Part one in a three-part series in understanding podcast consumption and measurement.
As many of you are aware, unlike digital marketing, there are limitations when it comes to some aspects of podcast measurement, including delivery and consumption. In response, Veritone One has created internal solutions allowing us to identify a variety of metrics to help reduce the impact of some of these measurement shortcomings. This is the first in a three-part series to help you navigate and interpret the current measurement platforms and understand how they can all coexist.
Step one: a definition of terms.
Defining a Podcast Download
First, there is a difference between a podcast download and someone actually listening to the podcast. A download is measured when a consumer requests the episode and has it delivered as a file to their device. Some networks and measurement tools consider the request a download, and some wait until the full file is delivered, so it does vary depending on the measurement tools. Veritone One considers a download when the file is complete, not when it is requested.
The download is the most significant factor in terms of measurement as somewhere around 80% of podcast delivery is in download form (note this number varies dramatically based on show platform, etc.).
There is another type of download, called a progressive download, which is where an app/browser begins downloading the episode to a device and begins playing once enough of the program is buffered. The distinction between a progressive download and a stream is that the file is being transferred to the device and eventually resides on the client device. This is the behavior for many podcatchers (podcast apps) and what Apple uses in the default iOS Podcast app. This behavior typically occurs when an episode is played from the podcast app when the file has not been downloaded. This typically happens when the user is listening to a show that they don’t subscribe to. At this point, this is a smaller group of podcast requests but continues to grow.
Some platforms rely solely on streaming as the delivery method of podcasts to subscribers, but this remains a very small portion of podcast consumption, with Spotify being the largest podcast host to utilize this extensively for podcast delivery. Depending on the show and factors like the production and marketing strategy of a show, the percent streamed can vary dramatically for an individual show—for example, a Spotify exclusive show that is not released through iTunes or other platforms will have a significantly higher stream percentage. Spotify distributes its content only by streaming so episodes are not downloaded to their client software. Streaming metrics significantly impact a small number of shows as the Apple ecosystem still owns a majority of podcast discovery and consumption.
Even after dissecting the different types of delivery, there are a number of variances in measurement from platform to platform. This has led to the IAB’s push to create podcast measurement standards, looking at things like what is a completed download, what is a unique audience, etc. and what are best practices for counting. While the IAB standards are accepted by many leading publishers, given the huge number of independent podcasters, it has been adopted by a small number of podcasts, so we continue to work with networks and shows to help identify how they differ from IAB standards so that we can better estimate audiences across networks and hosting platforms.
This is just a glimpse into a few variables that impact podcast measurement, so you can begin to imagine the difficulties in creating industry-wide reporting that is representative and comprehensive. We provide you with this background so that it gives you greater insight into evaluating industry-wide analysis.
Continue to the next blog in this series, COVID-19 Impact on Podcast Listening – Publisher Analysis