About a year ago, I described in length what my podcasting workflow for #DevJourney looked like. It's time for an update. Let me warn you, I will only describe the delta here, so read the previous article first if you want to get the full picture.
In the last months, I used Twitter intensively to find new interesting guests. I searched for:
- Active users,
- Tweeting about interesting topics,
- Which were followed by people I like,
- Which are open for direct messages!
Then I sent them a direct message introducing myself, asking them if they would be interested. I then offered to send them more information and thus asked for an email address.
The invitation (canned response in Gmail) I send per Email is way more complete than it used to be. I extended the invitation to contain all the information the potential guests could dream of. From the history and philosophy of the podcast, examples from previous guests, the recording process, todos for the guest and time estimates.
For a while, I had the feeling that it was too long and could be scaring off some guests. But it seems to be working fine.
A few potential guests never responded to my Twitter DMs, but almost all those who got my email responded.
Scheduling via Calendly
I settled on the tool Calendly for my scheduling. I set up default time windows in which the guests can pick a slot. I also hooked up my Google Calendar in Calendly, so that it can reduce my availability accordingly. It is not perfect, but it does the job pretty well.
The two features I miss dearly are:
- Support for multiple calendars (on the host side). I had to juggle a bit to merge my professional calendars and my Google Calendar.
- Pick multiple slots and let the host have the final say.
New: Extensive Notes
As soon as the recording is scheduled, I create an extensive markdown file for the guest containing all the things I have to do before the recording day, just before the recording, during the recording and after it. This helps my remain on track with everything. Here's an extract:
- [x] Fill up BIO
- [x] Fill up Kickstart Question
- How do I pronounce your name right?
- How is the timebox?
- Are there any subjects we MUST talk about?
- [x] Write show notes
- [x] Gather links
- [x] Extract quotables sentences
- [x] Write the title
- [x] Edit the episode in Audacity
- [x] Crunch the episode in Auphonic
- [x] Upload the episode in Buzzsprout
- [x] Create Videobite on Buzzsprout
- [x] Prepare Tweet & LinkedIn updates
- [x] Save initial tweet in Airtable
- [x] Write further tweet(s)
- [x] Create Poster
- [x] Buffer Tweet(s)
- [x] iOS Reminder for LinkedIn Update
- [x] Reddit Update
The note also contains useful information like the SEO tags I always use when posting a new episode, links to the admin pages of my hoster etc. It looks like a lot, but I can get through the metadata part of it in about one Pomodoro (25 min) if I do it conscientiously right after the recording, when the interview is still fresh in my mind.
Recording with Zencastr
I dropped skype alltogether for Zencastr. Zencastr works in-browser (no install required), it records both ends of the interview locally and uploads it to my Dropbox at the end. The quality is immensely better since it is not distorted by the eventual low bandwidth... as you hopefully have noticed by now. No more sweating on my end when I can't hear the guest right. Thanks Zencastr!
I made a few improvements to my physical setup as well. As you can see on the picture below, I use a sleeping pad (Therm-a-Rest Z-Lite) to prevent sound reverberation. This sleeping pad - that I bought to go on camping trips - has some kind of egg-shell structure that absorbs sounds very effectively. Since it is folded (and not rolled like many other sleeping pads), it holds vertically on my desk and can enclose the microphone. It also serves as a separation between the PC and the microphone and prevents eventual fan noises coming from the PC.
I used to take notes with a pencil. I now changed for a fine-liner. Those make way less scratching noise.
For in-person interviews, I still have my Zoom H1 recording device which works like a charm. But I have had a few complicated interviews in the past, in rooms where the reverb conditions were pretty bad, so I prefer to record online when possible.
Edit in Audacity
I still do very few edits in Audacity. I remove eventual mishaps that happened during the recording: a cough, a sentence that the guest restarted or a place where I accidentally cut the guest while they were speaking. A couple times, the audio tracks produced by Zencastr had half a second delay and needed to be readjusted. With the right keyboard shortcuts I've gotten good at it. And since it comes with a tremendous quality increase, it is totally worth it.
I usually edit right after the recording and it takes about 0.5 to 1 Min editing for 1 Min recording.
Release and Marketing
I have intensified the marketing. I usually send out 3-4 different tweets during the 2 weeks period and consistently post on LinkedIn, Xing and Reddit too. Here are the kind of tweets you will see from me:
- A couple days before an episode airs, I tease the new guest and hint at the current one in the same Tweet to get more synergy
- The initial one on release day with a "videobite" (30s audio segment)
- A quote of the guest as a picture
- Another normal tweet during the second week
And when I think about it, I also retweet those a few days later as well. This whole process takes about one more hour of work spread out during the two weeks
A word on Statistics
I observed that the first week after the release brings the vast majority of the downloads.
- About 1/3 comes right away during the first 24 hours, probably from subscribers.
- About 1/3 comes during the first 72 hours, probably the contacts of the guest that the social media
- The rest comes during the rest of the two weeks window, from further social media appearances
Since the recording goes pretty well at the moment (I manage to keep a 3-5 recorded-interviews-lead), I am tempted to double the release rate of the podcasts... but I still fear that this increase will not be sustainable on the long run. What do you think?
All in all, I think my workflow cycle time increased a little, but mostly because I am consequent with everything now. I never manage to do less than 60 minutes post-production, but I also rarely go over 120... so I guess it is still OK.
The results speak for themselves though. Since the reboot of the podcast 9 months ago, the listener base has grown steadily. The podcasts airing right now get in 36 hours more downloads than the older ones still have in total. So I must be doing something right... and that's thanks to you!
Keep making some noise about it. Tell your friends about it. Share the links and share the stories. And don't forget to engage with the guests and with me. Your encouragements, comments and word of mouth counts triple in this marketing game!
Note: The Amazon links are affiliate links
Photo by Marvin Meyer on Unsplash