For the last couple of months, FabRiders has been working with School of Data on the creation of a podcast series called ‘Data is a Team Sport’ and thought it would be worth sharing what we’ve learned.
Why a podcast?
The word podcast combines ‘iPod’ with ‘Broadcast’. Podcasts were initially developed when digital audio players like the iPod came onto the scene and audio content was becoming more readily available via the internet. Listeners subscribe to a podcast feed, usually via an app often called a podcatcher, and download new episodes as they become available. Podcasts developed as alternatives to traditional radio broadcasts, allowing people to produce audio content and distribute to listeners.
Podcasts are still gaining popularity as the format and medium are being explored. Most popular radio stations are releasing their shows as podcasts these days for distribution and we are still seeing innovation and exploration of the podcast format.
FabRiders has been involved with the development of a handful of webinars and have noticed that what’s valuable about webinars is the recordings and outputs that they produce. In terms of how people interact with webinar recordings, more people will listen than watch, given attention spans, tech constraints and bandwidth. This has led us to the conclusion that webinars should err on being listenable over being watchable, so we had started to think of the format as being more ‘podcast’ like.
For School of Data, the emphasis was on producing outputs that individuals could engage with at any time and the idea of undertaking research as a series of ‘listenable’ interviews was very compelling. The podcast is an output of the research activities to explore the ever-evolving data-literacy eco system.
Step 1: determine content & goals
We thought a lot about what topics would work in an arc of a podcast series. We knew there was a few key things we wanted to explore – namely the data-literacy eco-system and how it’s evolving in the wake of emerging theme’s like post-fact. Focusing on data-literacy within an eco-system gave us a variety of different themes that we could focus on in individual episodes, such as ‘data-driven journalism’ and ‘data-literacy in advocacy organisations’.
We started listening to lots of different podcasts to better understand what would make it listenable. We looked for more fact-based podcasts and appreciated those that gave you the experience you might be sitting at a dinner table with a group of friends that were sharing insights. A big inspiration was Vox’s The Weeds, as it takes the relatively dry subject of Washington DC insider policy mechanics and makes it accessible and entertaining. This also made us think a bit about who we should have as guests, as we want to make sure they will be engaging and provide a comfortable listening experience.
We also knew that we wanted some of the dynamics of having an online live event like a webinar, in order to draw on an audience. We also realised that we would never get the guests together all in the same room as the School of Data network is global. We decided to use Google’s ‘Hangouts on Air’ which allows us to convene guest speakers from anywhere they have a decent internet connect, record and publish directly to YouTube, and also allows us to have an audience. We publicise the live event as being an ‘online conversation’.
Step 2: prep your guest speakers
Aside from the editing of the podcast, this is probably the place where you need to spend the most time. I try to speak and meet with each guest twice before we have the online conversation. The first meeting is to determine what they will talk about, and the second conversation is a dry run through that involves testing out the technology on their end. The aim is to make your guest speakers feel as comfortable and prepared as you can. We have the run-through within a 48 hour period before the online event to assure it is fresh in people’s minds and still have enough time to address any issues with equipment or bandwidth. The run-through is also recorded and made available to the guest speaker for review.
Step 3: record your audio
As I mentioned, the online conversation is recorded via hangouts on air which allows us to have guest speakers and an audience that can use the text-based chat room to contribute questions and comment. We remind everyone to stay on mute and the guest speakers come off mute when they are ready to say something. This is especially important for the audio quality. After the online conversation, I use a web-based youtube to mp3 converter to get my audio file for editing.
Step 4: edit your audio
This is the most time intensive part of the process. In doing research for which audio editing program, as a Mac user, it came down to either Audacity or GarageBand. As an open source advocate, I really wanted to use Audacity – but as a freelancer who is doing a lot of different things at once, I found Garageband’s ease of use to be a winner. It also has a lot of bells and whistles (literally) that can help an amateur like myself produce a more professional sounding podcast. That said, you may want to check out this review of podcasting software which points to Audacity.
What takes the most amount of time is the listening involved to do the editing. In editing and splicing the audio, you want to make your guests sound more straightforward and make it easier for listeners to get the point the speaker is making. People often talk as they think something through and it can be hard to follow. The ability to make precise splices and take out bits that don’t make sense is invaluable to having a finished more polished podcast. It’s important to give it a final listen all the way through to make sure you’ve got the flow right AND that you haven’t inadvertantly taken out important points.
Some other pointers:
- don’t be shy to rerecord your own voice, especially if you can make yourself sound clearer. I realise that when I ask a question, I’m trying to give un-needed context before I get to the question and rerecording it with the simple/straightforward question makes an easier listen.
- put in occasional podcast idents to remind listeners about what they are listening to.
- add music or other bits of audio queues other than voices.
Step 5: publish the podcast
This bit was a huge learning curve for me – I just assumed you would post it on your website and produce an RSS feed that people would subscribe to. Turns out that hosting audio content is a bit more than what most web-hosts are prepared to handle, so you have to find a suitable place to host it. I settled on Soundcloud but there’s a few others to choose from like Simplecast
Once you’ve got your episode uploaded you still have a few steps to make it accessible to everyone:
- generate an rss feed that you can distribute and people can use with their podcast player to subscribe to your podcast. Ours is: http://feeds.soundcloud.com/users/soundcloud:users:311573348/sounds.rss
- Submit your podcast to directories associated with popular podcast players such as the iTunes Store (for iOS’s podcast app) and Stitcher (for android and everyone else), and anywhere else you may have potential listeners.
Have also learned that titling things ‘Episode 1’,’Episode 2’ is not really necessary as they come out in sequence anyways. You also don’t need to include the title of your podcast series – for example, I titled the first episode ‘Data is a Team Sport, Episode 1: Enabling Learning’ – by the time I got to release episode 2, I realised that everyone was seeing ‘Data is a Team Sport’ twice. I also realised that people are getting the podcasts as we are releasing them and sequence is not that important given the topics, so I ended up retitling episode 1 as ‘Enabling Learning’ and episode 2 is now ‘Data-Driven Journalism’
- Mariel Garcia-Montes excellent podcasting for movement building