Emily Shaw’s first audio project began as a means of teaching herself Vietnamese. She had spent time in Vietnam during college and fallen in love with the culture and language. So the next time she went, she brought a handheld recorder and taped people telling stories. 

She didn’t know what to do with the material; it sat on her computer for two years. Then, in 2015, “I found myself just compulsively interviewing people,” she says now. “Like I would go to a barbecue and then end up in another room with someone, recording a conversation on my phone.”

A friend named George Rosenthal suggested that they make something together at his San Francisco recording studio. That show became Cozy Boat, a podcast about music, creativity and passion. (The name is literal: The Complex had an inflatable boat filled with pillows where Emily recorded her interviews.) 

Since then, she’s created the show Candy Ears, as well as working as a freelance executive producer and creating content for companies like YouTube, Google, Meta, Spotify, Headspace, and Gimlet.

We talked to Emily about how sometimes it takes a few false starts to get good at freelancing, as well as how she balances a creative day job with her creative hobbies This interview has been edited and condensed for length and clarity. 

How did you learn audio editing?

I found the Duke Center for Documentary Studies. Unfortunately it doesn't exist anymore, but they used to have these intensive audio documentary institutes during the summer. So I signed up for a week-long intensive, and made my very first audio piece.

We learned Hindenburg in that class. I used that to edit the Cozy Boat episodes, although if I was doing it today, I would use Descript because it's much more beginner-friendly. 

Interviewing is definitely a professional skill. What did you learn as you did more of it?

Let me talk about what can go wrong. If you don't know anything about recording interviews and you're just chatting and asking about whatever you're curious about, something that can happen is that then when you're editing, you don't know what you're looking for. You don't know what you want to hear. And so it can be hard to know what the focus of the finished piece is. 

Now I do a lot of research about whoever's being interviewed. Ideally, I'm listening for a specific story with a lot of detail. If someone's being really general, that's not interesting, so I might follow up and ask, “Hey, could you give a specific example of that? Tell me what happened, or explain what happened step by step, in as much detail as you can remember.”

One last thing that is helpful is, when you ask a really long rambling question, you're going to get a really long rambling answer. So to try to be really clear and direct. Specific and short in the questions is helpful too. 

How did you start thinking about moving away from having audio as a hobby and towards pursuing it as a career?

I didn't have a public radio internship. I just made my own stuff, and got better by taking classes and making my own projects. I was working my way up creatively in audio until I had the skills for someone to pay me to do it.  And at that point it was almost like starting out as an executive producer, ‘cause my very first projects were directly with clients. I didn't have anyone overseeing me creatively. 

Basically what happened was, I made Cozy Boat. I started an AIR membership—the Association of Independents in Radio—and I had a profile on there. It linked to my very DIY looking website, which I'm horrified to look at today. Amazingly, this intercom company found me in the AIR directory and listened to Cozy Boat, and long story short, they were like, “We want Cozy Boat, but about our intercom company's 50th anniversary.” It’s not like, apartment intercoms—they make intercoms for rock shows, and NASA uses it, and the Olympics. 

I had some questions, but it actually turned out that they genuinely had a really interesting story. It did not end up being exactly like Cozy Boat, but I think hopefully there's a little bit of that fun attitude. 

Early projects like that, I was just figuring it out on the fly. Like, how do I do a contract? Everything took a million times longer than it does now because neither of us really knew what we're doing. 

What were some of those important early lessons?

You just have to learn by doing. One of the most important things that I tell people to do is track your time. That is the most valuable thing you can possibly do. So I tracked my time, and then I just had that information. 

You can keep building that with every project you do. I actually have a spreadsheet that's time tracking for every project I've ever done. So when a client reaches out to me and they're like, ‘Hey, we want this kind of thing’ I can say, ‘Cool. Let me check the three other projects I did similar to that.’ And then I feel more confident to make an estimate.

What kind of work were you doing while you were building up your freelance business?

I was the first employee at a mobile app startup. It was me and the two co-founders and then a few more people were hired. I was the customer support, sales, operations, all of that.

I was there for like a year and a half. And at a certain point, the co-founders announced that it wasn't working out.  And so I was like, I’ve been making Cozy Boat, and I’m really passionate about it. I didn't really know how I could make a living doing it, but I felt like I needed to give it a shot.

When that company shut things down, I was like, “Okay, I have six months of runway. I'm just going to try to do this.”

It did not work out that first time. So I ended up getting another day job. And to be honest, I was like, “Oh no, this is my passion, and it didn't work out.” It was a pretty low moment. But I just didn't have enough experience. It takes a long time to build enough network and connections and experience to actually sustain yourself working independently. 

So then I got a day job at Patreon in customer support. I was able to move down to four days a week at one point, and I would just spend that one day applying to jobs and making stuff, and just focusing on audio. I was able to get a full time contract at Pandora from that.

That was towards the end of 2018, and ever since then, I've only been doing audio. I've had the ups and downs, but I've been able to make it work. I want people to know that just because it doesn't work the first time, or even if you have to get a day job at some point, it doesn't mean that it's all over. 

How did you bounce back creatively from that first moment of “oh no, what if I can’t make this happen?”

I think a really key part of my path has been being uncompromising in terms of my own projects. My own projects have always been, this is 1000% what I am 1000% stoked to be making for myself. Because of that, there were no gatekeepers. 

But the trade off is that I didn't get paid for it. If you pitch a story to a show, then you get paid something. And you have to deal with the editor who gives you feedback. 

In terms of my own project, I was proactively seeking out feedback from friends and colleagues. I wanted it to be the best that it could be. I was always curious and hungry, like, “Why is this not working? Please help.” And so I think that counteracted some disappointment. 

Because I think you can kind of tell when something's not working. And I'm pretty comfortable letting go, because I want it to be effective. I want it to be the best it can be. I probably cut too much sometimes because I don't want things to be boring. 

I'm not trying to monetize my own passion projects. That's kind of unusual about my path. I have pretty much only ever made money from client work, meaning that an organization or a person wants to make a podcast, and they hire me to make a podcast for them.

But it all ties together: I make passion projects, and then clients hear those passion projects, and that’s what makes them want to hire me. Getting paid from client work then gives me time and space to focus on more artistic work. 

How did you start Candy Ears?

Previously I had always posted my passion projects on SoundCloud, and I felt like it was difficult for people to listen there. It's so funny, because my job is literally making podcasts, but it occurred to me—I was like, oh, I should start a podcast. I mean, Cozy Boat was a podcast, but that was a really long time ago.

I think one of the challenges is that this is my job now. And so a lot of perfectionism would come up for me around it—this feeling that anything I put out there needs to have perfect sound quality. People are going to think I'm bad at my job if I let it be a little rougher. So the idea of Candy Ears was like, “Oh, I can make this whatever I want and I can break all of the rules.”

Candy Ears is audio sketches, stories, and experiments. That gives me that freedom of, this is just a sketch, you know? This is just an experiment. It really celebrates the pleasure of listening. Everything I make, I want it to feel like ear candy. I want it to be something you want to listen to again.

I've been having so much fun giving myself permission. One of my most popular episodes is 30 seconds. Actually, in a couple of the episodes I use Descript. I created an AI generated version of my own voice in Descript, and then I juxtaposed that with some archival recording from the early 1940s—actually right after the Pearl Harbor attack. I think it's interesting to lean into that contrast between what feels organic and feels really human, and then something that's not human, and to play with that. 

One of my favorite pieces that I made with Descript was, I created as many different versions of my AI-generated voice as I possibly could. So there's this prompt that you read to train the voice. I'm like, I wonder what it would sound like if I recorded it from really far away, shouting. What would it sound like if I record it whispering really quietly? What would it sound like if I record it really high, really low? What does it sound like if I sing the prompt? A lot of them, it actually wouldn't let me. It didn't work because it's too crazy. 

So then basically the piece is just a collage of all these different versions of my voice reading the prompt. Some are real, some are the AI generated ones. It starts with, I take a breath. So one of the ones I did was, I took a deep breath and I read the whole thing in one breath. 

So I start with that, and then it's the first couple of words, and then it's all the different ones. And then the very end is me with the last little bit of air getting it out, and then huge deep breath in. So I feel there's an interesting juxtaposition of me as a human who needs to breathe air, and then these alter egos, alternate versions of myself. 

‎What does your technical setup look like?

I was very lucky with Cozy Boat that my friend, George, had really fancy, nice microphones. I don't know what type of microphone it was, but they had these long mic stands so that we could be like, lounging in the boat and have the microphone just exactly in the right place. George was very generously monitoring everything from the control room so I didn't have to worry about the sound at all. So that was a very luxurious experience. 

In terms of my current setup, I most often use a Shure SM7B with a Cloudlifter, and record either into a Zoom H4N recorder or through a Scarlett 2i2 interface. For client projects, I almost always use Riverside. I'll actually usually send a microphone to the guest, and send headphones.

And what software do you use to edit?

I got started in Hindenburg, which is a little bit more user-friendly than Pro Tools. When I started that job at Pandora, I was very lucky that they hired me—they use Pro Tools and I didn't know Pro Tools. I learned Pro Tools working there.

I still use Pro Tools today for finer edits and sound design. But for the paper edit, I use Descript. I don't want to make this sound like a commercial, but it has been like a total game changer. It saves so much time, just being able to copy and paste things. 

I remember talking to Amy Standen who works on the show Ear Hustle. She's an extremely talented producer and story editor. I was very resistant, because I was just using Pro Tools. You have your whole process, you know? And I was so resistant to trying something new. She was like, No, it saves so much time. You have to. 

I kind of explored it a little bit, and I was like, no, it's too hard, too scary to change. And then I was working with the sound engineer Zach McNees, who's amazing. And he was like, no, we need to use Descript. 

It’s just a total game changer in terms of saving time. I use it for classes, if I take a class or even do a skill share with someone, I'll record the Zoom meeting and then put it in Descript and print it out and like review that. I honestly am a super fan. Like I know that's not like the vibe of this conversation, but just being real. 

Listen, we love to hear it! How have you thought about bringing an audience to your shows?

It’s mostly just word of mouth. I'm not trying to build a huge audience; it’s more like a business card. If someone asks about my work, I want them to be able to find it really fast.

I try to spend as little time on social media as I can. I would rather have the free mental space to be creative, versus feeling distracted by who's liking and commenting on my things.

And I'm not trying to monetize it with advertisers or anything like that. 

It’s very, very, very hard to make  money making a podcast, to be honest. Those shows that have a really big audience and are able to have it be their full time job is like, such a small percentage of all podcasts. 

Once I learned that I was like, I'm not gonna try. It doesn't feel worth it. Especially because I want to really enjoy the process of creating and do it for myself. 

One benefit of it is that because I'm able to be so free creatively, that does show potential clients what I'm capable of, and leads to higher paid, better projects. If you never have that opportunity to really make what you want to make, then people don't get to see what you're capable of. 

How do you balance having a creative job with keeping some of that energy for yourself and your own projects?

Do you have another hour? There’s so much.

One thing is, it’s okay to do other stuff creatively. There have been times when I felt satisfied with what I was doing for client projects, and then for fun, I would like, go for a walk outside.

I always tell people my favorite sound is quiet.  So it's nice creatively. Like we actually need to have some free space for our brain just to process and have a tactile sensory experience in the physical world. It's totally okay to take a break. 

Option two is doing other creative things. I took some really fun watercolor classes. It was good to try something new, and I didn't have the baggage of perfectionism and all of that relating to it being my career. It's more about the process of your brain opening up than it is about what medium you're using.

Option number three, which I've actually been doing more recently, is having a daily creative practice with audio where I'm really focused on the pleasure of the experience. There have definitely been times when I felt a lot of shoulds around audio, like I should be crafting really narrative pieces. Even though I love doing that, it feels like work! 

So, side story, I'm in a triathlon club, which is also a really nice way to physically be in my body and not like in front of a screen all the time. And I love the sound of bicycles.  So long story short, I got someone from the club—we just planned a time to get together and try to make as many cool sounds with a bicycle as we possibly could. We ended up being able to record this crazy sound where it sounds like a gamelan. 

One of the client projects I made for YouTube was about ASMR. Working on that got me in tune with a different experience of audio that's more like a sensory experience of like, oh, you can feel things like when you're listening. It's not just a story being told. 

What advice do you have for someone who wants to start a podcast of their own?

I would say just make it. You have to make it in order to improve your skills. Ira Glass says this, that when you first get started your taste is so high, and that's why you're into something, but your skills are not there. You have to have the perseverance to keep doing it while you're not meeting your own high standards and taste.

Stop thinking about it. Do it. If you don't have a microphone, pick up your phone, use your phone. Just make something. It’s so easy to get in your head and plan and plan and plan. Once you start working on it, it's going to be different from what you thought anyway.

One other practical thing I would say is set a certain number of episodes that you're planning to do. I think one thing that can be daunting is people are like, once I start a podcast, I have to do it forever. And so it could be like, “Okay, six. I'm going to make six episodes and then I'm going to reassess. Am I enjoying this or not? Do I want to stop or continue?”

‎