The Biggest Challenge For Podcasters Today Is…
In which this author gets in a pointless argument with an AI robot about podcasting.
Welcome to Dispatch #38 of The Audio Insurgent. There are 100 new people from last dispatch two weeks ago. Welcome!
You get a TWO-FER this week: Today’s newsletter is focused on podcasting. Then, tomorrow, you’ll receive another with the same question focused on radio and public radio.
What you are reading was written last Friday. It is an extremely quiet morning in my house. It’s barely light out. My wife is traveling. My son just left for school. The dogs are passed out on the office floor. My business partner is very busy managing more than 60 costumes for her daughter’s school play. No one is expecting much from me until later today.
So, I thought I’d write something. First idea in my head: “I’m going to write a dispatch about how every successful podcast solves a problem for its listeners.” Then, after I already started writing it, I got a weird deja vu feeling and searched my archive. Turns out I already wrote a dispatch on it… two years ago in Dispatch #4. You should read it if you haven’t, it’s pretty useful.
People regularly ask why I write The Audio Insurgent. My answer is often that what I write about is either (a) something I’ve brought up in multiple recent conversations, or (b) ideas I’m trying to work out in my head.
My preferred method of working out ideas is to argue with people. I not-so-secretly love to argue (in a constructive way). It has generally fallen out of favor today but if you look at my best friends, they all love to debate… anything. I’ll even regularly argue a point-of-view that’s different from what I believe, just to for the fun of seeing it from a different angle.
As I mentioned, none of my usual debate partners are available this morning, so I thought I’d have this argument with ChatGPT.
Generative AI! It is the “metaverse” of 2023! It seems that no one can stop talking about it. There is an endless stream of writing about it. So I thought I’d ask it for its opinion on the challenges podcasting faces today. And then I will argue with it.
Oh, and at the bottom, a few pics from the London meetup last week, which was super fun.
[TODAY’S FOLLOW-UP THING: THE IMPORTANCE OF UNDERSTANDING THE CUSTOMER EXPERIENCE]
Before we start, I want to flash back to the last dispatch of The Audio Insurgent, where we talked about lessons we can take from Spotify.
One point I raised that I expected to get some pushback on was “Senior leadership needs to understand spoken word audio.” And, to my surprise, and despite lots of conversation about that dispatch, not a single person pushed back on that idea.
Just after I wrote that dispatch, I ran across an Axios post on the turn-around of Barnes & Noble. It is a fascinating (and quick) read. If I was to summarize: Barnes and Noble has been on a tear over the past few years, bucking a lot of predictions for the fate of book retail, as well as for their company specifically. Their secret? Two things. First, be successful at being a bookstore by actually focusing on books. Radical, I know. But if you have been in a Barnes & Nobles store in the years before the turn-around, you could be forgiven for thinking you’d accidentally walked into Target instead. The second key to their turn-around: putting book lovers in charge of their stores and giving them a lot of autonomy over everything from who works there to inventory to store layout. There is a great quote towards the end from retail analyst Neil Saunders: “Shops succeed when people running them understand their customers.”
Imagine if podcast companies were run the same way.
[TODAY’S MAIN THING: THE BIGGEST CHALLENGE FOR PODCASTERS TODAY IS…]
AI! AI! AI! AI!
(I’m talking about Artificial Intelligence, btw, not shouting out to someone who uses the short form of “Alfred.”)
I am so sick of people talking and writing about AI. So I took that as a cue that perhaps I should write about AI too.
Yesterday someone asked if I thought AI was a threat for journalism, writing, or podcasting.
Short answer: No.
Not-short answer: It is a clear threat to crappy journalism and crappy writing… and possibly crappy podcasting. While it does have some utility for creators to help with research and tasks like grammar/copy editing. I think it has a long way to go to be a threat for anything innovative and creative.
Those are words I might eat someday.
The reason I don’t get too breathless about AI is right out in the open. The makers of AI engines like ChatGPT tell you up front what’s happening behind the curtain: ChapGPT learns by basically reading everything on the Internet.
Think about that.
What do you find on the Internet? A lot of great stuff, sure. But for every great thing, there is a pile of things that aren’t great, if not bad or untrue or sloppy or lacking context or wrong or just too simple for the question at hand.
Want to know the five best Westerns filmed outside the United States? Use ChatGPT to look into that. Need an idea…well. Get some inspiration by focusing on problems you can solve (a reference back to Dispatch #4—again!--problem solving is a powerful framework for podcasting!).
(And if you are curious about Westerns, here is ChatGPT’s instant verbatim response: “(1) The Good, the Bad and the Ugly (1966): Directed by Sergio Leone and filmed in Spain, this film is considered a classic Western and is known for its iconic score by Ennio Morricone. (2) Once Upon a Time in the West (1968): Also directed by Sergio Leone and filmed in Spain, this epic Western stars Henry Fonda, Charles Bronson, and Claudia Cardinale and is often regarded as one of the greatest Westerns of all time. (3) A Fistful of Dollars (1964): Directed by Sergio Leone and filmed in Spain, this film stars Clint Eastwood in his breakthrough role as the "Man with No Name." (4) The Magnificent Seven (1960): Directed by John Sturges and filmed in Mexico, this film is a classic Western remake of Akira Kurosawa's Seven Samurai and stars an ensemble cast that includes Yul Brynner, Steve McQueen, and Charles Bronson. And (5) Unforgiven (1992): Directed by Clint Eastwood and filmed in Canada, this film stars Eastwood as a retired gunslinger who takes one last job. It won four Academy Awards, including Best Picture and Best Director.”)
So…back to podcasting.
There are lots of times where humanity’s collective wisdom is…useless. But there are times that collective wisdom can be helpful. When we ask ChatGPT a question like “The Biggest Challenge For Podcasters Today Is…” its response isn’t magic or what it thinks (because it doesn’t think), or what its programmers think. Because it draws from what is written online, what it is doing is telling us back what WE think.
So what are the answers? What do we think?
According to ChatGPT, the biggest challenges for podcasters today are monetization strategies and audience growth.
No duh, right? That’s neither surprising nor all that revealing.
But what I found fascinating are the bullet points it listed under each of those categories.
Audience Growth and Retention
High competition in the marketEver-changing listener preferencesContent saturation
Monetization Strategies
Exploring various revenue streamsBalancing advertisements with contentCreating valuable partnerships
So let’s start with Audience Growth and Retention. We tell ourselves these things all the time…and they aren’t true. Sure, depending on whose numbers you quote, let’s say there is an average of 1,000 new podcasts every day. I often start off workshops with that number…and it deflates people. It feels like an impossible mountain to climb–and if you are making something unexceptional, heard-it-before, boring, or run-of-the-mill–then yes, it is a huge mountain to climb.
So I asked ChatGPT, “Isn’t the prescription for that to specialize and be unique?”
It answered: “Yes, one strategy to gain and retain listeners in a crowded podcasting market is to specialize and be unique. By focusing on a particular niche or topic and offering a unique perspective or approach, podcasters can differentiate themselves from the competition and attract a dedicated audience. For example, a podcast about general news and current events may struggle to stand out in a crowded market, but a podcast that focuses specifically on technology news or political analysis may be able to carve out a unique space and attract a dedicated following.”
I was nodding my head until the last bit.
I replied: “But even making a podcast about ‘technology news’ isn't specific enough. There are literally thousands of them. Isn’t the key focusing on a very specific niche. Instead of ‘technology news,’ focus on ‘news about the Nintendo 2DS’ or ‘news about software development in Malaysia.’”
It said back: “Yes, you're right. By narrowing the focus to a specific niche, a podcast can become a go-to source for in-depth coverage of a particular topic or area of interest, building a loyal following among listeners who are passionate about that subject. This approach can also make it easier to market the podcast to a targeted audience and attract advertising or sponsorships from companies interested in that particular niche. Of course, it's important to ensure that the niche is specific enough to be truly unique, but not so narrow that there aren't enough potential listeners to support the podcast.”
Then I switched to talking about monetization.
I wrote: “I’m not buying what you are selling about monetization. Podcasting’s problem isn’t monetization schemes, it’s the industry's over dependence on advertising.”
It replied: “Yes, it is true that podcasters are often heavily reliant on advertising as their main source of revenue, and that they have been slower to develop other revenue sources compared to other digital media. While advertising can be a lucrative revenue stream for successful podcasts, relying solely on advertising can be risky, as it can be impacted by factors such as changes in the advertising market. To address this challenge, podcasters are starting to explore other revenue streams, such as live events, merchandise sales, direct listener support, and subscription-based models.”
I replied: “But isn’t that true of everything? Any revenue source can have good periods and dry periods. Isn’t the point that what’s most important is to invest in a diversity of revenue sources to support podcasting?”
It’s answer: “Yes, you are absolutely right.”
It is hard to argue with someone who agrees with you.
It continued: “In addition to providing financial stability, diversifying revenue sources can also help to create a stronger connection with listeners and build a more engaged audience. For example, offering merchandise or subscription-based models can provide a sense of community and exclusivity for listeners, while live events can provide an opportunity for podcasters to meet and interact with their audience in person.”
Me: “So it all comes back to having a healthy relationship with a growing audience, doesn't it?”
It replied: “By building a strong relationship with listeners, a podcast can create a loyal following that is more likely to recommend the podcast to others, provide feedback and suggestions, and support the podcast through direct listener support, merchandise sales, or attendance at live events. This, in turn, can help to ensure the long-term success and sustainability of the podcast.”
ChatpGPT agrees! It is all about audience.
My biggest concern, and the reason I wrote this today, is the “collective wisdom” we hear back from ChatGPT is a reflection of us–the “truths” we tell ourselves and each other. And like it is in many cases, the “truths” we tell ourselves and each other…aren’t all that true. Often truth is in the subtext and detail–and the truisms get repeated and oversimplified. Do I think monetization is a problem for podcasting?
No, actually, even now, I don’t. I think the problem is our ambition to grow new sources of revenue. That’s different enough to be important.
Do I think audience growth is a problem? Actually, the problem I see is a lack of centering on the audience. We worry about what companies and creators want, but are slow to realize that those who win and thrive are the ones who think about the audience first.
Right as I ended the chat session, I was surprised how much I was agreeing with what I was hearing from my robot friend, despite trying to pick a fight with them.
I wrote: “Gee, this sounds like I’m talking to myself about this stuff.”
It responded: “You're correct, Eric Nuzum has talked about this topic extensively.”
Well then.
Remember, you’ll get another dispatch tomorrow morning with my chat (and thoughts) on what ChatGPT has to say about public radio and radio generally. The results were…more provocative.
[TODAY’S FINAL THING: LONDON MEET-UP]
The London meet up for readers of these dispatches was something new–a bit of a risk and an experiment–and it worked out great and was total fun.
Thanks to those who came out to The Star and Garter in London on a cold, rainy Friday evening. We had people from a number of different places: BBC, Listen, The Telegraph, Audible, Message Heard, Apple–plus a number of independent and aspiring creators too. I forgot to really count, but we had around 20 people, which made my wallet grateful (pints are now £6 each!), but also the right size so that I could have a chat with almost everyone and people could actually meet others. The original idea was that my old Audible UK crew would come for a while beforehand and then we’d see if others showed up for the meet up. All but one of the Audible folks were almost an hour late (not surprising), so the two events became kind of a mash-up.
On a trip with a lot of fun things happening, this was the best part for me. I got to spend time with people I adore, and meet a bunch of people who all share a love for making audio.
And no one got COVID!
It was a bit of a risk, as everyone is busy and in-person events are still a bit of a wonky thing today, but I’ve never been afraid to take a risk that will publicly blow up in my face. This, thankfully, wasn’t one of them. As one attendee wrote to me afterwards, “So great that you organised such a nice little night out. I feel like that never happens anymore. It was a treat!”
Perhaps we’ll try something like that again soon somewhere else.
Okay, that’s it for today. See you tomorrow with the same question, same AI robot, but about radio and public radio.
If this was forwarded to you or you read this online, would you mind subscribing?
And while this is free, you are also always welcome to buy my book or (even better) buy me a beer.
Make great things. I’ll be listening.
--Eric