The Right Length for a Podcast is ...... D.W.M.T.

The debate continued last week in trade papers and blogs over how long a podcast episode should be.  Many with long experience in podcasting point to the top ranked podcasts, which typically run about an hour, as the ideal length.  A significant number are produced by public radio with a history of hour-long show blocks, and while some are time-shifted from over-the-air broadcast, most - 24 of the top 30 podcasts - are organically built as podcasts. 

The argument from longtime podcasters, is that since it is a lean-forward medium in which people select a show, podcasts appeal to an engaged audience with an expectation and desire for longer and deeper dives into topics. 

The other side of the debate comes largely from commercial broadcasters and podcast aggregators who know the tyranny of getting listeners to hang through precious quarter-hours of over-the-air broadcasts and worry about listener drop-out.  

While indeed there is currently a bias to longer podcasts, there are two new entries, which are tearing up the podcast charts and their length may be instructive.

The Daily from the New York Times runs about 20 minutes and has had 27 million downloads since its inception in February.  NPR’s new “Up First” began a few weeks ago and rocketed to the top of the charts and runs about 12 minutes each day.  Both are thoughtfully constructed with busy commuters in mind and clearly the short length is part of the appeal and success of these programs. 

The optimal length for a podcast is not easy to determine. There are hard-to-listen-to shorter podcasts and engaging longer ones. The larger, and more complex issue goes back to an old Winston cigarette commercial; "It's not how long you make it, but how you make it long." Quality rules.

Here is another thing to consider; audio is generally consumed while people are engaged in other activities, whether it be driving, walking the dog, cleaning the dishes, or at the gym. Through many focus groups and studies of listening patterns I have seen over the years, people stop listening when they are done with an activity.  No one sits in the car at the parking lot at work.  No one stays on the treadmill longer – okay that one may just be me. 

Content should be designed with the end-user in mind.  One clear finding in just about any media use study is people report an abundance of choice and a scarcity of time. What's the listener attrition on longer podcasts episodes?  Do people look at time codes and determine the length they want?

Print and digital journalists, have experienced a change in consumption largely dependent on time, place and device in use.  A brief mobile news update in the morning is potentially different than a weekend read. 

While at The Worldwide Radio Summit last week, I was reminded of the importance of actual use patterns driving podcast length. Fred Jacobs, President of Jacobs Media debuted Techsurvey '13 in which they collect data from thousands of audio listeners. The accompanying chart shows varied activities - note people could choose more than one option for this question.


When audio creators look at content development through the prism of the end-user, and factor in how and when people actually consume the program, it should inform some of the architectural aspects of the show.

Here's a good axiom to follow - DWMT: "Don't Waste My Time"

More than ever, listeners penalize media outlets for wasting their time. So here's a good axiom to follow - DWMT: "Don't Waste My Time."   

Go as long as you must......then stop.

Few will stick around if they are not being informed or entertained. They have choices.  

At Podcast Movement this August, my company, Amplifi Media and Nuvoodoo will team to present a breakthrough video study in which we interview real people about how they relate, react and listen to podcasts. Do they look at time codes?  Do they bail on longer podcasts? Do they select by the subject line?  It is revealing and might change some thinking about the relationship between podcaster and listener.