05/02

This Is How Long Your Podcast Should Be

At NAB, Rob Walch, VP at Libsyn (and a very smart guy) attempted to debunk the “myth” that shorter-length podcasts are more popular than long-form shows. And he did it by sharing Libsyn data indicating that “84% of podcasts with more than 100,000 downloads were more than 51 minutes long while just 9.9% were less than 30 minutes.”

By contrast, tons of largely attitudinal research has shown that listeners want shorter rather than longer shows, and the length of a show is one of the obstacles to its consumption.

So what’s the truth? And can you argue with actual data?

Of course you can.

First, the “myth” isn’t that shorter-length podcasts are more popular than long-form ones. The assertion (not a “myth”) is that shorter, in general, is more appealing than longer to more people. Appeal takes into account potential usage, not simply historical and habitual usage. Rob’s data is purely historical and ignore the vast numbers of folks who listen to podcasts little or not at all.

Second, the more engaged you are in any piece of content the more you generally wish it to last longer. This is what makes a fan different from a casual listener. But without casual listeners you don’t grow new fans.

Third, length arguments skirt around the deeper issue, which is quality. If something isn’t worth listening to it can’t be short enough and if it’s really outstanding an hour is not too long.

Fourth, length is relative not only depending on the quality of the content but on the type of listener introduced to that content. A “new to the show” listener is likely to be more impatient and less tolerant of length than a “faithful fan.” Yet it’s certainly possible that a longer show signals to a new listener that the investment of listening is “too much to bother with” if she lacks the motivation or time to consume 51 minutes of show. While it’s true that any listener can stop listening whenever she wants, if you see a time-code for the show and it’s long, you may believe a little bit of listening is worse than none at all. What other research is really showing is that length is an obstacle for new listeners. Even my own research has shown that.

Focus less on how long your show is and more on how good it is, and how to effectively reach and introduce listeners… Click To TweetFifth, the Libsyn data is using what appear to be ambiguous labels. For example, does “100,000 downloads” mean 100,000 total downloads of all episodes ever, thus giving an advantage to the podcasts with the largest libraries, especially those that have been around a long time, are established, and are likely to come from public broadcasters or those who emulate them and tend to favor the hour-long clock? What about using the show as the basis of measurement rather than the episode (if, in fact, episodes are what’s being measured here)? What about separating established shows from new shows? Repurposed-from-radio shows from original online shows? Professional from amateur? Etc.

Fifth, what is the proportion of shorter to longer podcasts? And do they have fewer episodes to add to the total figures, thus creating a bias against them (if “downloads” represents all downloads of all episodes).

Sixth, it’s quite clear that downloading a podcast and listening to the entire show are not the same thing. I have hours of podcasts which have been passively downloaded but I have barely or never heard, and given that podcasting is driven by power users I am quite sure I’m not alone. And you know which types of podcasts that I am less likely to have heard completely? The longer ones.

So this is not to argue that you should create short or long shows per se. That debate is stupid because the “right answer” depends on the power of the content and the type of listener you’re out to attract. It also ignores how you open your show and tease what’s coming. These programming formatics are not irrelevant, even if they are ignored by too many show-runners.

What is certainly true, however, is that the more unfamiliar a listener is with your show the less likely she is to sample even a minute of it, let along 51 minutes of it. Perhaps to this listener a “shorter ask” will reduce the barrier to listening. Brevity alone is not nearly sufficient to promote sampling, of course, but it helps.

The biggest challenge facing podcasting in general is not show length, it’s the many listening barriers and the ecosystem’s dependence on distribution mechanisms filled with speed bumps and partners with no skin in the game (that’s you, Apple). That’s a big reason why today a relatively small fraction of Americans listen to even one podcast in an average week.

Ponder that as you debate how long or short your show should be.

My advice is to focus less on how long your show is and focus more on how good it is, and how to effectively reach and introduce listeners to it.

* = required field
  • Rob @ podcast411

    Here is my rebuttal for those not actually at the NAB session but reporting on it with out all the facts. http://podcast411.libsyn.com/sorry-virginia-there-is-no-such-thing-as-the-perfect-length-for-a-podcast

  • Well considering your “rebuttal” was written before my post I don’t know that you can properly call it that, right? Pre-buttal? Non-buttal, maybe. That post ignores most of my points (as it should, being written before my piece). Plus from my reading we come to the same conclusion regarding length, so who cares anyway. And I called you a smart guy and you would think I’d get some credit for that from you. 🙂
    Keep in mind, Rob, that the title of the article in Radio Ink covering your presentation was, and I quote: “NAB Show: Listeners Want Longer Podcasts” – and that’s written by somebody who was present in the room. So you may think your argument is more nuanced, but the sense in the room from at least one headline-writer was what it was, regardless.

  • Tony, actually there are lots of people who say “that movie could have been 30 minutes shorter.” And it’s one reason why screenplays have a standard length and movies beyond that length are much harder to get made. If you read the entire piece I make your point and do so in bold. But I also believe – and research has shown this – that length can be an obstacle. I make that argument in the piece too. Thanks for your note!

  • Well, welcome to the behavior of TV viewers with DVR’s (actually only a minority of them, based on the data, but still). Commercial necessities may require “pre-rolls” (which aren’t always “pre” by the way), but there’s no reason you can’t drop some content before them (as I did in Inside Psycho – and as TV shows do all the time). As for housekeeping or absurd waste-of-time banter, those things are hugely irritating to me and pure explorations in vanity on the part of podcasters.

  • Rob @ podcast411

    In your article you state “Fifth, the Libsyn data is using what appear to be ambiguous labels. For example, does “100,000 downloads” mean 100,000 total downloads of all episodes ever” since you did not actually reach out to me for clarification I thought I would put that link above where it is clarified – and was also mentioned in the session. It is a myth about people wanting shorter content. They don’t – they just want good content – and the length does not mater. To say it is not a myth – is not accurate – as it is based on nothing – no data to support people preferring shorter podcasts exists. Because all of the data shows longer shows are more downloaded on a per episode basis. My article is to stop the stupid and inaccurate myth that shorter is better or preferred – it is not. At least that is what big data says. If you ever have any questions on our data please let me know.

  • Yes, you clarified that 100,000 downloads means overall across all shows. So my points about viewing this from the show perspective rather than the episode perspective still stand. Further, all my points about the difference between historical and potential listenership likewise still stand. Data is not just about what people do but also about what they choose not to do and why. And that data is not contained in your set. Behavioral data is always better than questions about behavior where both exist, but behavioral data does not necessarily explain WHY it says what it does. It does not, for example, explain why only 15% of Americans listened to even one podcast in an average week. The obstacles to more widespread use are my primary concern. And plenty of available data point to length as one obstacle. So that’s no myth. Nor does it mean that every show should be shorter, of course.

  • Rob @ podcast411

    Let me try to explain this simply. I looked at all the episodes released in January – of those episode measured to the end of February that were over 100k downloads – I then looked at the average length of the episodes for those shows. 84% were over 51 minutes. 9.9% were less than 30 minutes. We are looking at the top 1% of shows overall – the ones consumers downloads (voted they like the most). And one thing stands out – having a long show did not hurt these shows. Saying length is a barrier is not supported by data in this set. Because if it was a barrier – then the longer shows would not have won the audience they did and would not have grown. You would have seen more shorter shows in the top group. There is no measured data in the podcast world I am aware of that shows length of episode is a barrier to growth. NONE. There are survey’s of Radio industry and historic Radio data that sorta indicates that – but Podcasting is NOT Radio. To take what people learned in Radio and try to put that square peg into the round hole of Podcasting is wrong and the data is not there to support it. You can spin it however you like – but the data set from podcasting – does not support the myth that length of episode is a barrier to growth or entry. If anything it shows the opposite. Please don’t mistake the Radio world for Podcasting – they are very very different.

  • Well I believe I have finally met the Donald Trump of the podcast world – thin-skinned and patronizing, even when the conclusion of the piece he’s commenting on mirrors his own independent conclusion. He can’t take “I agree” for an answer. Primarily because he is Hell-bent on sustaining the idea of podcasting as a niche rather than recognizing that it is part of the larger tapestry of audio consumption to consumers who care more about what they hear than how it comes to them. The “you just don’t understand podcasting” routine doesn’t work when you’re dealing with people who not only work in the space but create content in it. Meanwhile, he has no idea what research I do, evidently, and is little interested in it if it disagrees with his predisposition.
    The truth is this: No, “radio” and “podcasting” are not the same, but – and this is important – podcasting and radio are audio, and that’s all consumers care about. By his logic, TV is not video and newspapers are not digital. Speaking as someone who has spent his career asking questions of people who listen to audio across platforms, I do not see a “radio world” or a “podcasting world.” I see a world of audio powered by consumers and content. And that world is the only one I care about. Indeed, it’s the only one any content creator or distributor should care about.
    I’m finished with blog-fighting, Rob. If you want to talk about this further, I would encourage you to call me directly: 858-485-6372.

  • I now produce both short and long form content on my podcast. If both helps me have more fun (just by mixing it up), and also the listening, knowing they can scan for the format that meets their need that day. Long form only podcasts are a big commitment for the listener.

  • Cody Gough

    Not a fan of the clickbait headline. “This” indicates that an actual length might be suggested at some point, which you don’t really do. “Stop focusing on the length of your podcast” would actually be accurate.

  • Hi Cody, I think the fact that there is no one answer is a fitting conclusion to the piece as titled. Otherwise I could simply tweet your title and skip the piece. 🙂
    Mark Ramsey

Dive Into The Blog