There are three key ingredients for a successful podcast: message, consistency and quality. JFrog’s developer advocate and chief sticker officer Baruch Sadogursky shares the things he’s learned from his varied podcasting experience in this entertaining and practical talk from DevXcon San Francisco 2018.
Why we listen to podcasts…traffic, air travel, it’s good enough, right? Okay. Let’s try it again. It works.
So yeah, I’m going to tell you a story and a lot of technical details on what you should and what you shouldn’t do – about DevRel Radio, but also a couple of other podcasts that I participate in. Before we start, my prepared page, which is actually the idea of show notes comes from a podcast generally becomes the show notes. The slides are already there and I wanted to take a video on my phone and publish it like tonight, but I crushed my phone, so we will wait for the organizers to publish the video, so no video for you. All the links, that’s the important part. I speak about tools and platforms and everything, all the links are there. Comments, this is where you say that I rule, right? And ratings, five is the best, not one. Little raffle to thank you for being here, an Amazon Echo Dot for the people who love their stuff.
You know what? Let’s start with the poll. Let’s see how it works. And that will be a reason why we do podcasts. So who loves to speak? Who loves to write? I mean, it didn’t work. I tried. Well, yeah. I usually, me personally and my colleagues prefer to speak than to write and this is why we do podcasts and not blogs generally. So if you’re like us, I’ll tell you a story about three different podcasts. They are quite different. I participate in all three of them, so I can compare and share with you the differences. The first one is…any Russian speakers in the audience? Okay. So it’s only me and you. Okay. Max was here but he left. Okay. I tried to find you all day because you are the only people who will connect to this one. Razbor Poletov is a podcast about IT in Russian, 160 episodes, two plus hours each. That’s a lot of hours. Since 2011, I would say, million-plus downloads, about 10,000 downloads per episode. Think Software Engineering Daily, but in Russian and two hours long and every two weeks. But everything else is the same. Well, not so much. But the idea is we talk about stuff that we care about and have six permanent residents. That’s why it survived so long because, kind of, when we’re sick of it, we take a break. And its known for excellent audio quality. No sarcasm; I’m going to speak about the importance of that.
The second one is very, very different. And that’s Groovy Podcast. Anyone heard about Groovy Podcast? Those are my two listeners, they’re right here. What a coincidence. That’s great. Well, 66 episodes since 2014. Last one is number 57, deal with it. I have no idea how that happened. And 60,000 downloads which means, well, moderate success. I’ll call it that way. And it’s a podcast about Groovy obviously as a programming language. If you have anything to do with it, check it out. It’s basically me and Ken Kousen who is amazing and does all the work. I just come for the laugh. And so, yeah, we’re going to speak about that. And the thing of Groovy Podcast is that it’s extremely low-key production. It takes literally nothing for us to get aligned, record the podcast, and put it out there. And I’m going to speak about that as well, of course.
And the third one, the meta of them all is DevRel Radio. So DevRel Radio is, well, the first season is lost in the time and space continuum. There are a couple of episodes scattered all over the internet if you want to search for them. I didn’t find any, but they’re there. We rebooted it this year in January and have five episodes out. Anyone heard about DevRel Radio? Yeah, a couple of people and that’s because we have an astonishing number of downloads and you’ll see the popularity here. But today everything changes, thanks to you. And yeah, it’s a podcast about dev rel. And we did a triple meta episode last week. We recorded an episode of DevRel Radio preparing me for this stalk at DevRel conference about DevRel Radio. Triple meta, awesome stuff.
It’s basically three people that I’m going to introduce in a second. And it’s known for very, very chatty and annoying hosts – and guests – because we love to speak, as I mentioned. So Viktor Gamov, he’s my co-presenter at DevRel Radio, @gamussa on Twitter, solution architect and undercover developer advocate. He tries to get this position. They won’t let him out of solution architecture. And he’s from Confluent. And interesting fact, he’s co-author of the Russian podcast as well. The amazing Tim Berglund, @tberglund on Twitter, a senior director of developer experience. And he really should be here, but for some reason, he couldn’t come. Also at Confluent. And DevRel Radio was his idea. The name was his idea as well and he’s an amazing guy. And myself, I’m chief sticker officer at JFrog, we spoke about the importance of stickers, right? Yeah, but also head of developer relations, @jbaruch everywhere and this is my business card and then this part right here is actually peelable stickers. Stickers!
Okay. So I have one thought leadership and kind of inspiring slide. You can see by design. It’s like important. And that will be the only thought leadership and inspiring slide. The three most important things about podcasting when you think to run one are: Message – you need to come up with the idea of what your podcast is about, Razbor Poletov is about ranting about IT shit last week that we care about. Groovy Podcast is Groovy, DevRel Radio is dev rel.
Consistency – that’s kind of important. DevRel Radio didn’t do a very good job of it. We’re trying to get better. If you have episode once in a year, probably no one will listen to you.
And quality – we’re going to talk about quality as well.
What do you need? You need to plan or you don’t need to plan. You need some kind of equipment. You need post-processing. And you need website or show notes. And you need promotion, measurement, and feedback process. We’re going to talk about all those in the following seven minutes.
Okay. So let’s start with planning. What I’m going to show to you is a scale from very low effort to significantly well done. And I’m going to express my opinion on where you should stop in this quest for the top because done is better than perfect. So planning, low planning is let’s not plan. We have half an hour. We have an hour. We call each other or text each other. Let’s get online and record a podcast. Boom, the podcast is ready. It’s sometimes nice if you just want to rant about stuff and you’re creative enough to come up with enough content. Usually, you want some planning of the content. Yeah, so at Razbor Poletov we never plan anything. We just set up a date and then rant about stuff for two hours. A tentative list of topics is nice to have. And that will be the DevRel Radio and the Groovy Podcast.
We come up with a list of new stuff or stuff that we want to talk about and that’s good enough. You will also reuse this list later for the show notes. I’m going to talk about that. And the third one…I don’t know what’s going on with the fonts…anyway, what’s written here is basically this. So this is another podcast that I was invited as a guest, L8ist Sh9y, and take a look at that. They measured by the seconds. “From 2 minutes 17 seconds to 3 minutes 26 seconds, we’re going to talk about JFrog Artifactory.” That’s absolutely insane. I don’t remember if we met those seconds when we actually spoke, but that was really impressive. That’s before, right? It’s planning. It’s not actual timings, so somewhere in the middle, I’m going to say that a lot, somewhere in the middle sounds like a reasonable solution.
Equipment. You remember how quality is important? This is what brings you there. The low key is whatever built-on crap you have. It will be your mic or in your computer or your phone, which will probably require most post-processing. And I’m going to talk about that later. You can buy kind of mid-level microphones, stuff like Yeti or Rode Podcaster. They are very, very good in providing quality so high that you almost don’t need post-processing. And that’s important. If you feel professional and have too much money, you have the third level of a Heil and Shure and a soundproof room in your house or basement or whatever.
Anyway, sometimes you wanna record episode on the road. This is very nice because you can meet people in person, both your cohost and your guests, and then you sit together and you record with them. You can do it from your smartphone. Again, probably not the best idea for quality. For a couple of hundred bucks, you can get very decent equipment like Zoom H4N. This level of recording equipment records a group of people in decent quality. So that’s a very good idea. And again, mid-level of investment. And of course, if you have too much money, there are tools like Zoom Field 8 or Zoom Field 6. But for that, you will have to buy a set of wireless microphones. We’re talking about thousands, so if you feel like that.
Talking about software, when you actually record this stuff. So again, entry level, you can use Skype, you can use Hangouts. I personally hate both. Skype for those. These are guidelines of Skype broadcasting. The bottom line here is very interesting. Every 15 minutes you need to announce to your listeners that you actually do this thing on Skype. You didn’t know. Now, you know. Enjoy Skype.
And Hangout has just been horrible over the years. Constantly spins up the open fans of my laptop. I feel like in a helicopter. There is a very good solution, which is free, too cheap, and it’s called zoom.us. Zoom.us or BlueJeans or there are a couple of other decent conferencing tools. They allow you to record, of course, and that’s exactly what you need. You can turn on video. You can turn off video. But you will grab the audio and that’s a really, really good idea. I’m a big fan of zoom.us. I highly recommend it.
So that’s for recording. You have the file, you upload it to the platform we’re going to talk about in a minute. And sometimes you wanna stream because you, for some reason, think that people want to watch you live speaking about stuff. Usually, it’s not, but you might think. And then you can use stuff like Hangouts on air. It has all the downsides of Hangouts. Or you can use, again, zoom.us with webinars plugin which allows you to stream to Facebook or to YouTube or in other staff like Belive.TV. They’re decent platforms for doing that.
And we come to post-processing. So post-processing, again, low key: don’t do anything, just release. You recorded whatever horrible sound came out of your microphone on your computer and you just release it. People will hate it regardless of what you have to say. Podcasts with bad quality won’t take up, period. So either you invest in decent microphones or you do some kind of post-processing. And mid-level post-processing, which is very easy to do and won’t require much time is just using an open source tool that’s called Audacity. It has automatic post-processing with automatic levels and noise cancellation. Very nice stuff. So if you have mid-level microphones and you invest a little bit in post-processing, you get a really, really decent quality that you can roll on. If you want to get crazy, there is stuff like two-pass normalization with noise gate based on FTP. I have no idea what I just read. Viktor does that. He’s really a quality freak, but I don’t know.
Show notes. It’s very important to give people the ability to go to the website and see the stuff that you are talking about because you are going to mention tools, like I just did. And especially when there is no text on the slides, like it happened for some weird reason and you need a place that people will go and see. So that’s the show note page. You can do that with wherever you publish your podcast. Usually, you can create like readme for this podcast episode and you can put the links there. And this is fine. Well, not so much. Next step will be a dedicated page, GitHub pages or whatever you use for your own website will be just fine. There are static websites generators. This is something that we use. And the really next step that I personally very much recommend if you do this podcast for purposes other than just fun is a dedicated website with trackable and measurable conversions.
And this brings me to KPIs, which you might care or might not care about. So, first of all, the KPI that you definitely care is the download numbers. We are there for the glory and being vain, so we want to see our podcast popular. You need to measure download numbers. Every podcasting platform will give you that. So this comes out of the box. But, of course, if you do it as a part of your job as developer relations specialists, you want to measure conversions. I’m not talking about buying a product of my company, but any engagement that you care about can be measured from the show notes page. So you know how many people visited the show note page, clicked on different links, and then proceed further for different actions. It’s very easy to do. And this is how you convert your air audience which you have zero engagement with to actually something measurable that you can follow up on.
Last but not least, the hosting platforms. There are tons. There are three which are the leaders. I kind of selected the middle ground in their offering. So there is Podbean with a plan called Unlimited Audio and it’s very cheap. It’s nine bucks a month and it provides you with unlimited storage and advanced stats. And this plan on Soundcloud will be Pro Unlimited. Again, unlimited, advanced stats, 15 bucks. I don’t know why it’s almost twice as expensive but it’s not so bad because there is Libsyn which limits your storage to 400 megabytes a month and charges actually much more for that. So for me, Podbean is a winner. We do our Groovy Podcast there. It’s really good.
And that will be all. So just to remind you, show notes, that’s the same show notes but not from the podcast but from the conference talk. The idea is exactly the same. I will measure you like Heil. DevXCon is when you praise the stock on Twitter just now, I’m @jbaruch, @devrelradio. This is where you go to actually start pumping up those numbers that I spoke about. And obviously, we’re hiring. If you want to work for a chief sticker officer, please do. Thank you very much.
Can you make good release notes by collating your commit messages? Eva Parish argues not.
Can negative feedback from customer satisfaction surveys have a positive impact on your developer experience?