So What? AI Podcast Growth: Unlock Your Show's Potential - Trust Insights Marketing Analytics Consulting

So What? Marketing Analytics and Insights Live

airs every Thursday at 1 pm EST.

You can watch on YouTube Live. Be sure to subscribe and follow so you never miss an episode!

In this episode of So What? The Trust Insights weekly livestream, you’ll learn how to use AI to achieve substantial podcast growth. Discover how to leverage audience data, uncover hidden patterns, and use those insights to reach more listeners. You’ll gain tactical advice on refining your podcast’s messaging, optimizing your content, and developing data-driven marketing strategies for sustainable podcast growth. Don’t miss out on valuable tips for maximizing your podcast’s reach and impact!

Watch the video here:

So What? AI Podcast Growth: Unlock Your Show's Potential

Watch this video on YouTube

Can’t see anything? Watch it on YouTube here.

In this episode you’ll learn:

How to analyze listening data with AI to understand your audience better than ever
Using AI to personalize your message, optimize ad campaigns, and find the perfect collaborators
We’ll share real world prompts you can use to generate results

Transcript:

What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.

Christopher Penn – 00:29
Hello again, and welcome to So What?, the marketing analytics and insights live show. John and I are holding down the fort this week while Katie is off searching for mystical artifacts in the urban mazes of New York City’s under sewers.

John Wall – 00:44
All right, that sounds— that’s Ghostbusters-level stuff there.

Christopher Penn – 00:47
Absolutely, absolutely. This week we are talking about AI podcast growth: what can you use generative AI tools for to help grow your podcast? We’re going to talk about uncovering audience data and hidden patterns, and then using that data to find more opportunities to market your show.

So John, talk me through how for Marketing over Coffee so far, you’ve done audience analysis to understand who listens to the show.

John Wall – 01:19
Yeah, I’m the perfect use case, the total Luddite example of how to do this. We get— reviewing profiles on LinkedIn, that used to be the best thing for us. That has dried up years ago, but that was part of it. Surveys are a big part of it, just surveying the audience via the email list and getting a feel for different types of— it’s all the classic stuff that the advertiser wants: how many are there, what’s the gender breakdown, how much buying power do they have, what size organizations do they run, B2B versus B2C. And then where in the world are they, too? Libsyn has a bunch of stuff on the backend, too, that’s actually not too bad as far as listener data.

John Wall – 02:01
Like an interesting one for us is about half the audience is US and then another 30% or so fall into ten different countries. So it’s kind of all over the world, but that’s about it. And of course they care about downloads, and they care about leads. Like that’s the saving grace for us, is that if they get 20 leads, they really don’t give a damn about all these made-up metrics they’re really interested in: are they getting some business? So that saves us. We’re not like one of these B2C plays where they need to show all kinds of stuff to convince Johnson & Johnson or somebody like that this is where the money’s supposed to go.

But yeah, I’m thrilled to dig into this because this is like a list of stuff that obviously I’m not doing, and I’m sure there’s going to be stuff that we could do.

Christopher Penn – 02:46
Okay, so let’s just— today we’re going to build a podcast audience profile, similar in concept to an ideal customer profile, but this is specifically for podcasting. So we’re going to need two things: we’re going to need information about the audience and information about the podcast. Do we have on the Marketing over Coffee website a really solid description of the show?

John Wall – 03:15
No, no, I just lost audio for a second, but I’m back.

Christopher Penn – 03:18
Oh, no, I was saying do we have a really solid description of the show? We actually don’t.

John Wall – 03:24
Oh yeah, About Us page, we always say at the intersection of marketing and tech, covering topical issues and article interviews.

Christopher Penn – 03:32
Okay, so step one, if you’re going to be doing this process, you need to have a solid description of your show. Since we don’t, we’re going to have to make one. I’m going to go ahead and move to our shared screen here, and let’s go to Perplexity. Perplexity is one of those AI-based search engines.

I’m going to start off by saying something like, “Say what do you know about the Marketing over Coffee podcast at marketingovercoffee.com? What is the show about?” So, let’s see what Perplexity comes up with, and of course it’s going to go out and digest down a bunch of information. I’m going to open up a text document in another window here just so that we can copy and paste chunks of this. There’s the first part of our answer: it is all this stuff here.

I’m going to get rid of the citations; we don’t need them for the purposes of this. We’re going to go look through this very quickly and say, is any of this factually wrong? We don’t record in a coffee shop anymore, we haven’t recorded in a coffee shop in years.

John Wall – 04:39
Oh no, I say that still stays in. They believe that— the other one is it’s published— it’s actually published Thursday night, so it’s available Friday morning, so maybe it should be Friday morning.

Christopher Penn – 04:52
Okay, so let’s see. The podcast is recorded— it does— well, we’ll leave that, we’ll leave that for now, but it is— we don’t really— it doesn’t say in here for the day, week. Okay, next: what major topics are discussed on Marketing over Coffee?

John Wall – 05:15
Yeah, I have to say, like this is already— the Perplexity is cutting edge, like this is way ahead of anything I’ve seen previously when we tried to do this.

Christopher Penn – 05:25
Okay, so we’ve got a nice summary here of the major topics of the show. And again, we’re going to trim off the citations, we don’t need them for this purpose. So, we’ve got kind of a— essentially a company profile, if you will, for this. This is pretty decent, so we’re going to call this “MOC profile.”

Now, if you don’t have a show or your show is just getting started and it’s not known by search engines, you need to do this manually. So you need to write down: what are the key points about your show, what are the topics you’re going to cover on your show, and things like that.

So, that you have that— if you’ve got a show and it’s reasonably well-indexed from an SEO perspective, then an AI-based engine like Perplexity is going to be able to put together a profile. So that’s part one. Anytime you’re doing any kind of profile work in generative AI, you need your stuff and the audience, and those are the two halves. If you’re missing one, you’re going to have a really hard time.

So next, let’s talk about where we get our audience profile. You mentioned Libsyn, right? So, Libsyn has statistics, we can download these statistics, they come in— they come in PDF form. Well, I mean, you can get them in just about any form. So, let’s try doing— let’s grab all of our stats here, let’s look at the last 90 days, and let’s get it in CSV format.

And that should give us a nice pile of data. And indeed, it does. So we have, by episodes, we have our episode title, the release date, and then our metrics. These are, I would say, the most popular shows in the last 90 days. We have our countries, we have our destinations.

Now, the one thing I’m going to do here, I’m going to get rid of the high-level stuff. I want to— if you load a CSV file into generative AI systems, they have a tendency to try to do data analysis on them, and I don’t want it doing that because there isn’t a lot of data analyzed here in the first place, and I don’t want it to treat it like a data file because then it’s going to invoke Python and do a bunch of stuff that really isn’t helpful here.

Instead, what I want it to do is just treat it like a text file. Now there’s a programmatic way of doing this. If you’ve only got a few files, the easiest way is just copy and paste all manually into a text file. I’m going to do it the not manual way because I like coding, but you are under no obligation to do so. So, we have now this nicely combined text file of these— of this stuff. So let’s go ahead and put that here into our podcast live stream folder.

So that’s our main podcast data. Next, we need to know— we should probably know— about what the audience is interested in. There’s two sources for this that you can get— well, at least two. One of them, like you mentioned, is LinkedIn.

So if your show has a LinkedIn page, a company page, you have analytics for your audience. So, let’s go ahead and look at that. Going to go to Analytics, go to— go to Followers, last 365 days, and let’s see what happens when we hit this export button: it’s our last 365. Let’s see what comes out.

So, it has spit out a spreadsheet.

John Wall – 08:50
This is crazy, this has all been updated again, too, like this is a new dashboard.

Christopher Penn – 08:55
Yep, yeah, the pages stuff has gotten more useful. So from LinkedIn, you now get— let’s see, these are our followers, the locations of our followers for the page, the job functions, the seniority, the industry, and the company size. So, this is pretty decent. So let’s go ahead and again, I want this treated like text, I don’t want this treated like a spreadsheet. So I’m going to do— let’s call this “Marketing over Coffee LinkedIn Page Data,” and what I’m going to do is I’m going to write this in a language format called Markdown. It’s just a type of plain text.

So, we have company size, we have industry— that’s a lot of industries. We have seniority, we have job function— what do people do for work? We have location. We are huge in New York City. Okay. So that’s the audience, demographic data from there.

So we’re going to call this “MOC LinkedIn.” So that’s our LinkedIn data.

Now, the next set of data that we can get is demographic data about audits— what they’re interested in. To do is we’re going to use Google Analytics 4. Again, if your podcast doesn’t have a website, it should, and then you should be using Google Analytics 4. Even though Google Analytics 4 is no one’s favorite analytics software, one of the key reasons to use it is because you do get really good demographics, sampled data.

So let’s go ahead and start an exploration, and let’s add for our dimensions under demographics: age, gender, and interests. And under our metrics, let’s go to “user” and add in “active users.” We’re going to make “active users” our value, and we’re going to put in— let’s put in age and gender first. So, let’s do age. Oops, there’s gender. And this should be a freeform table as the wrong place— that’s interesting.

So, we obviously have the unknowns, but from this time period— I’m going to back this up, let’s back this up. Can I get 12 months on this? I cannot.

John Wall – 11:49
Yeah, so many people get, you know, through podcast apps, the site actually doesn’t get a ton of traffic.

Christopher Penn – 11:55
Well, that’s okay.

John Wall – 11:57
Yeah, I mean, if we get percentages that gives us— Great.

Christopher Penn – 12:00
That’s interesting. So we actually— according to Google Analytics, in the last 92 days, we have more female listeners than male listeners.

John Wall – 12:07
Yeah, no that’s not surprising because we have a lot of female guests. And I think the gender breakdown for a lot of marketing careers is definitely close to, if not leaning to the, female side.

Christopher Penn – 12:19
That’s interesting. We also get— we skew younger and female for audience. The two top two audiences— huh. That’s not what I expected.

Okay, so that’s not bad. Let’s do this: let’s take out age, and let’s put interests. And what we get here is the topical interest. So you have your gender, and you have the topical interests that people are interested in: technophiles, shutterbugs, business professionals, avid investors, entertainment news enthusiasts, social media enthusiasts— let’s go to 100 interests.

So, now we have this really long table. Now this is derived from Google’s DoubleClick advertising network, so this is a lot of information here. We’re going to go ahead and extract this out as a CSV as well. And again, we’re going to find that download, which we can see here. And let’s just give this a short name, “demographic interests by gender” as the profile name. And we’re going to change that file to a Markdown file.

Okay, so now let’s put that in our desktop folder. And now in terms of ingredients, we have the Libsyn stuff by PDF, we have the Libsyn stuff— let’s sort this here. We have our general profile from Perplexity, there is our Libsyn data in text form, there’s our LinkedIn profile data, there’s our Google Analytics data.

So, we now have a lot of ingredients, so now it’s time to start sewing this all together. Let’s go over to— choose the large language model of your choice. The paid version of ChatGPT is good. The paid version of Google’s Gemini is good. The paid version of Anthropic Cloud is good.

I use Google’s Gemini mostly because I like the integration with Google Drive and stuff like that, but by no means do you have to use this particular one. You can use any language model that supports uploads of stuff and has a large context window, large short-term memory.

So, I’m going to move to Gemini Pro. I’m going to turn off all the safeties because I like to live dangerously, as you do. “Podcast Audience Profile.” So we’re going to start off as we always do with our basic RACE framework, which again, if you have not gotten your copies of this, go to get the RACE framework at trustinsights.ai/race. You can get the PAIR framework at trustinsights.ai/pair, and the TRIPS framework at trustinsights.ai/trips.

All right, so we’re going to start off saying: “You are a podcast marketing expert, skilled at podcast growth. You know how to grow podcast audiences, find podcast listeners, and retain podcast audience members. First, what are the general best practices for building a podcast audience profile?” And we’re going to see what it says.

We want to assemble some of these components. We are doing the priming— model priming— step of model building here where we’re going to try and understand who is in our audience. We’ve got some basic stuff here. We’ve got demographics, psychographics, listening habits, and stuff like that. So we’ve got now essentially the model has pre-prompted itself. Great. “Next, I’m going to give you a series of files about the podcast we’re analyzing today. We want to develop a comprehensive picture of the audience.”

“First, what is our first file, and what should we start with?” Let’s start with our LinkedIn file. “Here is the audience profile of the listeners from the show’s LinkedIn page. Analyze it and provide your conclusions in outline format.” I like outline format because outline format— it lets it be more thorough without like, totally foaming at the mouth. So, there’s our LinkedIn file, and we are at 1,700 tokens for that. Let’s see what it comes up with.

Small to mid-sized businesses dominate, industry: marketing and related fields with a business focus, seniority: experienced professionals, job functions and marketing, geography: major US cities, international interest from London, Bengaluru, Addis Ababa, Vancouver, and Toronto. So, that’s LinkedIn, great.

“Next, I’m going to provide you some listening data from our— from the podcast hosting service Libsyn. Analyze it and produce your analysis in outline format.” Now, if you do PDFs, you can load the PDFs; you have the text file, you can load a text file. I’m actually going to load the PDFs because I think that would be illustrative.

All right, so we’ve got quite a bit of information there, mostly on the episodes. Let’s see what Gemini comes up with. Dominant US audience, strong international presence: the UK, Canada, Australia. Global interest. Episode popularity: new episodes consistently garner higher downloads, duh. Early episodes with lower downloads. Apple Podcasts— Spotify is a growing force. Niche podcast users interesting. Did you know were mostly Apple Podcasts?

John Wall – 18:23
Yeah, oh yeah, that’s always dominated.

Christopher Penn – 18:25
Okay, US audience, okay. So we now— we’ve got that, great. “Next, I’m going to provide you with gender and interest data from Google Analytics 4. Analyze it and reduce your analysis in outline format.” Now let’s go ahead and go to Google Drive here and put in our Markdown file. “Analytics, overall audience: predominantly female, top interest: technology, technophiles, banking and avid investors, informed and educated consumers, passion for travel and food. For the male audience, similar but less pronounced, great interest on sports and gaming. Niche interests. Overall conclusion—” Okay, now so we’ve got a bunch of different data sources. Great.

“From all of this audience data about the show, please create a comprehensive podcast audience profile for marketing for the podcast. Incorporate and synthesize the data together for a clear, coherent, single view of the audience. Produce your profile in outline format.”

“Be as thorough and comprehensive as possible, leveraging insights from all the data.”

Let’s see what it comes up with. Core audiences: marketing professionals and small to mid-sized businesses, US-centric with global curiosity, predominantly female, seniority: experienced professionals, values: professional growth, business acumen, staying informed, interests: business and technology, media, entertainment, travel, and food, gender-specific nuances, platform preference: Apple Podcasts, tech-savvy, content preferences: recency bias, guest and topic appeal: drawn to well-known guests, timely topics, and practical tips for applications, marketing implications: content strategy, platform strategy, marketing channels.

So, this is a pretty solid profile of who this audience is. So, let’s go ahead and now take this. We’re going to copy the Markdown of it. Go back to our text editor. So this is now really our major audience. I’m going to trim off those next steps because for an audience profile, we don’t want it making recommendations of the profile, we just want to have the audience itself.

That’s a pretty solid understanding of who Marketing over Coffee is when it comes to our show.

Now, great. “Next, here is the existing description of our show. Using our comprehensive audience profile, rewrite the description to be as persuasive and compelling to our target audience as possible.” So with any kind of audience profile, the next thing you can do and should do is inspect all of your existing collateral that you’ve got for your show and say, “How would you rearrange the content so that it appeals to the core audience,” assuming this is the core audience that you want.

And again, if your show is new, you’re going to provide a lot of this data manually of who you want your audience to be. And equally true, if your audience is not who you want to be right now, you’d want to change that profile to say, “Our aspirational audience is Fortune 50 CMOs,” or whatever.

So we’ve got this very nice description now of the show— here’s how we’ve improved the description, and it has a stronger hook and things like that. So, this is something that we could take and immediately put to use on the website. So, let’s go ahead and just take a look at this. I’m going to trim off the Gemini portion. “Tired of marketing fluff and empty promises? Us, too. That’s why we have this.” Join our— this— here’s what you can expect in every show— that’s not bad.

John Wall – 23:07
Yeah, that’s funny. I like how it’s— I mean, it’s definitely hype, right? This is hype copy. I like the— you’ve got all the keywords in there. You’ve got “SEO, no BS style that respects the schedule”— that’s definitely dead on. Yeah, that’s tight. That can definitely plug into the About Us page.

Christopher Penn – 23:27
Exactly. So we’ve got— we’ve met now— managed to reinvent the show itself, at least from our audience perspective. Now, if you recall, gosh, I forget when it was, it was probably about a month ago now, you and Katie did an episode of the livestream, which you can find on our YouTube channel, on your podcast media kit, right?

John Wall – 23:53
Yeah, right, right, for all the promotion stuff. Yeah.

Christopher Penn – 23:56
So, now this audience profile you could use to validate what’s in the media kit to say, like, is our media kit accurate based on this profile? And of course, you can take all the raw material data and put that in the media kit, as well. But the next things you can do is use this profile to do things like edit your podcast descriptions.

So, for any individual episode— let’s go to the website, let’s go to a recent episode, let’s take the Cassie Breunig episode. All right, what do we got here? So, we have— this is the basic show notes of it. Do you have the transcript for this episode?

John Wall – 24:45
No, no. This just went live, so I haven’t added on even— yeah, transcripts are still not part of the workflow. I’ve not added that in.

Christopher Penn – 24:54
Oh, okay. I feel like we should.

John Wall – 24:57
Yeah, for SEO definitely. It should go in there, it’s just— again, that’s not something I’ve made time for.

Christopher Penn – 25:04
Got it. We covered that on the last episode in terms of how to use AI tools to produce those transcripts, but one of the things you’d want to do is then take that transcript and put it against the profile and say, “Now, write an intro paragraph for this episode that would be compelling to our audience profile and make them want to listen to this specific episode.”

So, it’s this: this audience profile is everything that you’d want to apply to your creative, to the copy, to show descriptions, maybe even the— if there’s a cover image for each show— evaluate this cover. In fact, let’s take a look at this. So, we have some funny cover up, but any of these things, those images could be taken and say, “Okay, is this image as appealing as it could be to our target audience?”

Okay, next, knowing this information, let’s say, “Great, what do you know about targeting ads in Google Ads?” Let’s use Google Ads as the example.

John Wall – 26:28
Yeah, another one to run to. I would be interested in what it says about YouTube because that didn’t come up as far as platforms. I mean, we’ve kind of— I’ve always known that we haven’t done enough work on that front and would be interested in there.

Christopher Penn – 26:43
Okay, so here’s the Google Ads targeting, here’s what it basically knows about. Place of targeting, YouTube Music, Google Podcasts, and stuff like that, great. “Build me a Google Ads tactical roadmap for running ads for our show, using our podcast audience profile and mapped against the overall plan for Google Ads you outlined above. Include specifics such as what demographics and interests to target. Build the plan in outline format.”

Unsurprisingly, Gemini knows an awful lot about Google stuff. Huge surprise, right? But what this gives us is: this gives you kind of a shortcut for how to do your ad targeting based on the audience profile. Campaign structure, YouTube Music, Google Podcasts, niche targeting outside of retargeting, ad creative audio snippets, visuals, copywriting, et cetera.

So, if you wanted to take a Google— you know, take some budget to promote your show— I’m going to turn off the spanner here because it understands the structure of a Google Ads campaign really well— this will save you the brain work of, like, what should I be doing for my campaign, how do I structure it, what do I target? So, just follow the instructions.

Any ad system that provides documentation, ad specs— you can load that into a tool like Gemini and say, “Okay, here’s— you know, here’s Meta’s Instagram ad campaign details. Build me a campaign to promote my show on Instagram Stories or Instagram Reels.” For podcasters— particularly if you’re— you know, you don’t have the backing of a major corporation, it’s a lot of time to get up to speed on this stuff. This is a good way to at least get a starting template and maybe throw— you know, $10 a day at it to see if it will work for you.

John Wall – 28:50
Yeah, that’s fantastic. This is the kind of thing that— again, it’s getting involved in— getting it up and running is such a pain. And so, to have all this stuff ready to go, just cut and paste, is huge.

Christopher Penn – 29:01
One of the big challenges that people have said with marketing your podcast— and I think this is a generally true statement— is that it’s kind of overwhelming, right? In terms of all the things you can do. In fact, if you recall, we have in the past talked about how to market your podcast and come up with this massively huge, gigantic map of, “Here’s everything you could possibly do to market your podcast,” right? And this goes on and on and on and on. By the way, if you would like a copy of this ridiculously absurd map, you can get it for free. We’re going to put it in our Slack group. Go to trustinsights.ai/analyticsformarketers. If you’re not a member, join, totally free, doesn’t cost you any money and this will be in the group from today’s livestream.

You have to join sooner rather than later, posts vanish after 90 days because we don’t pay for Slack for them. So, it is a time-limited offer, but one of the things that you can do— huge surprise— is export this map as Markdown, which means that Google, Gemini, and any language model can read it. Say, “Great, here’s an overall podcast marketing strategic map based on our audience profile.” Let’s see, what is my section here?

“Tactics and Methods: Review the Tactics and Methods section and help me prioritize which two to three tactics I should focus on that will reach our audience. Since there are so many tactics to choose from overall, based on your knowledge of our audience from the profile, which tactics should we start with?” And now we load up our Markdown file and we run it.

“LinkedIn Ads for Professionals: Why?—It’s a strong fit. Your core audience consists of marketing professionals heavily skewed towards those in small to mid-size businesses. LinkedIn is the platform for reaching this demographic, audience with laser focus. Use your targeting—” Here’s your ad formats.

“Email Marketing: Why It’s Essential— Building an email provides a direct communication line of communications for your most engaged listeners,” and so on and so forth.

“Events, specifically virtual events and webinars: Why Strategic?—Your audience values professional development, connecting with industry peers. Virtual events provide a scalable way to deliver on those needs: webinars, virtual summits, live Q&A sessions.”

What do you think, John?

John Wall – 32:12
Yeah, that’s interesting. Well, that definitely tracks, right? We’ve always known that LinkedIn— we’ve been partners with LinkedIn forever, so that’s dead on. And the email list has been— you know, at the core of our successful promotion stuff. But, events, that’s funny that’s up there because that has never entered my mind. I never thought of actually doing, like, a webinar or some kind of live Q&A thing, and that does kind of make sense. That’s interesting. That’s something that I’ll have to kick around a little bit; maybe that’s something worth exploring.

Christopher Penn – 32:41
And bear in mind, this is out of— let’s see how large is our “How to Market Your Pocket—”

John Wall – 32:47
Oh yeah, like there’s well over 64, 65 tactics on that list.

Christopher Penn – 32:51
Oh yeah, there’s— yeah, there’s quite a bit. In fact, there are 170— yeah, different tactics and stuff. So those are the three that it says from for promoting your show, for getting that podcast growth, these would be the three for our show.

Now, the one thing I would caution folks watching is: you have to do this for your show. This is the answer for Marketing over Coffee based on our audience profile that, you know, we’ve just spent, what, 30 minutes constructing. Do not just copy and paste this for your show because if your show is not Marketing over Coffee, this is not going to work. In fact, I would bet you if— went through this exercise, and I just know I’m going to get a message from Katie the moment this show finishes airing, saying, “Now do this for the Trust Insights podcast.”

John Wall – 33:38
Right, yeah.

Christopher Penn – 33:40
Right? You’ve got to do this for your show. So again, remember the podcast strategic map is going to be in the Slack group. You’ll want to download that PDF and feed it into a language model to make use of it, to say, “You know, here’s what I know about my show, the show that either I have or I want to have.”

Maybe you might even say, like, “Here’s how much money I can afford to spend promoting my show.” In fact, let’s do that. Say, “Great, I also have a total of $100 per month to spend on advertising my show. Given the audience profile and given the strategic map, where should I spend this money?” Bear in mind, it doesn’t have to be on ads, it can be anything that is a good investment. Let’s see what comes up with—

John Wall – 34:42
It—

Christopher Penn – 34:42
—Could— it could come up with tchotchkes.

John Wall – 34:45
Yeah, right, free stuff.

Christopher Penn – 34:50
“Optimizing Your $100 Monthly Budget: LinkedIn Ads— allocate $50 to $70. Email Marketing Growth— allocate $30 to $50.” So you’re 80 at the low end there. “So immediate listener growth, long-term audience building, additional budget-friendly options is applicable: podcast guesting and content repurposing.”

So it’s saying LinkedIn ads and email marketing growth. I think the email marketing growth you’re going to spend more than $30 a month on.

John Wall – 35:17
Yeah, well that was the one that got me, like where should that money be going? So, it’s saying, create more content— that’s basically design costs, then, to create a PDF or whatever. “Lead magnet website and social—” Yeah, “Lead Magnet,” dude, you hit it, that’s tchotchkes. Lead magnet is tchotchkes.

Christopher Penn – 35:36
So, how would we build a lead magnet?

John Wall – 35:40
Yeah, give away something for free. I mean, that’s an interesting idea because the idea of, like, say we took transcripts from five popular episodes and made that a PDF— you know, it’d be a paper thing— but that’s an interesting idea, crossing over to a different media from audio to see if that attracts attention.

Christopher Penn – 36:01
So, let’s talk about that because I think you hit it exactly on the head in terms of what you’d want to do. How would you go about doing that?

John Wall – 36:15
Yeah, that’s it. The big thing would be grab all the guest interviews because, for the show format, for people don’t know, we usually alternate every week, you and I do topical stuff and cover what’s going on in the industry. And then, on alternating weeks, it’s usually marketing professionals or authors. So, to do a quarterly summary of the author interviews would be— And then I guess, yeah, I didn’t think about it too, actually: quarterly summary of the stuff we do is even more topical. And we’ve actually did that in the past. I did do a Marketing over Coffee quarterly report where we grabbed everything— you know, that you and I had covered that was topical and put that into an eDoc— but that’s just, like, an insane amount of work. That’s definitely $100 a month is not going to cover that.

But yeah, now with— you know, gen AI to eat the transcripts, I mean, to have it come up with a summary of, “Okay, give us the— you know, five guest interviews over Q1 2024 and give us the— you know, five most important points from each of those.” That’s an interesting take.

Christopher Penn – 37:16
I don’t know, you would even— I think you could make decent amounts of stuff just from a single episode. I mean, think about it: so a single episode interviews, what, about 30 minutes?

John Wall – 37:27
Yeah, 25 to 30, yeah.

Christopher Penn – 37:29
Okay, so if you take 30 minutes, people speak at an average of, what, 150 words per minute? So that’s 4,500 words for a single episode. If you were to take that episode— do we have any transcripts laying around? I don’t remember if we do or not.

John Wall – 37:47
Yeah, go grab, like, one of the Seth Godin, Simon Sinek interviews, those have transcripts.

Christopher Penn – 37:52
No, I do have one. I have this one with Marketing over Coffee 803; 803 was with Robert Rose. Okay, so let’s hit “Save” on this podcast audience profile, and let’s talk about building something with this Robert Rose transcript.

I’m going to switch over to Pro in Gemini, I’m going to turn off all the safeties. “You are a podcast marketing expert with a specialty in content marketing and building long-form content, such as ebooks. First, examine this transcript from the Marketing over Coffee podcast and provide an outline of the show.” So, let’s see what we can do with just a transcript.

I remember the show because one of the folks in the Slack group said, “I would love to share the show, but my staff is all Brazilian and doesn’t speak English,” so I had to download this transcript initially so that we could translate into Portuguese. Okay, so “Introduction, content marketing deep dive, three pillars of content marketing: balance, strategy, operations, and measurement, six steps of the content lifecycle, flipping the script on content creation, storytelling and brand, the future of—” this actually would also be decent show notes.

John Wall – 39:21
Yeah, right? This is another way to run the show notes.

Christopher Penn – 39:24
Okay, great. “Let’s now take all the topics discussed in the show and rearrange them in a logical order for a marketer interested in building a content marketing strategy. Arrange them in a logical flow from the start of a marketer’s project to the completion of the project.”

So, we’re going to take that same content now and have generative AI rearrange it. “Define your ‘why,’ establish a content strategy, prioritize content ideas, flip the content creation script, build your content operations, pressure-test your brand story, integrate AI strategically, measure, analyze, and adapt.” I don’t know about you, but that actually sounds like Robert’s most recent book.

John Wall – 40:24
Yeah, that’s covering most of the points.

Christopher Penn – 40:27
Great. “Now, for each section, build a list of actionable items and takeaways/practices a marketer can do immediately.” Define your ‘why’ action, takeaway, so on and so forth. “Set SMART goals.” Okay, great. “From Section 1, build out the start of a chapter or a book based on the interview, actions, and takeaways. Write as though it were a tutorial, recipe, or lesson.” And we’ll put in “Chapter 1.”

So, what generative AI is doing here is digesting down that stuff in the podcast, adding inference, adding its own essentially— you know, background knowledge— and then building out content.

So, you could take five episodes on content marketing, five episodes on advertising, five episodes on this— merge and synthesize the knowledge of the guests together, and build out a full-size book, chapter by chapter, from all of that background knowledge, crediting obviously all the interviews, and maybe even providing the interview— the cleaned-up interview transcripts as part of the book in the appendix, so people can go and read the original things, and of course, linking to the original authors’ stuff.

But podcasts are such a huge repository of knowledge, they are so rich because you get very often— obviously, you get people who are promoting their books and their works or whatever, but you get extra knowledge often that isn’t in the book things.

And if you were to now merge down this stuff topically, you could say, “Okay, well, here’s— have Gemini or any language model write a book derived from that source knowledge and have it be coherent.” So, it says, “Dive deep in your customer’s world. For each persona, explore pain points, goals, information sources, and content,” right? That’s pretty straightforward.

And now you can say, “Great, write eight paragraphs on action item two.” Now you have a book, or several books with 800-some-odd episodes. We have a huge repository.

You could take, for example, all of the Seth Godin interviews over the years, and make the annotated Seth Godin—

Christopher Penn – 43:28
—So if we go back, and we think back to our podcast marketing in terms of long-form media: books, transcripts, action guides— you could turn again any of that content from those— all the episodes from Seth Godin or David Meerman Scott, whatever, into action guides, into cards, into things that people would want to have, because it’s super-valuable information and a good chunk of it is timeless, especially folks like Seth Godin. He pretty much writes mostly evergreen content.

John Wall – 43:59
Yeah, right. Like you said, and it’s interesting to have that view across all of the works— you know, because things have evolved and changed over time— to be able to put that all together under a single topic. That’s really interesting.

Christopher Penn – 44:11
So to recap what we’ve done today— and we’ve done a lot— we talked through how you build your show profile from existing data, particularly if you don’t have a great description on your website. We’re going to be doing that. So, that’s feeding the data from Perplexity or another AI search engine into a tool to die— or getting out of Perplexity. We talked through how do you extract data from LinkedIn, from your podcast downloads, from Google Analytics, to build that podcast audience profile, and then how do you run that profile digested down to something coherent, so that you can talk to it constantly and say, “Here’s this upcoming guest, how should I promote their stuff on our show? How do I write the email about it? How do I make the email compelling? How do I write the show description? How do I write the ads promoting this upcoming guest— this upcoming show?”

And then, from there, turning it into— you know, revising your website copy, revising your show notes, and then thinking about the things, getting your marketing strategies. So, we talked through how to— how you build marketing strategy with a pod— from a podcast or for a podcast, from— let’s can this here.

What different things should you be looking at? Here we are, this is one, so we look through building Google Ads plans, building email marketing plans, identifying and prioritizing tactics, building tactics from in a specific channel, all this using generative AI.

Ultimately, we want to get to that point where podcast growth is— you still need to steer it as a human being as the podcaster, boy, these tools can get— help you get a lot more work done on promoting it.

Any final thoughts, John?

John Wall – 46:01
Yeah, well I think the— you know, the big one is being able to carve that list down. Like you said, there’s over 160 things that you can do, but you need to figure out where to start. And so, getting some direction on those three is a huge advantage for you in saving time and making— you know, what little time or money you’ve got be able to put it to work for maximum impact.

Christopher Penn – 46:21
Exactly, so that’s what traction, right?

John Wall – 46:24
Yeah, yeah. Traction is— the traction model is similar, with— Yeah, so much of the work is: you have no data on which of the programs work, so anything that gives you any kind of edge into how you whittle that list of 160 down to three is— that’s what makes all the difference.

Christopher Penn – 46:42
All right. So I think that’s going to do it for this week’s show. Next week I believe we’re going to be talking— what are we talking about on, because this is part two of— next week we’re talking about— we talk about the future of AI podcasting, and the way that AI technologies are going to continue to evolve and how you might be able to use some of these technologies, and also a whole bunch of “what not to do.” So stay tuned for next week’s show.

Thanks for— thanks for being with us on the live stream, and we’ll talk to you soon.

Thanks for watching today! Be sure to subscribe to our show wherever you’re watching it. For more resources and to learn more, check out the Trust Insights podcast at trustinsights.ai/tipodcast and a weekly email newsletter at trustinsights.ai/newsletter.

Got questions about what you saw in today’s episode? Join our free Analytics for Marketers Slack group at trustinsights.ai/analyticsformarketers. See you next time!

Need help with your marketing AI and analytics?

So What? AI Podcast Growth: Unlock Your Show’s Potential

So What? Marketing Analytics and Insights Live

airs every Thursday at 1 pm EST.

In this episode you’ll learn:

Transcript:

Leave a Reply Cancel reply

Subscribe to our Weekly Newsletter

Pin It on Pinterest