AI Podcast Planning — Topic Research and Episode Structure That Hooks Listeners

Why 90% of Podcasts Die Before Episode 10

The data is brutal: 90% of podcasts never make it past episode 10. The number one reason isn't equipment, editing skills, or marketing. It's running out of things to say.

AI fixes this by generating unlimited episode ideas, structuring each episode for listener retention, and building content calendars months in advance. You'll never stare at a blank page wondering what your next episode should be about.

Step 1: Define Your Podcast Positioning with AI

Before generating episode ideas, you need clarity on what makes your podcast different. Feed Claude this prompt:

I'm starting a podcast about [BROAD TOPIC]. Help me find a specific 
angle that's underserved. Analyze the podcast landscape for this topic:

1. What are the top 10 podcasts in this space?
2. What angles do they all cover? (this is the saturated space)
3. What angles are MISSING? (this is the opportunity)
4. Who is the ideal listener that nobody else is talking to?
5. What's a unique format that would stand out?

Be specific. "AI for developers" is too broad. 
"AI tools that save solo developers 10 hours/week" is a position.

Claude researches the competitive landscape and identifies gaps. Your podcast lives or dies on how specific your positioning is. "Another tech podcast" gets zero listeners. "The podcast that tests one AI tool per week and gives an honest 5-minute verdict" gets subscribers.

Step 2: Generate an Episode Backlog

Never launch with one episode idea. Launch with 30. That's 6 months of weekly content before you need to brainstorm again.

Generate 30 podcast episode ideas for a [NICHE] podcast. 
For each episode:
- Title (hook-first, under 60 characters)
- One-line description
- Why a listener would care (the "so what?" test)
- Which episodes it connects to (for listener retention)

Mix these types:
- 40% practical/tutorial episodes
- 30% interview/conversation episodes  
- 20% opinion/hot take episodes
- 10% story/narrative episodes

Order them so the first 4 episodes are the strongest 
(they determine whether a new listener subscribes).

Review Claude's output. Kill any episode that doesn't pass the "would I listen to this?" test. You should have 20-25 solid episodes remaining — that's still 5+ months of content.

Step 3: Episode Structure Templates

Every successful podcast uses a consistent structure. Listeners like predictability — they know what they're getting. AI generates episode structures tailored to your format:

Solo commentary episode (20-30 min):

1. Cold open (0-30 sec): The most surprising or controversial point from the episode. No intro music, no "welcome back." Drop the listener into the action.
2. Context (30 sec - 2 min): Why this topic matters right now.
3. Main content (2-20 min): 3-5 segments, each with a clear takeaway. Use stories and examples, not just facts.
4. Hot take (20-25 min): Your strong opinion that listeners will either love or disagree with. Controversy creates engagement.
5. CTA + outro (25-30 min): Specific ask (review, subscribe, share with one friend). Tease next episode.

Interview episode (30-45 min):

1. Cold open: Best quote from the guest (pulled from the recorded conversation)
2. Guest intro: 30-second positioning (why the listener should care about this person)
3. Warm-up question: Easy, builds rapport
4. 3-5 core questions: Each building on the previous. AI generates these based on the guest's background.
5. Rapid fire: Quick, fun questions that show personality
6. CTA: Where to find the guest + next episode tease

Step 4: AI Interview Prep

For interview episodes, Claude is your research assistant. Give it the guest's name, company, and recent work. Ask for:

• 10 questions the guest has probably never been asked (avoid the boring "tell us about your background" opener)
• 3 potential follow-ups for each question based on likely answers
• Controversial opinions the guest has shared publicly (great for spicy segments)
• Connections between the guest's work and your audience's interests

This turns a 2-hour research session into a 10-minute Claude conversation. And the questions are better because AI has analyzed more of the guest's public content than you could manually.

Step 5: Content Calendar

Use Claude to build a 3-month content calendar with publishing cadence, topic variety, and seasonal relevance. Map episodes to current events ("Release this AI tools episode the week of Google I/O") and cross-promotion opportunities.

The calendar isn't rigid. It's a safety net. When you're creatively stuck, the calendar tells you exactly what to record next.