Using predictive analytics to discover the next “Serial”

itunes top podcasts 1024x585 1

July 2024 Update:

As podcasts continue to grow in popularity, understanding what drives their success is crucial for creators and marketers. Predictive analytics offers a powerful tool to uncover the communication factors that can make a podcast go viral. For example, the podcast “Serial” has shown that storytelling and inclusive language are significant predictors of high iTunes rankings. Storytelling engages listeners by making content personal and relatable, while inclusive language encourages audience participation and connection. Analyzing top podcasts using predictive communication analytics can reveal the most effective elements, such as emotional language, clarity, and engagement. By leveraging these insights, businesses and content creators can craft podcasts that captivate audiences and achieve higher rankings. This approach not only applies to podcasts but can also enhance other communication formats, ensuring content is both high-quality and effective. Embracing predictive analytics in content creation is essential for staying competitive and resonating with today’s audiences.

What makes a podcast popular?

Podcasts have risen in popularity by more than 25% year-over-year, now counting nearly 40 million American listeners.

Some of this success can be attributed to “Serial,” a podcast from the producers of “This American Life,” which tells the story of a murder case from 1999. According to an article in The New York Times, “Serial” has been downloaded or streamed more than five million times, with an average of 1.5 million listeners per episode. To put this virality into perspective, “This American Life” took four years to reach one million listeners, while “Serial” reached that audience size in a month.

Given the rise in popularity of podcasts, and as corporations consider the value of producing their own podcasts, we wondered:

Can we use communication analytics to predict the future popularity of a podcast, based on its content?

Predictive model goal: Identify communication factors that best improve iTunes podcast ranking.

Data source:

We started with a random sample from iTunes’ list of the top 200 podcasts as of 6/8/15. We ran the transcripts from each podcast through our predictive communication analytics platform to measure the message content and analyze each podcast on over 70 communications effectiveness metrics to find the factors most significantly correlated with iTunes’ podcast ranking.
iTunes-Top-PodcastsCommunications measures:
Want to create a viral podcast? Here’s what matters:

  1. Storytelling – a good story is personal, relatable, and includes emotional language.
  2. Inclusive language – a measure of the linguistic markers that signal an environment and culture where everyone is encouraged to participate in the discussion or share their opinions.
  3. Negative sentiment – exhibiting negative emotions, such as anxiety or inhibition.
  4. Clarity – a measure of how easy it is for the audience to follow and understand the content.
  5. Innovative – a measure of the language used to articulate a departure from the norm through achievements, new products or ideas.
  6. Engaging – The ability of the speaker to capture and hold the attention of the audience.

Prediction:
Using a linear prediction regression, we found the following two communication metrics to be the most significant predictors of a top iTunes’ podcast ranking:

  1. Storytelling – a one-unit increase in storytelling leads to a 2.5 boost in iTunes rank.
  2. Inclusive language – a one-unit increase in inclusive language leads to a 0.8 boost in rank.

Applying these findings to creating a podcast (or to any communication setting)

As podcasts continue to grow in popularity, so does advertising revenue for businesses and direct purchases for advertisers (63% of podcast listeners bought something based on a podcast ad.) With so much upside, more and more businesses and advertisers will jump on the podcast bandwagon. But to increase the likelihood that your podcast will be a success, give listeners what they want: great stories using inclusive language.

While this analysis focused only on podcasts, the findings can be applied to other communication settings as well.

    1. Storytelling

      A great storyteller takes the audience along with them, step-by-step, through the events. They allow listeners to make their own judgments based on the information they’ve been given.

      Given its popularity, let’s use “Serial” as an example of great storytelling. “This American Life” host Ira Glass explains the new show this way:

      “It’s a case where what really happened is actually much more complicated than the jury ever heard when this thing went to trial. And each week we will go with Sarah on her hunt to figure out what really happened. And we will learn the answers as she does.”

      Episode 1 of this podcast scores in the 90th percentile of our communications database in terms of storytelling, largely because of the descriptive language used to relate the events as the narrator experiences them.

2. Inclusive language

Inclusive language is a good way to make the story relevant to the audience. By asking them to think about your content in terms of their own lives and experiences, you encourage them to participate in your story.

In Episode 1 of “Serial”, host Sarah Koenig uses inclusive language to bring in the audience:

“Before I get into why I’ve been doing this, I just want to point out something I’d never really thought about before I started working on this story. And that is, it’s really hard to account for your time, in a detailed way, I mean.

How’d you get to work last Wednesday, for instance? Drive? Walk? Bike? Was it raining? Are you sure?”

Because the audience is asked to apply the narrator’s theory to their own lives, they are further drawn in to the story.

Want to tell a more effective story?

Whether through a podcast or any other medium, walk your audience through the events as you experienced them in order to keep them coming back for more.

The moral of this story: Regardless of the communication setting, businesses can use predictive analytics to better shape their content to help increase both quality and effectiveness.

To learn more about how we can help your team use predictive analytics to transform and improve your communications, contact us at info@quantifiedcommunications.com.