Sruthi Pinnamaneni

Sruthi Pinnamaneni

Intro from Jay Allison: Sruthi Pinnamaneni of Gimlet's Reply All has a theory about building radio skills that she calls, "Scaffolding." You start close to the ground and create a solid foundation (get good tape), add a layer (character) as your stories get longer, and then move to the upper levels (motivation, narrative, style, and the rest).

Sruthi lays out a comforting framework for a beginning producer, but fills it with smart tricks for people who have been doing this a long time. Finally, this manifesto is more than advice about building stories, it's advice about building a ramp to launch you in the right direction toward becoming a top producer. My bet: this turns into a curriculum.

Scaffolding: Starting Small

Learning radio on your own is hard. I used to think I could listen to stories I liked, figure out the process that created those stories, and then, you know — just replicate the process. One time I tested out my theory on a Radiolab piece: Lulu Miller’s A Bus to Nowhere. I sat alone in my apartment, dutifully transcribed the tape, wrote out all the Jad & Lulu lyrics, and then just stared at the words. How was this made? It was like trying to learn architecture by drawing a cathedral and guessing how to backwards construct the whole thing.

Still, I was optimistic. I found a newly published scientific paper that seemed interesting enough. I interviewed the scientists. I recorded a conversation with a radio friend, where I told him about the paper and played him the scientist tape. And then I put it all together into 40-something minutes of the most boring thing I’d ever heard.

I mean, this is a surprise to nobody. Yet it’s still the way most of us try to learn how to make compelling radio documentaries — by jumping right in and trying to make one. And then feeling supremely disheartened when we can hear that every part of it is broken.

Now I know I had it all wrong. You can’t learn all the parts of radio at once. I’ve become a big believer in the idea of scaffolding: that the way you learn to build a great story is by starting small. Make a humble little structure — a radio shack, if you will. With each small story you make, try to just get one thing right. Once you figure out the one thing, try to learn another. Keep building. And one day you’ll look back and realize you’ve made something that ta da, looks like a decent radio doc. It’ll have a great character! Maybe a question it’s trying to answer! And if you’re really the master architect — the path to a surprising end.

For me, the smallest structure — the one to start with — is the 3-minute story. Before I got a job at Reply All, I’d always been a freelance producer, first out of Vienna, Austria, and then New York. I had to find and pitch local stories. I had to do most of my interviews in the field because I didn’t have access to a studio. All my stories were short news features, but I dreamed of making documentaries. (Hence my tragic attempt to copy Lulu’s story.) I built 3-minute stories over and over again, until I was brave enough to pitch my first 8-minute feature. And I’m now at the point, ten years later, where I can just about comprehend what it takes to make a fairly entertaining 30-minute story.

But forget I said all that; let’s just focus on the first shack, that 3-minute piece.

[Quick disclaimers, I’m going to use clips of my stories here. Not because they’re great stories (they’re not) but because they usually have one tiny good thing in them that I was practicing. So I could eventually make slightly longer, slightly better stories. Also, I refer to 3-minute or 8-minute stories but those numbers are approximate stand-ins for short or medium-length stories.]

Scaffold Level 1: Tape

I once made a story about a payphone. I was a freelance producer and was beyond excited that a national radio show had given me an assignment — to stake out New York City’s highest-earning payphone and find out who was using it.

Listen to “Sruthi Pinnamaneni”

Let’s check off what’s broken here:

  • The tracking is bad. It sounds like “fake fun!”
  • There’s no logic in the structure. I seem to have picked my favorite pieces of tape and then pinballed from one favorite to the next.
  • There’s one good scene when the drug dealers show up and I ruin it by writing all over it, over-explaining but also explaining the wrong things.

But the one thing I got by on was the tape. It’s decent tape! I got a city official to sound human. And yeah sure, I should have cut out the tape about the filthy payphone smells — it doesn’t fit the story — but it is tape that makes me feel something.

Let’s actually talk about that for a minute — how to get tape. There are Transom manifestos on that subject from producers far more brilliant than me (just to name a few: Alex Blumberg, Marianne McCune, Joe Richman), all of which you should study religiously. But here are three things I found most useful when I was starting out:

1. Getting good tape takes time.

I did two interviews for that 3-minute Marketplace story. The interview with the city official lasted over an hour and I spent almost two hours with the guy who loved the smell of pay phones. A more skilled reporter could have collected the tape more quickly, but I needed all that time to try a hundred things, ninety-nine of which failed. Just know it takes time. You can’t rush getting good tape. Tell the people you’re interviewing that it’s going to take a while, that you want to make sure you have time to go into details, and that you’re trying to get beyond the usual sound bites.

Also, don’t forget — I spent hours staking out that damn payphone. I accosted lots of strangers, interviewed the Citibank security guard who watched the payphone all day and had lots to say about it, and I stuck around long enough that the drug dealers showed up. All of that required a little patience and lots of fear that my editor would be disappointed in me. My opinion: both those things — patience and fear — are essential ingredients for good tape.

2. Don’t put people on pedestals.

Alex Blumberg talks about modeling in interviews: as in, if you’re overly formal, it’ll make the people you’re interviewing feel as if they ought to be too. I think about that a lot: how am I behaving in an interview and is it helping me get good tape.

And here’s my main rule on the attitude I bring to the table: I want to be respectful to my interviewees because I’m thankful (and eternally surprised) that they’ve agreed to give me their time. But I also don’t want to put them on a pedestal. People on pedestals make for bad tape — they sound as if they’re handing down aphorisms, or as if they’re too otherworldly to make you feel anything. So, treat the person you’re interviewing as if you’re both on the same level. Don’t be afraid to interrupt them. And definitely, definitely challenge them — even if it’s something you agree with. As in: Why? Why do you believe that? Help me understand that. Or my dumb tick, which somehow works . . . “Really?” Gently nudging people will get you better tape.

3. Use the mic to take control of the interview.

You know what object I love more than almost anything in the world? A shotgun mic. Truly, it is your conductor’s baton. Here’s a way to use it: when interviewing in-person, position yourself as close as (comfortably/respectfully) possible to the interviewee — for instance, on the corner of the table right by them instead of across the table.

Put the mic under their chin so it’s out of their line of vision. Tell them, “I know it’s weird having this big pointy mic here but I promise you’ll forget about it soon.” And because you’re taking your time, not rushing the interview, they will.

And then during the interview you can use the mic to make people say less, or say more. If they’re going on and on, and you need to interrupt or direct, you can move the mic back to yourself, saying something like, “That is interesting; but I was wondering about this other thing” . . . Or, if the person said something thoughtful and lovely, you can look at them expectantly, and leave the mic in position without asking a question. Often, there will be a long pause, as they keep thinking, and then . . . they say something simply perfect.

. . .

Back to what all this has to do with scaffolding: just focus on one thing in the beginning. Pitch news stations or podcasts that take shorter features. Don’t get fancy with your writing and structure. Try to keep things simple, disciplined. Don’t worry about making your tracking sound great; as long as it’s not distractingly bad, it’s fine. But the one thing you’re going to get right is the tape. If there are four pieces of tape in the story, make sure every piece is perfect, an item worthy of display on a shiny trophy shelf.

And when you’re feeling like you’re pretty comfortable in your radio shack (it’s more like a prim cottage at this point) — your tape skills are decent and you can imagine scaling up — you can try your hand at the next thing, an 8-minute feature.

This is a length where you can really start to sketch out a person — and have space for that person to be complicated.

Scaffold Level 2: Tape + Character

Here’s a clip from a story I made for Studio 360. It’s about a problem that museums like MOMA were trying to solve — they had collections of new-media art dating back from the 70’s, with old CRT televisions and radio parts, and when those components would break, nobody could fix them. Because there are no more TV repair men and even if there were, nobody stockpiles those old electronic parts. Except this one man in Chinatown, CT Lui.

Listen to “Sruthi Pinnamaneni”

Again, I’ll check off the parts that don’t work.

  • The tracking is 10% better but still pretty bad. Stiff and sing-song-y.
  • The writing is better only in that there’s less of it cluttering up the scenes. There are slight ideas in the story about friendship and technology and death but I don’t know how to prop them up with writing.
  • The stakes are unclear. Why am I listening to this again? What happens to all that artwork when the one repairman is gone?

By the time I made this, I’d been practicing my scaffold 1- tape skills for eight years. So the tape is pretty good — the people sound comfortable, and there’s action and feeling in what they’re saying. And on top of that, I’m doing a better job standing in for the listener — I’m reacting to what CT is saying and restating things that are confusing.

But you know what else the story has that’s working? It has a character. I love CT Lui. He’s funny and charming, and says things that surprise me. And I found that as my stories got longer, I had to find people like him to be at the core of them. People who had layers to them. As in, the more you talk to them, the more interesting their motivations become. With CT Lui, there’s the obsession with old gadgets, and when you peel that back there’s a friendship that made him feel like he’s part of a rarified art world, peel that back and there’s loss and a need to keep things alive.

So, how do you find those kinds of people? Honestly, it’s mostly luck. But here are a couple of things you can do to increase your chances of finding a rainbow:

1. Don’t be afraid of the telephone.

I e-mail people a lot, asking them to talk to me or to introduce me to someone I want to talk to. But I vastly prefer calling to emailing. So when I see a phone number alongside an e-mail, I’ll use it. Or, rather than explaining myself at length over e-mail, I’ll write the bare minimum and ask them to call me.

A few years ago, I read an article online about the Hasidic philosophy around technology. It was by a woman who’d left her Hasidic community. I couldn’t find her contact info so I emailed the editor of another website, Unpious, where she’d also written stories. My translations are in red.


I’m a producer for Reply All and am working on a story about Hasidim and the internet. (I wasn’t actually working on it. Just looking for a story to pitch.) Would you be able to put me in touch with F. Vizel? I loved her story in Tablet.

Also, I’ve been thoroughly enjoying the essays on Unpious. Perhaps you’d have a moment to chat as well? (I can’t fish for a story via e-mail! Let’s talk.)

Some background on our show: Reply All, a show about internet culture. It’s hosted by Alex Goldman and PJ Vogt. It’s part of Gimlet Media, which was founded by Alex Blumberg and documented in the podcast series, StartUp. (Reply All was only a few months old at this point. So I’m trying to piggyback off the bigger show.)

If you have questions, feel free to call me at


I talked to Shulem on the phone. Asked him questions like, “What do you think about the Hasidim and technology? How do you spend your time online?” He told me about a website he couldn’t stop visiting. A place he called “Hasidic reddit,” but said he’d need to give me the tour in person so he could translate the Yiddish. I went to his place and recorded him showing it to me. He was a good talker –meaning, he spoke in anecdotes, told stories with dialog. And when I was leaving, he mentioned his estranged family. He had children, whom he hadn’t seen in seven years. His story had layers to it, and each layer made me feel something different. Here’s his story. I’m glad I asked him to call instead of just emailing him for contact info.

2. Look for obsession, dig into motivation.

I like people who are obsessed with things. CT Lui was obsessively collecting old parts. Shulem was obsessively trawling Hasidic reddit for glimpses of his old life. But why are they doing these things? There are reasons behind their obsessions, and those reasons can be powerful stories. Try to find people who have become the expert in an obscure field. Or are on some hyper-specific mission that you don’t quite understand. Make sure they’re okay with you following them around with a recorder. And when you’re trying to understand why they’re doing something, don’t just accept the first reason they give you. Watch them and develop your own theories about why they’re doing this. Run the theories by them and see what they say.

And remember, these kinds of stories take time. It’s hard to get someone to open up the first time you meet them. So maybe do the interview in stages. Meet the person a few times and they’ll be more comfortable each time. I met CT Lui three separate times — once before I pitched the story, once for the main interview, and one last time for follow-up questions, specific bits my editor was wondering about. Each time, I spent a few hours with CT. Again, I’m sure a more skilled reporter might be able to do this quicker. But it’s the part I enjoy the most and I’ve (sort of) made peace with my inefficiencies.

. . .

For every 8-minute story I made, I left behind hours of tape on the cutting floor. And some of that tape I could have used, had I been able to pitch these as longer features, say in the 20- to 30-minute range. I’m glad I didn’t though because I didn’t have a clue how to build the sort of structure that supports those stories: how to find the right question that would propel the listener through; how to motivate the turns; how to direct interviews to create scenes.

Just to be clear, I still don’t know how to do the above things by myself. I’m lucky enough to work at a show with weirdo editors who like taking risks, and smart, capable producers who can pull them off. When you’re pitching long-form narrative stories, you’ll have the support of the show that accepts the pitch. But you will have to first demonstrate to them that you absolutely know how to get tape and how to identify the right characters.

And on top of all that, you’ll have to know how to make stories that get to the heart of a question.

Scaffold Level 3: Tape + Character + Question

A few months after starting at Reply All, I heard about a website called Ripoff Report. A listener had written in, saying it was evil and destroying many, many people’s lives. It’s a place where you can post an online review of any company or person. And it’s impossible to get the review taken down. I did some perfunctory research and found that the owner of the website, a man named Ed Magedson, lived in a bunker and sued every journalist who profiled him. I emailed a go-between and asked him to ask Ed to call me.

Listen to “Sruthi Pinnamaneni”

My fun checklist of broken things:

  • My tracking is still bad! I’ve gone from artificial enthusiasm to deadpan. Why so serious, Sruthi?!
  • My writing doesn’t acknowledge or accentuate any of the tape’s humor. Ed hands me that ridiculous moat joke and what do I do in my narration?  Nothing. I change the subject.

Less-fun checklist of functioning things: Tape is good. There’s a character. (Almost too much character?) But what makes this a Level 3 piece for me is that despite all odds, I came up with a clear-ish question for the story and a way to answer it. Which was very difficult. Let me tell you what happened.

Before the trip I’d made a plan and had prepped with my editors. My task was to get Ed’s backstory and confront him about the negative effects of his website. But once I showed up for the interview, it all fell apart. Ed is very ADD (his words) and every time I asked a question, he responded with an impenetrable knot of tangents. He refused to turn off his eight cell phones (yes, he actually had eight cell phones) and took calls the entire time. I returned from my trip with two days of tape and zero hope of making this work.

I did however force myself to listen to all the terrible tape and some nine hours in, I realized I did have a question: how did Ed, a person with such good intentions end up with such a malicious website? And how does he not see how malicious it is?

I also realized that the terrible tape WAS the answer: that the ticks that made Ed impossible to interview were the same ticks that blinded him to the consequences of his website. That’s why he couldn’t see how he was hurting people. Here’s the end of the story we made:

Listen to “Sruthi Pinnamaneni”

It’s by far not the slickest ending, but it does the basic job, right? You feel like you understand Ed; you kinda like him. I wasn’t just telling you things about him, I was showing you the evidence in tape. I raised a question about Ed and I answered it. And that’s the last thing I want to talk about: how to handle the questions that motivate a story.

1. Make a story that creates the question.

When I find a character I love, I’ll often make up a question just to have an excuse to tell the person’s story. It’s a fake question like — what drives this person? Or how does the person do this thing? That works for a while. It’s actually liberating to be able to talk to someone without being too hung up on a specific question.

But at some point, when you start putting the story together, you need a real question that motivates a listener. You can’t assume the listener will keep listening just because you think this character is amazing. The story itself — the way it’s structured, the emotion in the tape — has to create a feeling of, “I don’t understand and I want to understand” in the listener.

I originally pitched the Ripoff Report story as a profile of Ed Magedson. Our editor at the time, Alex Blumberg, told me I first had to find a victim of the site, someone who’d been truly hurt by it even though they’d done nothing wrong. The search took months. There were tons of victims, but it was hard to find the person whose case seemed unimpeachable. And when I did, we put her right at the beginning, hoping her story would create the question that we’d answer through Ed.

2. Be honest with yourself.

So you have a question that you are sure is the engine of your story. But is it, really? Ask yourself that when you’re in edits. And be brutally honest.

Don’t listen to the draft as a nervous reporter, praying that all the parts are coming together as you dreamed they would. Listen with skepticism — as if you don’t know the story, you have a dozen other shows to listen to, and you’re deciding whether this one is worth your time. Are you bored? Are you listening out of a sense of duty? Or, are you into some parts of it — say, the tape and the characters — but the questions and emotions are all wrong. Like, listener-you has questions but they’re not the ones reporter-you is focused on.

It will take a while to force yourself to listen like this — you’re learning how to fluidly swap between reporter brain and editor brain, and that’s hard. But once you know how to do it, you can truth-test all the assumptions you’re making about why your story is interesting.

3. Don’t be afraid of breaking everything.

Here’s a thing I learned from my editor, Tim Howard. With every story I’ve worked on at Reply All, there’s a moment where things falls apart. Like, the story is almost done, we’ve fact-checked and I’ll call one more expert — just to be sure we’ve got it right — and the expert will tell me, sorry, this central premise is wrong. I throw up my hands in defeat; Tim rubs his in delight. I’ve never met a person more comfortable with wrinkles.

If that happens to your story, Tim would say: tell the listeners what happened. Try using that moment as a turn in the story. “We were sure this was the answer. But then this thing happened. I called Ms. X . . .” If that feels gratuitous, scrap that too. Most important, be honest with yourself. Do you feel as if your motivating question is no longer working? If so, dump it. Find the real human question, and then rebuild your entire story around it.

. . .

One last thing I want to say. I see people who are beginners in radio get ultra-focused on how to make a story sound perfect. Like, how to get the best audio quality, or track like a pro, or how to sound design. And I just want to stress that things like good tracking are just polish —  they help make the building pretty, but if that’s the only part you get right, the whole thing will collapse. For a story to work, it needs a foundation with heart — solid tape, a character you really care about, a question you want answered. If you can get those things right, the story will work, even if you screw everything else up.

And you know what? That feeling I had after failing to backwards construct Lulu’s gorgeous Radiolab story — disheartened and fully alone — I feel that way all the time. With every Reply All story I’ve made, there’s a moment where I think this one is beyond my skills. And then I remember:  don’t focus on the whole, just make the parts. You know how to make the parts. And then I get back to building.

Sruthi Pinnamaneni

Sruthi Pinnamaneni

Sruthi is a producer at Reply All. Before that, she worked on stories for The Economist, Studio 360, Radiolab, and Love + Radio. She’s also the mum of two loud boys. You can find her @sruthiri.


Your email address will not be published. Required fields are marked *


  • Jule Banville



    I teach audio storytelling at the University of Montana and use Transom ALL the time. But this piece blows me away at how amazingly, beautifully, simply Sruthi lays out the basics. Thank you for teaching me things!! I can’t wait to share this with everyone I know (and teach).

Your email address will not be published. Required fields are marked *