Tuesday, May 23

The world’s most efficient podcast production pipeline

Creating a new podcast and getting it online can be super easy if you use a service like libsyn.com – there are also many wordpress-based solutions (Podlove Podcast Publisher for example) but having done this podcasting business for 12 years, I like to take a bit more control while reducing the post-produciton time per episode to as little as possible.

I’ve seen many podcasts come and go and I believe one of the reasons a lot of people don’t stick with it is that they burn themselves out on the post production. All the making-it-sound-pretty and the shownotes and the promotion and the constant search to optimize the tech.

My approach is a little more like putting a puzzle together into a well-oiled machine that I feed a fresh recording and let it do the rest for me.

And as I just now brought a new podcast online (about film photography, in German), I invite you to have a look at that process.

Most important and at the very beginning is of course the decision to make a podcast at all. Having a topic is even better. And being able to do it regularly is an important step to building a loyal audience. Monika and I had thought about doing a film photography podcast for quite a while, but never really managed to kick it off, because of other commitments. Then after we recently got asked on Twitter if we knew any German film photography podcasts, and we couldn’t really think about any, we looked each other in the eye and decided to finally do this.

Here are some pieces of my very personal getting-a-podcast-off-the-ground puzzle. Your own approach might be very different from this:

  1. Decide on recording infrastructure: should we use a professional full-fledged headset setup with sound-proofing in the room and editing each episode or would a small microphone based ad-hoc production setup be sufficient? This one was easy to answer as we already have enough experience producing and cleaning up iPhone-based recordings. So iPhone it is (iPhone 6s in our case). The audio quality is remarkably good (if you don’t breathe straight into the microphone), it allows us to record wherever and whenever we want, and we can directly publish from the very same device we use to record. Flexibility wins. Based on previous experience, this decision took us just a few minutes.
  2. Decide on packaging (music, editing, etc.): what should it sound and feel like? Do we want an intro, an outro, jingles, music, ear-candy of any other form? We decided on a short and simple music intro and a slightly longer music outro with some voice-over telling people where to find more. Other than that, the rest of the episodes are just us talking. This took a bit longer, especially to find the right music.
  3. Get voiceover for outro: get someone to record an couple of sentences for us to put in the outro. “You’ve been listening to…” – luckily we know someone who has the perfect voice for that and who was nice enough to do that recording for us. This took a day from asking someone to getting the finished recording back.
  4. Create/edit intro/outro: build intro and outro from the above pieces, test in context to see if it works. Toss audio back and forth in the editor, until it feels right for both the intro and the outro. This step took an hour.
  5. Set up podcast publishing system (firtz in our case): add subdirectory on server or set up new domain, install system, edit config files, test. This probably took the longest. Firtz is one of many ways to publish a podcast, we’re using it because in conjunction with auphonic (see below), it allows us to set up an almost fully automated publishing pipeline, which is pretty much the only way we can produce yet another podcast. With over a day, this one took the longest. You’ll need to be technical and know your way around at the command line.
  6. Decide on and create podcast logo: graphic design, minimum requirement is a square podcast logo at 1440×1440 pixels. No podcast without a logo. We took an existing one and modified it. Took maybe an hour and a couple of iterations until we were there.
  7. Set up tracking (subscribers, downloads): optional but we’d like to know if people actually listen to what we do. We use feed.press for this purpose. Setup of this took about an hour, especially to get the redirect right that is required on your podcast server to make sure feed.press works correctly.
  8. Set up post production pipeline on auphonic.com (templates, metadata, etc.): that system is the best thing since sliced bread. Takes your audio and massages it to sound good. Adds metadata and intro/outro automatically, then uploads to your server. This in conjunction with firtz is the world’s best production automation pipeline if you want to be as hands-off as possible when it comes to the audio. Setup time was under an hour and the resulting time savings are immeasurable.
  9. Test post production pipeline: make sure everything works by sending some test content through the pipeline. Had to adjust a few parameters before everything went smooth. But it’s set it and forget it. Works well now. Time for test was a just a few minutes.
  10. Record and publish several episodes: we like to have a couple of episodes on a new podcast before it goes live, just to give people a bit more to choose from and to make it look nicer. First episode was meta (why are we doing this, what is it?) second and third episodes were already with content. We did the first two in one go, the third one a day later. Recording time under 20 minutes each.
  11. Submit to iTunes and wait for confirmation: Very last step after everything else is done. iTunes is by far the most important directory to get your podcast on. Submittal is easy enough and if everything is golden, Apple will take a day or two to add it. Time for this is just a couple of minutes. Give them a link to your feed and some additional information, then wait.
  12. Tell the world about it: finally it’s online and can be subscribed to by the world, so we better tell everybody about it on social media and then brace ourselves for the initial feedback.

What’s awesome about this podcast production pipeline is that recording and publishing an episode won’t take much longer than the actual recording itself. Once you’re done, all you’ll have to do is send it to auphonic right from the phone you recorded it on, fill in some metadata and hit publish. The rest happens automagically.

In case you’re interested to listen to two film photography nerds (who wrote the book on it) talk about film photography, in German – or if you doubt that a smart phone-based recording can be any good, here you go: Absolut Analog – Der Filmfotografierpodcast

Let us know what you think.