, that’s using download data. These listening apps have better data than your podcast hosting company when it comes to figuring out actual popularity. Podcast listening apps know how listeners interact with an episode. How many times they play the episode again and again. How often they hit fast-forward. How many times it takes them to fully consume an episode. And much more listener-initiated data.
Perhaps the app could blend their own internal data with the information these hosting companies are considering of providing via the RSS feed?
One other often overlooked popularity signal is inbound link data. It’s an easy argument that the more times people choose to share a link to a podcast episode, the more popular that episode is.
Tracking the quantity of shares across social media or the web is doable, but it's rather processor intensive. Perhaps an existing social monitoring service would be interested in doing something like this for podcasts?
Though this is going to be a data normalization challenge as well. Think of all the various links that are possible to use to share an episode. A link to the episode detail page. A direct link to the media file. Or a clip of the episode shared from a podcast listening app. Or a deep link to that episode inside a specific podcast directory.
I wonder what I missed? If you have ideas on ways we could better discover the most popular episodes of podcasts, please shoot me an email. If you're a podcast service provider. I'd love to know what you think you could do with “popularity” data and how that would make your podcasting service better.
The Boostagram Corner is silent today after a busy yesterday. You can fix that for tomorrow if you’d like by visiting the Value For Value page on Podcast Pontifications. You can choose from many of the many options listed on that page. Probably too many options, come to think of it. Maybe there's a business opportunity for someone to build a better “ways you can support the podcaster” plugin or template?
I shall be back tomorrow with yet another Podcast Ponntifications.