The Party Corgi Podcast is mostly an interview-based podcast at the moment, which means researching, finding, inviting, etc guests. This is mostly manual work. For my first three episodes I did a full "these are the questions I want to ask you" sheet before hand and sent that with the invite. This turned out to be time consuming but moreso turned into a mental block on getting more invites out. This caused me to lag for months on getting enough invites out to start the podcast.
I've since switched to a more ad-hoc model. Other hosts I've talked to say they can do enough research in about an hour for a guest and I'm finding this is pretty accurate. 99% (fake percentage) of invites I've sent out have been very ok with not having the initial sheet, with a small contingent concerned that they don't know what to talk about. With the later group it usually takes a bit longer, but I frontload more research (I do some before inviting as well) and come back with topics.
The very first thing I need to do once I have a guest I want to have on the show is send the invite. I have a mostly templated invite that I can send out. I think this is both a good and bad thing. The template ensures that I include all the important points but (and this is my own feeling) it also feels... templated. Templates always feel less personal (to me) so I try to personalize the message with an intro or similar when sending.
Notably, calend.ly is pretty critical for scheduling. I can't imagine going back and forth with every single potential guest and scheduling it all manually. I personally wouldn't be able to sustain that due to personal quirks.
The calend.ly link shows up in my inbox (and calendar), which I then create a zencastr episode for manually (I wonder if I can automate this...). Zencastr allows me to record each guest, as well as myself, as individual wav tracks that I can edit independently later. My backup recording if this site fails (which has happened exactly once) is Zoom, which works OK in a pinch.
Zencastr runs in the browser, so there's nothing for the guest to sign up for, they just click a link. There's no video and guests need to hang around after the recording because the recording happens in the browser, so you need to wait for the upload.
After I get the recordings, they sit for a little bit until I have time to edit them down. I have a small, probably automatable process for my own audio and a very similar yet different for each guest process for guest audio. I pull the files into Adobe Audition, cut off the noise floor, normalize the audio to the same volume, throw the interview audio files into a multi-track session, drop the intro and outro pre-recorded files in, then record a custom intro segment for the post-intro/pre-podcast spot.
Once it's all edited I drop an .mp3 export onto my phone and listen to it in Pocket Casts, usually at 2x speed.
I also pull out an excerpt of 30-60s of audio from Audition while I'm editing. This gets dropped into After Effects along with an image background. The image background is a Photoshop template I drop an image of the guest and some text into. The audio is connected to a waveform and exported to .mp4 for a socially shareable video for sharing on Twitter.
I use SimpleCast to publish the podcast episodes. It has a draft button which allows me to set up the episode title, show notes, and other stuff ahead of time. The drafts can be scheduled to go live as well, which means I can detach the editing process from the publishing process and they'll just go out whether I wake up on time or not.
I take the excerpt .mp4 and schedule a tweet using Buffer to go out shortly after the podcast goes live. I make sure to include a very short text teaser and to tag the guest so they know the episode is live. The tweet goes out from the Party Corgi Podcast twitter account, which is the only way people find out about the podcast right now. I don't market it anywhere else.
That's it. I'm into the double digits of episodes recorded/edited now and feel like I have a good workflow.