Podcast Analytics on Squarespace

It's been a while since I last wrote about podcasting, so it's time for some followup. I recently took inventory of my Brotherhood subscriber numbers across both Squarespace and Podtrac and discovered a pleasant, if somewhat overdue, surprise:

Weekly subscribers to Brotherhood (according to Squarespace)

Weekly subscribers to Brotherhood (according to Squarespace)

Nearly all of the subscribers that Squarespace had been reporting for the past few months are gone. Perhaps I should be sad, but since those numbers were for an RSS feed I've never shared with anyone, I can't help but feel relieved. That's the closest I think I'll be getting to a "Congratulations on correctly diagnosing the system, Dylan! Here's a sticker and some ice cream." Meanwhile, my Podtrac numbers have steadily grown 200% their original size since I began paying attention.

This has prompted me to think a little more deeply about Brotherhood and other potential future podcasts. In some ways, these numbers are reassuring. In others, I'm more than a little worried.

One thing is clear: never rely on a single source of information when it comes to who's listening, and always question where those numbers are coming from. Subscriber numbers are already notoriously tricky for podcasts to report. Having a second, more reliable opinion can only help you better understand them.

But that's not all I've learned looking at a years worth of podcasting data. There are a few more points worth making, here, that we may find useful in trying to understand how podcast listenership works.

Posting Regular Episodes Isn't That Important

It can be hard to tell if podcasts have a growth curve as tightly coupled with regular posting as YouTube does. How can you tell if a regular schedule or a large repertoire is more important when the two are so frequently seen in unison? Which comes first: the benefits of a regular schedule or the benefits of a back catalogue. Thankfully, I have a decently-sized show with a patchy upload schedule to help isolate those variables.  

Unlike my Squarespace numbers, which skyrocketed during the hiatus and have dropped now that I'm regularly posting again, my Podtrac numbers have almost never dropped, regardless of when or whether new content was published. I trust Podtrac more since podcast tracking is their whole job, so I'm willing to bet their version of events is closer to the truth. That would seem to indicate that there isn't a similar growth curve at all. Posting regular content, it seems, isn't as important as just posting content.

Which is a relief, really, since editing these shows can take a long time. And now that we've added a new host, editing Brotherhood has only taken longer. Posting that show every week has been far more stress than it's worth, especially given how small and relatively unknown the listenership is, right now.

Discoverability

There's a bigger reason why podcasts don't seem to grow like YouTube videos. YouTube has discoverability. Podcasts do not.

YouTube analyzes your viewing habits and curates content it believes you're likely to enjoy. It makes recommendations based on your own history, not just the popularity of the channel. I've discovered some real gems this way -- channels with a microscopic subscriber count which perfectly suited my tastes. iTunes does nothing remotely similar. Stitchr may be trying to combat this, but most people still only listen through iTunes.

When last I wrote on this subject, I theorized that Brotherhood might be gaining more phantom subscribers because of its popular subject matter. Bots -- I assumed -- must be searching for anime related RSS feeds and my podcast was included in the mix. What I did not take into account was that real people might be required to search for the podcast themselves.

This means that you have only two options if you want to be discovered online: latch onto existing trends that people are currently searching for, or create content for a service that algorithmically curates content like YouTube. Most of the interesting ideas for podcasts fall into neither of those categories.

In Closing

Podcast analytics are a quagmire. I'll most likely be writing on this subject again. And I may have to make some uncomfortable decisions about my work in the future. That will probably come up again, too.