Lessons Learned From Killing a Podcast
The Bitesize Irish Gaelic Podcast ran for 2+ years. It was mainly an interview show, where I would interview a new person each episode.
The podcast was a success. It went through tough times, at times, because I didn’t have guests lined up to keep to the publicised schedule.
If you’re looking to set up a podcast, and looking for ideas, maybe this can help you. The assumption is that it will be more sustainable if you have other people there to help you.
Put your team in place
Who you need to support your podcast:
- Audio editor: forget editing audio yourself. Record the podcast, and share it somewhere for someone else to pick up.
- Researcher: that’s the best name I have for the person who will contact new potential guests, schedule interviews, and follow up with them afterwards.
- Content editor: edits the shown otes page to insert a featured image into that blog post
I used Upwork to find an audio producer (by testing three producers on the same episode). The researcher was in fact the Bitesize Irish Gaelic customer support person.
We used Google Docs to share a document that had SOPs for the three roles:
- Audio Producer
Plus, we had a shared spreadsheet that listed podcast and future episodes. It included dates of when the episode would be published, its recording deadline, and its audio producing deadline.
As an aside, don’t promise future work to potential contractors. Define one specific piece of work to post on Upwork, like adding intro and outro music to our episode. Select multiple applicants. I’m always surprised at the tiny number of actual good contractors there are, after huge numbers of applicants.
Assign a regular task of finding new guests
Our podcast’s weakest point was not having guests lined up for the next show. You might have the same problem with your podcast, in that you don’t have a topic set out.
We always struggled to keep a buffer of recorded interviews. The goal was to have a couple of recordings in the buffer, lined up for publishing.
Somehow, though, I didn’t have in place a system to contact new guests. That process was too simple in the end:
- The researcher was assigned a weekly task of contacting two new potential guests. If there were no potential guests in our list of potential guests, then the researcher’s role is to find new ones.
If two contacts per week wouldn’t be enough, we could always adjust that number later.
Define steps in the process for SOPs
The podcast was always run according to standard operating procedures, thanks to inspiration from Dan at Tropical MBA.
But the podcast process was pretty complicated! It included delegating to the researcher the interview scheduling, guest research, and interview followup, plus production SOPs for the audio editor.
There were lots of steps for the researcher to do, like updating the podcast production spreadsheet with guest’s information, and keeping our related Trello board up-to-date.
If I was going to refine the process further, I’d define tasks based on events (like the audio file being uploaded to Dropbox). I suggest having specific steps for the researcher to take at each step of the episode process.
Our checklist defined these actions, but the different points of the checklist didn’t have triggers that actively asked for a step to be completed.
If you like SOPs, check out the bonus at the end of this post.
Hassle your guests for recommendations
As with running a business on word of mouth, the recommendations of guests for other interviewees were excellent.
Make it part of your system that each guest is asked to recommend three other guests.
To create this system, document it as a step to complete after an episode has been recorded. I assigned this task automatically with the help of Zapier (that’s another blog post).
Delegated Scheduling: Two slots per week
Scheduling was a really hard task to delegate to someone else.
I tried sharing my calendar with the researcher. Yet, there were last-minute clashes with interviews arranged and my own plans. I’d get and email saying “I’ve scheduled your interview tonight at 8pm.” But I already told my friend I’d meet him during that time!
It was important to build up a buffer of recorded episodes, and then to not let interviews be scheduled too close to the day. So there were a couple of outsourced scheduling principles I learned:
- Don’t allow interviews to be scheduled closer than 7 days away
- Reduce coordination complexity by only accepting interviews twice per week at specific times
Get the interviewee in on conversation
Make it part of your system that the guest is contacted with any comments left by listeners on the show notes page.
An interviewee will usually have reason to help out, since they’ve got a vested interest like promoting their blog.
Finish the podcast, don’t let it die
This past week, I was looking at my phone and saw a calendar notification that I had a podcast interview to record that evening.
I felt stress from seeing that notification. It felt like work. It was also going to affect my sleep that night, as I’d have to record it after the kids were asleep.
After talking with my friend, I decided that this was the end of the podcast, after 60+ episodes. It was an immediate weight off my shoulders.
Everything must die. Your podcast or blog is not eternal. It will die.
Don’t let it die a death of abandon, with no information on why you ended the publication.
I was very glad to realise that I wanted to end the show, and gladly recorded the last show to explain to listeners why I wouldn’t be continuing with it.
Track audio production against the schedule
I used Zapier to assign the task of audio production, by sending an email to the audio producer.
What I should have done was track that task somewhere more visbile (that’s on Trello boards, in the case of Bitesize Irish Gaelic). That would have let me see when an episode wasn’t produced and something had gone wrong.
Would you like to see the SOPs we used to run the podcast for two years?
Download our actual un-edited podcast checklists here (PDF)