This is the second post in a new series at Pacific Content, Common Client Questions. Check out our last post in the series: Should I Put My Podcast on YouTube?

If you’re anything like us (which, if you’re reading this, you definitely are!), then you’ve been noticing a difference in episodic consumption rates between Apple and Spotify. Mainly, when you’re digging into your stats, you’re finding that your Spotify consumption rates look different than your Apple consumption rates (i.e. they might be 10-15% lower in general, etc). This can seem quite alarming, and we have received a ton of questions from clients and colleagues alike wondering why this is happening. Do listeners on one platform like you better than the other? Could the slight variations in listeners on each platform play a significant role? Is your show banned to the shadow realm and doomed to fail before it even started?!

We can’t definitively rule out these theories (except for the shadow realm). But what if I told you the most likely answer is much, much simpler?

Turns out, Apple and Spotify calculate their consumption rates in different ways. They are not the same calculation—therefore, they are almost always going to look different.

So, how do these platforms calculate retention, then?

Apple Podcasts assesses the average behaviour of listeners. This includes how many times each person has listened, how much of the episode they listened to, how many times they’ve pressed play, any spots in the episode they skipped, and at what point they left. The average we receive is a total average of all of this data.

Spotify assesses the median behaviour of listeners. They share the point in the episode where 50% of the listeners who started were still listening.

Clearly, there is a difference in data presentation. Each platform has taken a point of view that they consider the most beneficial to us, the data nerds. However, when we pull back and analyse our two data sets, we should be seeing very little difference. We can expect our targetted audiences to behave generally similarly, regardless of what platform they are on (barring some sort of bizarre glitch-in-the-Matrix situation). Therefore, we should expect both Apple and Spotify charts to tell us a very similar story.

But Mary, how can I see that they’re telling a similar story?!

Don’t worry, I’ve got you. Buckle up, this is about to get detailed.

Both Apple and Spotify offer granular views of your episodic performance. If you take an episode and compare the performance it achieved on Apple and Spotify, you will see the graph looks very similar, with only slight variations.

But let’s start by making sure we know what we’re looking at.

A brief explainer of how to read the charts and data provided.

Let’s look at this episode for example (anonymized for privacy):

Apple Podcasts Connect (data provided for every 15s of the episode):

This is the episode’s chart in Apple Podcasts Connect. They have provided the listeners, engaged listeners (A.K.A. those who stuck around for either 40% of the episode or 20 mins, whichever comes first), and plays (A.K.A. how many times people pressed the play button on the episode). When we go by each 15s interval, we can see the amount of listeners listening at that moment. Apple has averaged all this behaviour into their Average Consumption value, which we see is 84%.

Spotify for Podcasters (data provided for every second of the episode):

Note: Spotify does not present an overall average of its quartiles.

This is the episode’s chart in Spotify for Podcasters. For each quartile, we see how many people are still around by the end of the quartile. So, at the end of the first 6 mins and 22 seconds, 75% of listeners were still present. At 12:44, 66% were still remaining. At 19:06, 63% were remaining, and at the end of the show, we’re at 40% still hanging around. 

This is corroborated by the median play time, which is listed above as 24 mins and 28s. We can see in the chart that at 24:27, 50.34% of people were still around, and then at 24:28, 49.66% are still around.

Okay, I can read these charts. How do we know they tell the same story?

Right off the bat, we can see that despite the vastly different data presentation, both of these charts look very similar. I’ve highlighted the same timestamp on both charts for reference. Both show a different data point at that exact spot, which can be confusing.

But instead, let’s look at the behaviours we can immediately see:

Apple Podcasts Connect:

Spotify for Podcasters:

I’ve detailed the important behaviours in both charts in what is the greatest graphic design that you have ever seen. We can immediately identify that both Apple and Spotify have shown a drop in the first minute of the episode (Drop 1), followed by long, steady listenership, and then at the end at around 24 mins when we hit the outro, people start bouncing at two distinct spots (Drop 2 and Drop 3). Even though we’re looking at two different platforms, the audience is still behaving the same! There are no significant outliers above. There’s no random spike or dip or gaps. We see slight differences due to two different groups of listeners, but mostly very similar behaviour.

So, if we combine what we know about Apple and Spotify’s data presentation differences and our granular view of our episodes, we can conclude that that there isn’t a problem with one platform over the other, or a mistake in reporting, or anything of the sort. They are simply similar data presented in two different ways.

Okay, okay, I get it. They’re different but also the same. What does this mean for me?

We can combine this information and use it to guide how we produce our episodes and any editorial decisions we may make. Did people like the episode? How well did the topic, guest or story perform compared to other episodes? Were there moments in the episode that listeners skipped or stopped listening? Are the majority of our audiences making it to the end of the episode? If not, why? Is it an issue with the content, or the marketing strategy? What can we do differently next time to ensure maximum success? Of course, make sure to consider how big of a share of your audience comes from each of those platforms before you do (or don’t) make changes to your editorial process.

What method do you prefer for your analysis? Apple’s averages, or Spotify’s medians?

Personally, I’m waiting for a third platform to arise from the ashes and toss a wrench in everything by presenting the mode for each episode.

Sign up for the newsletter for the latest brand storytelling and podcast inspiration, insights, right in your inbox, every two weeks. We promise to never spam you.

Got an idea or brand challenge you'd like to talk to us about? Contact us using this handy form.