The intellectual work comes first. A podcast dissertation is still a dissertation, and the design decisions need to be made before any recording starts:
Research questions come first. The questions should be answerable through the methodology you're proposing. If podcasting is your methodology, your research questions should be ones that benefit from voice — from hearing people rather than reading their responses.
IRB approval is non-negotiable. The fact that participants will appear on a publicly available podcast significantly changes the informed consent process. Work with your IRB early. Make sure participants understand what it means to be heard publicly. Give them genuine choices about participation level.
The "How to Listen" guide is a methodological decision, not an afterthought. Decide early what these guides will include and how they will function. They are the academic scaffolding that makes the episodes legible as dissertation chapters. They should be developed alongside the episodes, not added after the fact.
Convergent parallel mixed methods (as in Diamond's design): Survey data provides breadth; focus group data provides depth; the integration is where the meaning gets made. This is defensible, documented, and grounded in established methodology literature.
Purposeful sampling: Identify exactly what characteristics your participants need to have and why. Diamond was explicit: Black and Brown creatives, connected to higher education, with some familiarity with media or podcasting. The specificity matters for your methodology chapter and for recruiting the right people.
Episode structure: Decide before you start recording how many episodes you're producing, what each one will cover, and how they map to traditional dissertation chapters. You can refine this as production unfolds, but having a clear initial framework prevents both you and your advisee from getting lost.