Is This Thing On?

Last October, several dozen audiophiles gathered in a basement auditorium for an all-day conference about “the future of radio in a digital age.” Reddit founder Alexis Ohanian finished a talk he’s been giving to college campuses about the Internet and the transformative power it can unleash when it mobilizes a mass of people around an idea, a video, a website, a tweet. When he took questions, I asked: Why does the Internet so rarely mobilize around audio? What would it take to put audio on the Reddit front page?

Ohanian leaned back, contemplating the question, apparently for the first time. “That’s interesting,” he said. “I’m thinking of a lot of the viral content.” You could practically see the memes and GIFs pass across his brain. He started to point out that most viral videos are under three minutes, while the best audio storytelling was usually longer, but interrupted himself with a story about Upworthy.

When the founders pitched him on their plan â€" to make “socially good content” “go viral” â€" Ohanian invested “out of passion,” not because he thought it would work. Now Upworthy is one of the fastest growing media properties on the Internet. Sure, sound may not go viral today, but Ohanian is optimistic. “Probably someone here in the audience is going to show us all wrong,” he said, “and a year from now we’re going to look at the Upworthy for audio."

“So go make it.”

Easier said than done.

Cat Video Vs. The Cat’s Meow

Bianca Giaever has always been obsessed with radio. As a child, while she biked her newspaper delivery route, she listened to an iPod loaded exclusively with episodes of WBEZ’s “This American Life.” At Middlebury College, she stalked her classmates, dragging them to her dorm room to record interviews she edited into stories for the college station and smaller audiences online. “I was fully planning on working in radio,” she says. “My whole life.” That is until, the day after graduation, she became a viral video star.

When she painstakingly crafted moving audio narratives, her parents and brother listened. When she added video to her final college project, “The Scared is Scared” â€" a 6-year-old’s dream movie brought to life â€" “It just. Blew. Up.”

“At first I was like, ‘Wow. A lot of people are sharing this on Facebook,’” she recalls thinking, “‘I have such nice friends!’” Then it was friends of friends. Then strangers. By the time websites like Mashable and CBS News picked it up, she could only picture the audience as a number. Waiting on the tarmac for her post-grad vacation to begin, she watched on her phone as that number spiked into the thousands, then hundreds of thousands, seemingly crashing the site that hosted it. “These French people were yelling â€" because I had my phone on as we were taking off â€" that I was going to kill them,” she recalls. “They were like, ‘Is whatever you’re doing worth our possible death?’ And I was like, ‘Maybe? This is the biggest thing that’s happened in my life!’”

Of the 100 most-shared news articles on Facebook, three were from NPR, but none included audio. Two of these were reblogs of YouTube videos.

I’m a public radio reporter and this doesn’t happen in my milieu. There is no Google Sound, no BuzzFeed for audio, no obvious equivalent of Gangnam Style, Grumpy Cat or Doge. If you define “viral” as popularity achieved through social sharing, and audio as sound other than music, even radio stations’ most viral content isn’t audio â€" it’s video. A 17-minute video interview with Miley Cyrus at Hot 97 has nearly 2 million views. An off-the-rails BBC Radio 1 video interview with Mila Kunis: more than 12 million. In June 2013, the list of the 100 most-shared news articles on Facebook included three from NPR, but none included audio. Two of these stories were reblogs of YouTube videos (this one and this one), found on Gawker and Reddit.

“Audio never goes viral,” writes radio and podcast producer Nate DiMeo. “If you posted the most incredible story â€" literally, the most incredible story that has ever been told since people have had the ability to tell stories, it will never, ever get as many hits as a video of a cat with a moustache.”

It’s hardly a fair fight, audio vs. cat video, but it’s the one that’s fought on Facebook every day. DiMeo’s glum conclusion is an exaggeration of what Giaever reads as the moral of her own story: “People will watch a bad video more than [they will listen to] good audio,” she says.

Those in the Internet audio business tend to give two explanations for this disparity. “The greatest reason is structural,” says Jesse Thorn, who hosts a public radio show called “Bullseye” and runs a podcast network called Maximum Fun. “Audio usage takes place while you’re doing something else.” You can listen while you drive or do the dishes, an insuperable competitive advantage over text or video, which transforms into a disadvantage when it comes to sharing the listening experience with anyone out of earshot. “When you’re driving a car, you’re not going to share anything,” says Thorn.

The second explanation is that you can’t skim sound. An instant of video is a still, a window into the action that you can drag through time at will. An instant of audio, on the other hand, is nothing. “If I send someone an article, if they see the headline and read a few things, they know what I want them to know,” a sound artist and radio producer told me. “If I send someone audio, they have to, like… listen to it.” It’s a lot to ask of an Internet audience.

For some radio makers, social media incompatibility is a sign of countercultural vitality. Thorn has called his own work “anti-viral,” and believes that entertaining his niche audience is “still so much better than making things that convince aunts to forward them to each other.”

“That’s A-U-N-T-S,” he clarifies.

But when I suggest the situation doesn’t seem to concern him, he interrupts, “To say that it doesn’t concern me â€" it concerns me profoundly. I think about it all the time.” In his view, social media warps our consumption patterns, and not for the better. “It’s a serious problem in my life. And not just in my media-making life, in my day-to-day life.”

After Giaever’s video went viral, she turned down an internship at “This American Life” â€" “my dream since I was nine” â€" to become a “filmmaker in residence” for Adobe. She gets paid to make her own movies, which she still approaches as radio stories with added visuals. It’s the proven way to get people on the Internet to listen. “The entire concept of what I’m doing seems problematic to me,” she says. “What’s so beautiful about radio is you can’t compete with what people are imagining in their heads, right? And yet I still continue to do it.”

Because audio doesn’t go viral.

Except that sometimes, it does.

Kids Say The Darndest Things

Most viral audio wasn’t intended for the Internet. Recordings made for some other purpose are excerpted and uploaded: voicemails, speeches, and calls to 911 and customer service hotlines.

One category of viral audio is the document, bits of audio that serve as evidence in a news story. It’s easy to imagine text transcripts being distributed in audio’s absence: Bradley Manning’s testimony, the 911 calls of the Trayvon Martin case, Obama’s oft-quoted “clinging to guns and religion.” The primary advantage of audio over text is that it lets the listener confirm a quote with her own ears and determine if meaning is altered by nuances of emphasis or emotion.

Another category of viral audio is the rant or comic diatribe, where emphasis and emotion are the entire point. For instance, an irate San Francisco Chronicle reader chewing out the editor for referring to a “pilotless drone,” or a voicemail becomes an increasingly laugh-filled narration of the aftermath of a car crash. A transcript of these would be like lyrics without a melody.

Somewhere in between these two is a subcategory that could be called “celebrities gone wild”: Alec Baldwin cursing out his 11-year-old daughter, Christian Bale cursing out his director of photography, Mel Gibson cursing out his ex-girlfriend, etc.

These brief, emotional, sometimes-newsworthy clips of people speaking have cousins in viral video. In fact, the two are sometimes difficult to distinguish. Mitt Romney’s infamous “47% comment” was captured and distributed as a video featuring blurry donors’ backs. A recent viral “video” titled, “Potty Talk! [Original] 3 year old contemplates the effects of his diet on the toilet” is merely a shaky shot of a bathroom door. When documenting a primarily auditory event from the vantage point of a single recording device, adding a video camera to the microphone gives slightly more information, and the advantage of keeping the eyes occupied.

But these amateur, one-shot videos are a small and shrinking section of the viral video pool. “We’re seeing a lot more professional work in [the viral video] space, and I don’t just mean advertisers,” says YouTube trends manager Kevin Allocca. The “top trending videos” of 2013 were all intentionally shot and edited for an Internet audience: music videos (“What Does The Fox Say?”) and ads (Volvo’s “epic split” with Jean-Claude Van Damme) but also low-budget productions like the Norwegian army’s “Harlem Shake.” They’ve all have over 90 million views.

Analogous audio â€" deliberately constructed and virally distributed â€" is a rarer and more recent phenomenon.

Ask a public radio journalist for an example of viral audio, and one piece comes up again and again: “Two Little Girls Explain The Worst Haircut Ever.” It’s two minutes and fifty seven seconds of cute, as five-year-old Sadie and three-year-old Eva tell the story of an ill-advised haircut to their patient interviewer and father, WCPR reporter Jeff Cohen. For public radio, Cohen has covered gangs, unemployment, and the aftermath of the mass shooting at Sandy Hook Elementary school. He won a magazine writing award for a story in the Hartford Courant about Connecticut’s first Iraq war widow.

“I’ve done a lot of work as a reporter that I’m pretty proud of,” he says. “I will never be recognized for anything for the rest of my life, except for this.”

It, too, resembles a viral video: it’s short, self-contained and driven by cute children. But not only does it lack any images of said children, it isn’t a straightforward record of what unfolded in front of the microphone. Cohen recorded two interviews, one with each daughter, and then carefully edited them into a fast-paced, seamless whole. Unlike Alec Baldwin’s voicemail, “Two Little Girls” is a showcase of audio’s power to create what appears to be an unedited version of reality, but is in fact a tightly constructed story, with a beginning, middle and end.

To explain why millions of people have listened to “Two Little Girls” â€" and why this is still so exceptional â€" you have to look at its convoluted path to fame.

What We Mean When We Talk About ‘Viral’

Taken literally, “viral” brings to mind an infectious agent bumping around inside its host, spreading accidentally by breath or touch. When “viral marketing” emerged in the 1990s, the medical referent was apt. The disease vector typically took the form of email and “virals” â€" as such ads were then called â€" that lived in the inbox. Invisible to the wider world, they spread from individual to individual, as when Hotmail stuck a sign-up ad beneath its users’ signatures. Or when the movie “American Psycho” sent compulsively forwardable emails from its psychotic main character, Patrick Bateman.

Today, those seeking to “go viral” have the same essential goal â€" to increase their audience by reaching the audience’s audience (and their audience, ad infinitum) â€" but the web has changed beyond the dynamics of disease transmission. Instead of invisible, one-to-one emails, today’s Internet infections spread by a cascade of publicly visible, one-to-many “likes,” “shares,” “tweets,” and “reblogs,” accelerated and amplified by an expanding web publishing industry. “Sharing” implies a deliberate effort, but social media sharing skews toward a mix of self-representation and what Tumblr creative technologist Max Sebela refers to as “speaking in content”: You might share Rebecca Black’s “Friday,” not because you want people to watch the video, but to make a joke about the fact that today is Friday.

“How does it happen,” YouTube’s Kevin Allocca asked in a 2011 speech called “Why Videos Go Viral.” “Three things: tastemakers, communities of participation, and unexpectedness.”

Tastemakers are like virus broadcasters, picking up outstanding, or “unexpected,” Internet phenomena that might otherwise never spread beyond their initial communities, and spraying their spores onto larger followings.

For Cohen’s “Two Little Girls,” the key tastemaker, without whom it may well have languished in Internet obscurity, was Gawker’s Neetzan Zimmerman. (Note: I spoke with Zimmerman before he announced his plans to leave Gawker to become editor-in-chief of a social network startup called Whisper.)

Zimmerman is the closest thing to a one-man embodiment of what he calls “the viral industry.” When Gawker hired him in early 2012, his boss A.J. Daulerio approvingly called him, “a total freak” for his ability to methodically scour the corners of the Internet for the video, memes, and Internet ephemera that would grow to popularity after being seeded with Gawker’s audience. “Before I used to do basically 20 hours a day,” Zimmerman says. “Now there’s a night shift, so I don’t have to worry as much.” In the last three months of 2013, his posts were responsible for more than half of Gawker’s pageviews and two thirds of the site’s unique visitors â€" nearly 40 million in total â€" according to Gawker’s public stats. For comparison, that’s more than 1/3 of the traffic of the entire the New York Times website.

Zimmerman’s work is a more extreme version of the new, upside-down dynamic of web publishing. Instead of the publisher’s megaphone guaranteeing its articles an audience, the publisher only has an audience insofar as the articles “go viral.” Tens of thousands of readers see most of the dozen items Zimmerman posts each day, but millions see his blockbusters.

For those hits, the content and the clickbait headline are as important as the timing. He describes “going viral” like surfing: boarding a wave at the earliest possible point. “You don’t want to wait too long because you’ll miss that initial cresting,” he says. “It’s a race against everyone else.”

Zimmerman chooses what to cover by scanning for signs of that wave rather than looking deeply at the constituent molecules of content. “The way the system works is I keep a mental note of instances of occurrence on a certain tier of sites,” he says. This lets him identify “viral momentum,” even when his personal judgment might suggest otherwise. “The purpose of the system is to override my biases and to override whatever personal feelings I have.”

Sometimes this lets Zimmerman not only beat the competition, but also popularize something that might otherwise never bubble into the mainstream from a less-trafficked corner of the Internet. But the system â€" Zimmerman’s and that of the “viral industry” more generally â€" has an obvious bias of its own toward content that is already being shared on the Internet.

For Bianca Giaever’s “Scared” video, first college and radio friends shared it on Facebook, then Vimeo made it a “staff pick,” then major media websites like CBS News, BuzzFeed, Jezebel and Mashable blogged about it. Within three days, hundreds of thousands were watching.

For Cohen, it took four months, and a lot of luck.

‘Invisible As the Radio Waves Themselves’

Jeff Cohen had interviewed his daughters many times, in the same way other fathers shoot home videos. “I’m sappy that way,” he says. But he thought enough of the haircut piece to play it for colleagues at the radio station. “It was about five minutes long, and my boss and friends said, ‘Cut it down to three minutes and put it on PRX.’”

PRX is the Public Radio Exchange, and as the name suggests, its website is a marketplace where station managers shop for stories. After Cohen uploaded his new, tighter version of “Two Little Girls” in February of 2012, it was discovered and licensed by a handful of local stations: KOSU in central Oklahoma, KUT in west Texas, KSJD in southwest Colorado.

But to the Internet, all this was invisible as the radio waves themselves. “PRX is designed as a business-to-business marketplace,” says PRX CEO Jake Shapiro. “We’re not designed for listeners… yet.”

The circuitous route that “Two Little Girls” took to Gawker didn’t start with PRX, but at a monthly event called “Ear Cave” hosted by one of Cohen’s colleagues at a coffee shop in Hartford, Connecticut. “I call it BYOB, BYOE,” says the event’s creator Catie Talarski. “Bring Your Own Beer, Bring Your Own Ear.” She dims the lights, sets up chairs, and projects a photograph of an old radio, so the audience has something to look at while a chosen curator presses play on a laptop. That April, “Two Little Girls” was the grand finale.

“It was just a huge hit,” recalls Adam Prizio, an insurance auditor who was in the audience that night. Two months later, Prizio, with the voices of Eva and Sadie bouncing around his head, decided to google it. Finding the audio on PRX, he posted a link to community blog MetaFilter, with no description other than a mysterious quote (“It happens three times in every life. Or twice. Or once.”) and the categorization “SLAudio,” a riff on “SLYT” (Single Link YouTube).

Overnight, the comments swelled. “Amazing.” “Adorable.” “Better than the Car Guys.” “OH MY GOD THIS IS FUCKING BALLER.” There were fewer comments than a link published ten minutes later â€" “Fundamentalist Christian schools in Louisiana will soon be citing the existence of the Loch Ness monster as proof that evolution is a myth” â€" but they were comments of single-minded delight. The next morning, Zimmerman saw the thread in his morning Internet regimen, and within an hour had put up his own post that would go on to gather some 1.3 million views entitled, “Public Radio Reporter Interviews His Two Little Girls After One Gives the Other the ‘Worst Haircut Ever.’”

“It didn’t really matter that it was audio,” says Zimmerman. “It was more about how it was being received online.”

In one sense, it followed the same trajectory as all viral content, or what YouTube’s Kevin Allocca has defined as a combination of “community participation” and “tastemakers.” Something becomes popular in a niche community, whose public enthusiasm attracts the notice of a tastemaker, who then repackages it to suit a larger audience, where the entire process repeats on a larger scale.

But really “Two Little Girls” succeeded in spite of its immediate community. Cohen first had to be convinced to put it online at all, and even then it was on a website searched only by public radio station managers. While Cohen says it made the rounds of his Facebook friends, it only took off after audio enthusiasts heard it at a coffee shop.

Compared to other media, even young, tech-savvy audiophiles are less likely to share audio on a weekly basis, and when they do, they’re more likely to use email instead of social media.

The barriers that nearly blocked “Two Little Girls” from finding a larger audience are a mix of culture and technology. While home videos make the leap to YouTube all the time, audio makers tend to keep their scraps to themselves. When I took an unscientific poll (n=60), it backed up what I heard anecdotally: Compared to other media, even young, tech-savvy audiophiles are less likely to share audio on a weekly basis, and when they do, they’re more likely to use email instead of social media.

Several echoed the sentiment of occasional radio producer Laura Griffin, who said, “I tend to assume that most people don’t have the same patience and appreciation for audio that I do, so I am selective about what audio I share and with whom.”

Others pointed to technological limitations. The files themselves are large and often forbid downloading. Audio-hosting websites employ an inconsistent potpourri of players, many of which disallow the embedding that has helped make online video ubiquitous. (Some PRX audio can be embedded, but Gawker had enough trouble with its player that they uploaded the audio into their own.) “I often don’t share NPR audio because their player isn’t embeddable and requires going to another website to listen,” notes multimedia producer Will Coley.

There is one standard format for distributing digital audio, but rather than resolving these barriers to sharing, it may be their most perfect expression: the podcast.

The Podcast Problem

If you don’t know what a podcast is, you’re in the majority.

Technically, it’s an RSS feed containing links to files (“podcast” typically implies an audio file). Using podcast-listening (formerly “podcatching”) software, you can “subscribe,” setting your computer or smartphone to automatically download the new and get rid of the old.

It’s hard to appreciate in 2013 the enthusiasm with which this simple idea was met by the mid-2000s media.

“I haven’t seen this much buzz around a single word since the Internet,” computer programmer Carl Franklin told the New York Times in 2004.

By letting everyone become broadcasters (or really “podcasters”), it was supposed to disrupt radio in a way that was predicted to parallel that other online media format with a horrible portmanteau name: blogging. In fact, the name “podcast” was tossed off by the Guardian's Ben Hammersley between the alternatives “audioblogging” and “GuerillaMedia.”

It wasn’t all hype. Anyone can start a podcast, just as anyone can blog. The podcast did close the loop, in its clunky way, between where people download and where they typically listen. And aficionados can point to a long list of programs, especially covering technology and â€" more recently â€" comedy, which never would have existed otherwise.

12% of Americans listened to a podcast in the last month, the same percentage as three years ago.

But while much of online publishing now takes the form of the blog, interest in podcasting seems to have flatlined. According to Nielsen Audio (formerly Arbitron), 12% of Americans listened to a podcast in the last month, the same percentage as three years ago. It is a substantial niche, but smaller than the percentage of people who create online videos, and less than a sixth the number who watch them.

“There was a huge wave of initial excitement around podcasting changing and disrupting and turning upside-down radio seven years ago, or longer,” says PRX’s Jake Shapiro. “And then it kind of just petered out.”

While the number of podcasts has proliferated, the vast majority of episodes have audiences in the double or triple digits, judging from the experience of podcast hosting giant Libsyn. “If you want to do the average, our mean podcast? Now you’re looking at like 200, 250 downloads per episode,” Libsyn’s Rob Walch told NextMarket Insights's Michael Wolf. The majority of top podcasts, far from being grassroots disruptors, are major public radio shows: “This American Life,” “Wait Wait Don’t Tell Me,” and “Radiolab.” It’s the dominant way of finding an on-demand audio audience on the Internet, but it’s more Hulu than YouTube.

The absence of disruption is, in part, baked into the technology. “It’s clearly the number one barrier to wider listenership,” says Jesse Thorn. Apple gave the format a big boost when it brought it into the iTunes store in 2005, but that walled garden of a market has come to delimit the podcast’s reach. To watch a YouTube video, you click play, wherever it exists on the web. With another click you can immediately share it by putting a player in the feed of your Facebook, Twitter, Tumblr, or even LinkedIn accounts.

To listen to a podcast, however, you have to search for it on an app or in the iTunes store, sign up for it, wait for it to download. (Of course there are other ways to download podcasts, but the majority of podcast downloads occur through Apple.) Click “share” on Apple’s podcasting app, and you’ll be prompted to post an RSS feed, which is a bit like trying to share a new Tom Junod article and instead passing on a password that readers can use to subscribe to Esquire.

These hurdles don’t hamper podcasts that are already well known. Thorn’s podcast audience has been growing steadily by approximately 50% each year. “Radiolab” and “This American Life” â€" public radio shows that are among the most popular podcasts and the aesthetic guiding lights for young public radio producers â€" are both approaching a million digital listens for each new episode. For these shows, the occasional episode will get shared more than others, but that “viral” bump is on the order of 10 to 20 percent, and even that seems driven less by social media than old-fashioned word of mouth. “Google is a much bigger referrer to any given episode [than Facebook],” says WNYC’s Jennifer Houlihan Roussel. In other words, podcasts don’t go viral. Nor are they designed to.

As the Guardian’s technology editor, Charles Arthur, points out in the Independent back in 2005, “Podcasts take content and put it into a form that can’t be indexed by search engines or be speed-read, and which you can’t hyperlink to (or from). A podcast sits proud of the flat expanse of the Internet like a poppy in a field. Until we get really good automatic speech-to-text converters, such content will remain outside the useful, indexable web.”

A Cloud Atlas?

If there is any company attempting to create a modern web alternative to the podcast, it’s SoundCloud.

“Podcasting: It’s a fairly old school method of distribution,” says its co-founder and CTO Eric Wahlforss. “We are certainly of the opinion that SoundCloud is the superior way of broadcasting your show across the web.”

If you’ve played audio from Facebook, Twitter or Tumblr, you’ve likely seen it: the slow crawl of orange across a gray waveform. This omnipresent, embeddable player is what has most clearly attracted the moniker “YouTube for audio.” Hoping to make sound as sharable as video, SoundCloud delivers this content via a streaming player instead of a dressed-up file download.

In a Facebook message, data scientist Lada Adamic told me: “Soundcloud does seem to have a lot of sharing activity (everything is dwarfed by YouTube but soundcloud is holding its own) [sic].” SoundCloud was the 11th most commonly submitted domain on Reddit as of March 27, 2013, according to Reddit data scientist Chad Birch, above the Huffington Post, the Guardian and Vimeo. The number of YouTube domains submitted was almost 22 times as high.

But the SoundCloud content accumulating most on social media isn’t what the company calls “audio.” “In our world, in terms of viral content, the real viral content is actually music,” Wahlforss says.

For non-music “audio,” SoundCloud lets broadcasters and podcasters have it both ways, encouraging them to make their shows available on SoundCloud’s platform, while also creating a podcast-ready RSS feed. “We are trying to blur that distinction a little bit,” says Wahlforss.

“We’re on SoundCloud because they have a nice player for sharing on Facebook and Twitter,” says Seth Lind of “This American Life.” But the total plays of their hour-long episodes on SoundCloud peak at roughly 3% of its digital listenership, and are usually under 1%, hovering around 5,000. A look at SoundCloud’s “trending audio” page presents a similar picture: podcast episodes and radio shows, with listenership in the hundreds or low thousands.

Clearly, technology alone doesn’t ensure the virality of an hour-long show with a headline designed for consistency rather than clickability (e.g.: “#513: 129 Cars” from “This American Life”). “It’s probably not going to be as popular as a Gangnam Style,” Lind notes, dryly. The audio that has gone viral takes a different tact: short, tailored specifically for SoundCloud, and providing a near-immediate pay-off that fulfills the headline’s promise.

Much of it is some mix of rant and newsworthy document, like AOL’s Tim Armstrong firing Patch’s creative director, or Charles Ramsey’s 911 call after he helped rescue three kidnapped women in Cleveland.

But the most heard, and most truly social example of SoundCloud’s viral audio is a New Zealand radio host’s dramatic reading of a series of text messages from a one-night stand gone unhinged: “This Is What Crazy Looks Like Via Text Messaging.” “Fletch & Vaughan” host Vaughan Smith found the texts on BuzzFeed and performed them as part of a four hour-long drive-time show. He then uploaded it to SoundCloud and shared it on Facebook to appease callers who wanted to hear the skit â€" but only that one skit â€" again.

“At the end of the weekend it hit a million plays,” says Smith. “It was mental.” With more than six million plays to date, more people have heard the version from “Fletch & Vaughan” than have read the BuzzFeed article it was adapted from â€" a triumph of sound over text.

It couldn’t have gone viral without a player as sharable as SoundCloud, but perhaps more importantly, it couldn’t have gone viral without the active unearthing of comedic gold buried within a longer broadcast. “In public radio, only within the last few years has there been a big value seen in disaggregating content from shows,” says PRX managing director John Barth. “And there’s still a pretty big debate about that.” These concerns echo the now-largely-obsolete resistance of other media to the Internet. They want listeners to experience the whole enchilada, not take the ingredients and re-contextualize them.

As for creating a whole new audio cuisine â€" work cooked up specifically for a SoundCloud audience â€" the successful examples are elusive. “We mostly use it as a promotional tool really,” says Smith. “We use it to promote the podcast.”

The Message Is The Medium

Last October, Reddit's Alexis Ohanian told a basement full of audiophiles to go make "the Upworthy for audio," but in a sense, we already have the Upworthy for audio: Upworthy. With its scientifically-selected, clickbait headlines, it  is the reason nearly two million people have heard the future president of Ireland Michael Higgins dress down rightwing talk show host Michael Graham (“A Tea Partier Decided To Pick A Fight With A Foreign President. It Didn’t Go So Well.”) It’s the reason hundreds of thousands have heard Geoffrey Gevalt tell a small poignant story, set to music, about his daughter (“A Toddler Gets Totally Profound In a Way Most Adults Don’t”) and Summer Puente about her father (“Every Night This Dad Falls Asleep in Front of the TV. There’s a Beautiful Reason Why.”)

The Upworthy sector of the Internet economy isn’t just healthy, it’s insatiable and omnivorous in its appetite for content it can coax people into clicking and sharing. “Whether it’s audio, whether it’s video, whether it’s still images, whether it’s text: my system remains pretty much the same,” says Neetzan Zimmerman. “For me it doesn’t really matter.”

The viral industry can help solve audio’s skimming problem, but only if it can find the content in the first place. “Radio doesn’t do a very good job of marketing itself to the viral industry, for whatever reason,” says Zimmerman. “Maybe it thinks too highly of itself, or thinks of ‘viral’ as a cheapening of its content. I really disagree with that. I think there’s a lot there to be mined, and a lot that gets ignored.”

“Marketing” makes it sound like radio makers simply need to do a better job of drawing attention to their work. And it’s true: active, public sharing directed at non-audiophiles is how Zimmerman found “Two Little Girls.” If there were a website that showed what audio was “trending” in some smaller community, Zimmerman says it would become part of his system. “One hundred percent. No doubt about it.”

There are also plenty of short podcasts and single-serving radio stories that are poorly labeled on obscure web pages or presented in unembeddable players. “Nobody that I’ve seen, even the best of them, spends time thinking about how to create the metadata or the descriptions: the things that might actually catch your attention,” says PRX’s Jake Shapiro.

More fundamental than marketing is the question of where audio makers see a market. “So far nobody is producing audio, really, for an audience that might be scanning for things to enjoy,” says Shapiro.

“It’s somewhat of a chicken and egg thing,” he says, “Until producers have any kind of confidence that there’s an audience or some money to be made â€" or preferably both â€" they’re not really targeting it.”

“If it can’t be used for pornography it’s never going to be the most popular thing.”

Perhaps Facebook will tweak its algorithms to favor audio. Perhaps SoundCloud or PRX or Apple will make a social alternative to podcasting. “It’s possible that someone will make this app that’s all about sharing audio that will be the next Snapchat,” suggests Seth Lind. “That’s obviously not going to happen,” he quickly adds, to make sure I know he’s joking. “If it can’t be used for pornography it’s never going to be the most popular thing.”

But Jeff Cohen and “Fletch & Vaughan” demonstrate that audio makers don’t have to wait for a deep shift in technology to court a viral audience. They would, however, have to create audio not for already-dedicated radio and podcast listeners, but for the distracted, impatient crowd that is the web. Audio enthusiasts would have to evangelize on that work’s behalf, not just in coffee shops or emails to each other, but online, loudly, with the same manipulative, click-chasing techniques wielded by the rest of the web.

The day “Two Little Girls” went viral, Jeff Cohen tweeted: “I fear I may disappoint new Twitter followers once they realize that I mostly write on Hartford, government, and healthcare. Not my kids…” That is still more or less his beat, though he does also have a children’s book (“Eva and Sadie and the Worst Haircut EVER!”) due out this summer.

“I don’t know anything about the Internet, really,” Jeff Cohen says. But the way he sees it, although he got lucky, he also made his own luck.

“I didn’t cut anybody’s hair. But when you see an opportunity, you take advantage of it.”