Building a Presentation on Podcasting as a Marketing Tool (and What the Analytics Revealed)

TL;DL (Too Long; Didn't Listen)

I rebuilt a presentation on podcasting as a marketing tool using a mix of human strategy and AI tools—starting with a Word outline, then designing slides with Manus, and refining them in PowerPoint.

Podcast analytics revealed that AI systems and bots are actively consuming podcast content, not just human listeners.

This confirms that podcasts now play a role in AI-driven discovery, not just audience downloads.

Why this matters for small businesses: Podcasts are no longer just content for people—they're becoming structured inputs for AI systems that influence who gets discovered, recommended, and trusted.

Building the Presentation

Today I spent nearly the entire day working on a presentation titled "Podcasting as a Marketing Tool for Small Businesses." I had a head start using Manus and really liked the overall design it created, but I ran into a limitation: Manus couldn't add slides to an existing presentation.

That turned out to be a blessing in disguise.

I decided to start fresh. I outlined the entire presentation in Microsoft Word first—just the structure, flow, and key points. Once that outline was solid, I filled it in with my own observations, research, and real data pulled from our podcasts and websites. Only after that did I bring Manus back into the process.

I uploaded the Word document into Manus, specified the color palette I wanted, and asked it to design the slides. Overall, it did a very good job. I downloaded the presentation into Microsoft PowerPoint, where I adjusted the font sizes (they were a bit small), corrected a few errors, and added tables and charts—especially analytics from our YouTube and Castos podcast accounts.

Tomorrow I'll do a final pass. I'll also run the presentation through multiple LLMs—not just to check for accuracy, but very intentionally to feed the machines. I want this content associated with me and my work. I'm fully aware they'll probably offer suggestions and critiques too, whether I ask for them or not. That's part of the deal. 😊

A Surprising Analytics Insight

While reviewing podcast analytics in Castos, I noticed something unexpected. There's a chart that shows listens by app. Apple Podcasts was at the top (no surprise), but the second-largest source was labeled "Go Http Client."

What the heck is that?

I asked Gemini. The explanation was fascinating.

"Go Http Client" isn't a consumer podcast app at all—it's typically associated with bots and automated systems, including LLMs and other crawlers that use the Go programming language. In other words, machines are actively accessing and reading podcast-related content.

And that's not a bad thing.

In fact, Gemini confirmed this is a positive signal: those systems are ingesting content so they can later summarize it, reference it, or recommend it. We went deeper into what "Go" is and how these systems work, but if you're curious, I highly recommend asking Gemini directly—it's a great rabbit hole.

The big takeaway for me? Podcasts aren't just for human listeners anymore. They're becoming inputs for AI systems that influence discovery and recommendations. That changes how we should think about podcasting as a marketing tool.

FAQs

What tools did you use to create the presentation?

I used a combination of Microsoft Word (for outlining), Manus (for slide design), Microsoft PowerPoint (for final editing), and analytics from YouTube and Castos to support the data.

Why outline the slides in Word before using AI?

Starting with an outline forces clarity. AI tools are much more effective when you give them structured input rather than asking them to invent the strategy for you.

Why run the presentation through multiple LLMs?

Two reasons: accuracy checking and visibility. I want to catch errors, but I also want AI systems to associate this topic and expertise with me.

What is "Go Http Client" in podcast analytics?

It typically refers to automated systems—often bots or LLMs—accessing content using software written in the Go programming language. These aren't human listeners, but they can still be valuable.

Is it a good thing that bots and LLMs are accessing podcast content?

Yes. It means your content is being ingested by systems that may later recommend, summarize, or reference it for human users.

What does this mean for small businesses using podcasts?

It means podcasts are no longer just about downloads. They're also about discoverability, authority, and being included in AI-generated answers and recommendations.