How are Podcast Listens Calculated?

When you look at your podcast's analytics, the first number you probably look at is “Total Listens.” In reality, this number actually represents "Total Downloads", but that gets confusing.

Here's a quick breakdown of how we accurately calculate your podcast's analytics:

Downloads

A download is counted when a podcast app successfully pulls a small portion of your audio file and hasn’t done so again in the same calendar day.

Think of it like scanning a concert ticket: once we let a listener in, we ignore any repeat scans until the next day. This keeps accidental refreshes, brief scrubs, or tests from inflating your numbers. If a listener comes back tomorrow, that’s a brand-new ticket and a brand-new download.

We also attempt to ignore bots, which we detect using the User-Agent request header. If a request comes in from a known bot or an unknown user agent, we ignore the request.

Every few hours, Jellypod pulls the fresh counts, and updates your dashboard. Data may be delayed by around 24 hours if you just created your new show.

These rules adhere to the IAB 2.1 podcast measurement guidelines and are consistent with what advertisers expect. Accurate, honest stats let you charge fairly for sponsorships and understand what resonates with your audience. You can learn more about the IAB here: https://en.wikipedia.org/wiki/Interactive_Advertising_Bureau

Privacy First

In order to record accurate analytics, while maintaining your audience's privacy, we use an open-source analytics service, OP3, which adds a short prefix to your show's audio. When a podcast platform like Spotify fetches your show, the request first hits OP3, which records the download, and then we deliver your podcast to the listener normally.

When this happens, basic user data such as the timestamp, anonymized IP address, user-agent, etc. is recorded and then hashed for deduplication and preservation. OP3 never saves full IP addresses, and we only receive aggregate totals, not individual listener data. The most we know is how many people pressed play and the region they were in—nothing that can identify a person.

OP3 is MIT licensed and fully open-source, which means we can audit their claims of privacy and security by reading the code on Github.

Limitations

Although the current implementation works for all major podcast players, some sites have started to make the podcast ecosystem a bit more closed.

For example, if you upload your audiogram video to Spotify or Youtube, they will host your podcast on their own infrastructure - preventing us from getting notified when a user watches your video.

And, as of July 28th, 2025, if a user visits your podcast's custom website, we do not include video plays as a listen. Listens are primarily attributed to podcast platforms like PocketCasts, Spotify, Apple Podcasts, Youtube, etc.