As a podcaster measurement nerd, I sometimes envy people who work in other online media. Here’s why:

• A YouTuber can easily tell you how many subscribers they have on their channel
• A Twitter or Instagram personality can easily tell you their follower count
• The proprietor of a popular newsletter can easily understand the size of their email list

YouTube subs, social follows, and email lists all measure the size of your most loyal audience — the number of people who have opted in to hear from you on a regular basis.

But if you ask a podcaster for the equivalent metric (subscribers, followers, etc.) on their shows, you’re likely to get a shrug. 🤷‍♀️

It’s a simple question: “How many people have opted in to be notified about new episodes from my podcast?”

There should be a simple answer. But there isn’t.

That’s a problem.

Why subscribers matter

Jerod Santo from The Changelog said it best:

The stat that matters to podcast advertisers is downloads/listens, but subscribers matter to podcast creators because subscribers form a base for future growth.

Downloads are a useful yardstick, especially if you’re selling direct response advertising. But if your goals in podcasting go beyond selling ad inventory, subscribers matter.

Podcasts are a medium built on loyalty, habit, and an ongoing relationship with your listeners. If you care about creating those long-term, meaningful relationships, downloads alone don’t cut it.

Subscribers opt-in. Subscribers are telling you they want to hear more. Subscribers are in it for the long haul.

But here’s the thing: measuring podcast subscribers is tricky. Too tricky.

Why is measuring subscribers so hard?

Podcast subscribers are hard to measure because the listening app landscape is fragmented. And the podcast hosting landscape is fragmented.

There is no single source of truth for “subscribers.”

Some apps report a subscriber count. Others don’t. They all use different methodologies, and they often use the same word (“subscriber”) to mean slightly different things. Let’s take a look at a few of the players.

Apple Podcasts

Apple’s Podcasts Connect portal reports the number of “subscribed devices,” which is a count of “unique devices that were subscribed when users played content from an episode” across a certain date range.

The play requirement makes Apple’s definition of a subscriber very different from, say, YouTube’s definition of a subscriber. For example, I can subscribe to your YouTube channel, never watch one of your videos, and still count as a subscriber. But in order to count as an Apple Podcasts “subscribed device,” I need to have played one of your episodes within the defined reporting window.

Measuring RSS feed pings

Hosting services like FeedPress, FeedBurner, and Squarespace report a subscriber number based largely on the number of unique devices that have pinged a podcast RSS feed over a certain window of time. A unique device is usually just a unique combination of IP address and user agent, which means these numbers are subject to inflation, especially as mobile devices move around and acquire new IP addresses.

And, of course, not all podcast listening apps ping podcast feeds directly. Some podcast apps (like Google Podcasts) use a centralized feed crawler to check for new episodes, which means individual devices don’t show up as subscribers in services like FeedPress.

Reporting via user agent

Overcast/1.0 Podcast Sync (9462 subscribers; feed-id=795196; +http://overcast.fm/)

Certain podcast apps (like Overcast) use a centralized feed crawler, and report subscriber numbers in their user agent string. But as the folks at Castro have pointed out, podcast app makers don’t always know when users have uninstalled their apps (a de facto unsubscribe), so some app-reported subscriber counts are suspect.

And some other podcast apps (like Google Podcasts) use a centralized feed crawler, but don’t report subscriber numbers anywhere — not in the user agent, not in a private dashboard — nowhere.

Spotify’s “followers” metric

Every podcast listener on Spotify corresponds to a user account, which allows Spotify’s podcast dashboard to report a “followers” metric, which is akin to a “subscribers” metric:

Followers measure the number of users who clicked Follow on your podcast on Spotify. Followers represent your highest-engaged audience on Spotify and they automatically receive new episodes of your podcast.

Spotify follower counts are reported in the Spotify for Podcasters dashboard, and aren’t reported in their crawler’s user agent.

Apps that display follower counts publicly

Some apps (like CastBox and Himalaya) publicly display “follower” counts on their websites:

However, these apps don’t always report follower counts via their user agent (or an API). Some apps like Breaker display subscriber counts on the web and in their user agent.

Let’s make this better

It took years for the podcast industry to arrive at meaningful consensus on the definition of a “download.” The IAB’s Podcast Technical Working Group did important work in getting us there. What would it take to get us to a similar consensus on “subscribers?”

I’m not sure what it’ll take, but I do have a personal wish list for podcast app makers:

• If you don’t currently report on subscriber numbers, start
• Document your subscriber count methodology publicly
• If you use a centralized feed crawler, report subscriber counts in your crawler’s user agent
• If you offer a publisher dashboard, make subscriber counts available programatically, via an API, and permit your dashboard’s data to be intermingled with other apps’

Subscriber counts certainly aren’t perfect right now. But they’re important. And they deserve to be a lot better.

Because if we only measure downloads, we risk missing an important piece of the pie — our most passionate listeners.

Sign up for the Pacific Content Newsletter: audio strategy, analysis, and insight in your inbox. Once a week.