Your Podcast Consultant: Podcast Tips To Avoid Podcasting Mistakes

If you sent out a survey and got zero replies, that's actually an answer — it means you aren't connecting with your listeners or you asked the wrong thing. I know that stings, but hey, congrats, you found the sound of crickets. We need to sell the survey better, tell people how their answers will make the show better, and ask specific questions instead of "give feedback." Be persistent, offer a small prize, shout out anyone who helps, or take voicemails if forms feel stiff. I even tell a yellow-roses story to show that nice gestures alone won't cut it. If you want, I can help — we can fix this together. Takeaways: I put out a survey and got no answers, and that told me we are not connecting with listeners. We learned silence can be useful: it shows either a weak connection or bad positioning of the ask. I tell podcasters to sell the survey like an ad, explain benefits clearly, and ask more than once. I used the yellow roses story to show doing what you think is romantic can miss the point entirely. I advise asking specific questions, giving shout outs to responders, and using voicemail if forms flop hard. Links referenced in this episode: podcastconsultant.com schoolofpodcasting.com This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy

Do You Know of a Tool? I was scrolling Reddit the other day when I saw someone ask if there was a tool they could use to write their podcast script. I couldn't help myself. "Yes," I said. "Your brain." Maybe that was a little snarky. Okay, it was a lot snarky. But I stand by it. Here's what I'm seeing more and more: people are "AIing" themselves right out of their own show. It usually goes like this. Someone finds a YouTube video that worked well for another creator. Instead of watching it and asking, "how can I do this in my own way and still sound like me," they grab the transcript, feed it to their favorite AI tool, and say "write this in my voice." And on YouTube, where the algorithm rewards volume over connection, it kind of works. Then they ask AI to whip up a thumbnail that looks suspiciously like the original, because on YouTube it really is all about the thumbnail. The problem is, the person doing this often doesn't fully understand what they're now saying on camera or on mic. Getting My Electronics Degree That reminds me of my lab partner Gary from college. My first degree, before I got into teaching, was in electronic engineering. I'm pretty sure I only got that associate's degree because of Gary. He was brilliant. I understood the math fine, but the electronics side threw me constantly, so Gary did the heavy lifting, I copied his work, and I walked away with a diploma and not much real understanding of electronics. That's exactly what happens when you let AI be your thoughts instead of your tool. You end up reading someone else's words with all the warmth of a bored McDonald's drive-through worker, then wondering why nobody connects with it. You can't be a thought leader on somebody else's thoughts. Think about how backwards that is. And if that approach ever landed you a speaking gig, then what? You'd have to ask AI to write the speech too. But AI Helps Me Do So Much More! I know the counterargument: AI helps you get so much more done, so much faster. I get it, I use it too. But think about two bowls of mashed potatoes. One came out of a box. The other is Grandma's, hand-mashed with a stick of butter and a healthy amount of salt. I grew up on the boxed kind. It's fast, it's easy, and it's fine. But nobody goes back for seconds on boxed potatoes. Grandma's bowl, you eat until it's empty and your stomach hurts. That's the difference. When you let AI write your content, you're serving your audience the boxed version, and they can tell. They Come For the Content and Stay For the HOST Here's the thing people forget: listeners come for the content, but they stay for the host. My very first podcast, years before the School of Podcasting existed, was called The Marketing Musician. I'd meet listeners of my current show who'd bring it up, and I'd assume they were musicians too. Almost every time, the answer was "no, I don't play anything, I just think you're funny." They didn't stick around for the topic. They stuck around for me. If you let AI take over as the host of your own show, you're left with nothing but AI. And if I don't need you anymore, I don't need your show either. I can just go ask AI the question myself. How I Use AI So here's how I actually use it. I don't let AI write my scripts. I let it help with research and organization. For my next School of Podcasting episode, I recorded my reflections on the Empowered Podcasting Conference by just talking into a microphone. No outline, just a brain dump, the way you'd brainstorm out loud. I ran that recording through Otter to get a transcript, then asked AI to pull out the main topics I'd already talked about and organize them in a way that flowed better, since I knew I'd circled back and added points out of order. I asked for a list of bullet points from what I'd said. Not new content, just my own words organized better. And honestly, it came out better than my original brain dump. I also lean on AI for images, because if I had to design my own graphics, you'd be looking at stick figure theater. Use AI for the things you're genuinely not good at, but keep yourself in your own podcast. Need Help With Your Podcast? If you're just getting started and thinking, "Dave, I don't even know where to begin with any of this," you know where to find me. Head to schoolofpodcasting.com and use the coupon code LISTENER for a discount on a monthly, quarterly, or yearly membership, and you can work directly with me and the community there. We've got step-by-step tutorials to walk you through everything. I've been doing this for over 20 years, I love helping podcasters, and I can't wait to see what we build together. I want to be your podcast consultant. Key takeaways Using AI to write your script in "your voice" strips out the one thing listeners actually stick around for: you. People come for the topic, but they stay for the host — lose the host, and you're just competing with AI itself. Use AI for research, organizing your own ideas, and things you're not skilled at (like graphics) — not for generating your content from scratch. A good process: record a brain-dump, transcribe it, then ask AI to organize and summarize your own words rather than write new ones. Links mentioned schoolofpodcasting.com — use coupon code LISTENER This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy

Should You Publish a Bad Interview? (Even If It Hurts Their Feelings) You just finished recording. The audio is clean, the guest showed up on time... and the conversation was completely forgettable. Do you publish it anyway, or do you tell your guest "no thanks" and risk an awkward, guilt-ridden conversation? A Reddit question kicks off this episode: a new podcaster recorded an interview that just wasn't interesting, and now feels guilty about shelving it. I break down why that guilt is misplaced, how to spot the guests who never cared about your audience in the first place, and the exact line you can use to say no without burning a bridge. In this episode, Dave covers: Why a "meh" episode is sometimes on you, not the guest — and the research questions to ask before you ever hit record How to spot a guest who wants to pitch himself to you instead of serving your listeners The pre-interview trick that saves you from wasting a full hour on a guest who isn't a fit The exact script to use when you have to tell someone their interview won't air Why setting the "I reserve the right not to publish" expectation up front makes the hard conversation so much easier The goalie mindset: why protecting your audience from mediocre content matters more than one guest's feelings A line worth remembering: "Do I want one stranger mad at me, or my entire audience going, 'Is it time to swipe left and unsubscribe?'" If you're building (or fixing) your podcast and want help with exactly this kind of decision, Dave's been coaching podcasters for over 20 years at School of Podcasting. Use coupon code LISTENER for a discount, backed by a 29-day money-back guarantee. This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy

So Many Options I've been playing with AI (Claude, ChatGPT, Grok, and an all-in-one tool called Magi ) and I finally did the thing most of us avoid: I picked one and went deep instead of bouncing between tools. Hello Claude In this episode, I share what happened when I handed Claude a transcript and a few of my own blog posts and asked it to write my show notes, why I'm keeping AI on a short leash (folder by folder, not my whole Google Drive), and the bigger lesson underneath all of it: knowledge is only power when you actually do something with it. Whether it's AI, or finally plugging in that microphone that's been sitting in a drawer for three years, the principle is the same: knowing isn't the finish line. Doing is. On a Budget? I need to check and see what the difference of Claude in Magai, and a direct use of Claude. As Magai is $20 for a ton of tools, but much like many "all in one" systems, I may be missing some of the customizations. Key Takeaways Pick one AI tool and go deep, rather than chasing whichever tool looks coolest this week. They're all converging on similar features anyway. AI should polish what you've already created, not create it for you. I fed Claude a transcript and my own blog posts so the show notes would sound like me, not like generic "let's delve into a deep dive" AI-speak. Watching tutorials isn't the same as doing the work. I watch AI tutorials, then immediately try what I learned, and sometimes it's great, sometimes it flops, but either way I learn something. Start small and contained. I'm giving AI access to one dedicated folder on my computer, not my entire Google Drive, until I trust it more. The same principle applies to launching or growing your podcast. You don't need to know everything before you start. You need to plug in the mic, watch the levels move, and take the next small step. Resources Mentioned Magi, the all-in-one AI tool subscription I've been testing (~$20/month) Claude, the AI tool I chose to go deep with ChatGPT, Grok, other AI tools I've tested Wispr Flow, the voice-to-text tool I use to dictate show notes Craig Van Slyke, AI Goes to College: aigoestocollege.com Podcasting Observations newsletter: podcastingobservations.com Audacity, free recording software mentioned in the mic-setup example The School of Podcasting, weekly classes on growth and launching Work With Me I've been helping podcasters for over 20 years. If you've got the knowledge but haven't taken the next step, whether that's launching your show or finally growing it, that's exactly what I do as your podcast consultant. Head to the schoolofpodcasting.com to join a live weekly class, or reach out about working with me directly. From the Human I let Claude take a swing at the above show notes. I removed three of those stupid "M Dash" things, but the rest is AI. I'm not sold on how this is saving me time. It did write some social posts. This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy

Does My Podcast Idea Have Legs? There's Only One Way to Find Out Most of the time, I cringe when I hear someone say "just start," because it usually sounds like advice from a person who isn't looking where they're going. Start what? For whom? Toward what goal? There is, however, exactly one situation where I will look you in the eye and tell you to just start your podcast, and it's the situation I keep running into over and over again on Reddit. Homework is Done - But Can't Press Record It shows up as a question that sounds reasonable but can't actually be answered: "Does my podcast idea have legs?" The person asking has usually done their homework. They've researched, they listen to plenty of shows, they have a format in mind and a list of episode ideas ready to go. What they're really asking is for someone to promise them their show will work before they ever hit the record button, and nobody on earth can make that promise. In this episode, I walk through why that question keeps people stuck, the chocolate toast test I use to think about any untested idea, and how to get honest feedback once you actually have something recorded. If you've been staring at the same pile of Post-it notes for the 400th time, this one is for you. Mentioned in This Episode School of Podcasting (coupon " listener ") Fix My Podcast (auditing service) This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy