Nieman Foundation at Harvard
What’s in a successful succession? Nonprofit news leaders on handing the reins to the next guard
ABOUT                    SUBSCRIBE
Sept. 13, 2016, 11:47 a.m.
Business Models

And now a word from our sponsor: Host-read ads and the play between niche and scale

Podcasts have benefited from the unique intimacy of its ads. Can that strength survive the rise of programmatic and dynamic ad insertion? Part 2 of a five-part series on the business of on-demand audio.

“Right now is the moment in our podcast where we talk about our sponsor. Julia Turner, what have you got?” says Slate’s Stephen Metcalf. Twenty minutes into the weekly Culture Gabfest show, he is marking the start of a commercial as he turns to Turner, Slate’s respected editor-in-chief.

“The Culture Club Gabfest is brought to you by Club W. Club W is a club, as it sounds, that delivers wine right to your door,” Turner says. “They partner with innovative winemakers from all over the world to produce a wide variety of small-lot, handcrafted wines for their members. With Club W, you have the freedom to pick and choose the types of wines and the number of bottles you receive each month…The best part is you get a 100 percent guarantee on every single bottle…So take something off your to-do list. Just go to and take $20 off your first order now.”

The Gabfest is one of more than a dozen Slate podcasts. The company was in at the inception, launching its first podcast 10 years ago. Now Slate’s corporate sibling, Panoply — part of The Slate Group, owned by Graham Holdings — has become a key competitor in the new podcast network wars.

At first hearing, Turner’s commercial endorsement seems at odds with the old wall between business and editorial. What is Slate’s top editor doing pitching for a wine club?

Welcome to the odd world of podcast advertising. In the business, these are called “host-read” ads, a seemingly modern construct of the podcast revolution. In fact, of course, the form harkens back to the likes of Paul Harvey, that dean of radio “newsmen” in the latter 20th century — or, for that matter, to any baseball broadcaster who intersperses balls and strikes with pitches for Ford, Verizon, or Domino’s.

So would Slate have its editor promote products or services on itself?

Of course, not, Jacob Weisberg, chairman of The Slate Group, which includes both Slate and Panoply.

“I think if you did that on the site, it would create confusion. Julia writes articles. Why is she connecting with a product? It’s that thing where she’s endorsing Crest toothpaste — is that an ad or is that an article? I don’t think that would be clear. Which is the reason we wouldn’t do it.”

Weisberg puts the mild controversy — NPR for a long time wouldn’t allow host-read ads, though some have started popping up on some shows — into perspective.

“I think — and I always felt this in the magazine business too — that people forget the principles and get caught up in rules out of the principles.

“There are two principles,” he says. “Principle No. 1: The principle that you shouldn’t confuse readers about whether they are listening to an ad or an editorial.”

Principle No. 2: “The advertiser shouldn’t influence the editorial content. I don’t think host reads transgress on either of those at all. Everyone knows it’s an ad. Everyone knows they’re listening to an ad…It’s well-integrated into the program, and that’s why it’s effective and appealing, because it’s not interruptive the way a 30-second television commercial is interruptive. ‘Stop watching that and watch this.’ It’s coherent.”

Weisberg’s logic is compelling (and widely shared), if less than intuitive to many who have been around other media for awhile. He’s probably right: Podcast listeners won’t think terribly of the highly credible journalists suddenly pitching products. But no one’s really sure.

Certainly, the logic is a convenience that works for now. Host-reads on marquee programs fetch as high as $80 CPM rates, more than twice the price for which non-host-read ads will sell.

Those host-read ads — especially for “direct response” companies like Club W, MailChimp, Blue Apron, Squarespace, and Casper Mattresses — work well for those direct advertisers. Each proffered deal includes a URL that’s podcast-specific. Those allow direct measurement of ad results, providing return on investment data. Advertisers pay high rates simply because they work, returning a good return on investment for them. Those advertisers don’t need audience metrics. We respond to the “offer codes” and every day prove out the proposition that we like to try things our podcast buddies recommend.

For many companies, like Midroll Media and Panoply, host-read ads dominate all ad placements; Midroll’s Lex Friedman, who was just promoted to chief revenue officer, estimates 95 percent of all its ads get host renditions.

Host-read midrolls — played toward the middle point of podcasts — generate the $15-80 CPMs, while pre-rolls can start at $10 for more established shows.

Those are great rates, with much of network commercial radio going for $3 and digital advertising averaging in the higher single digits.

The growth in podcast advertising

Money is chasing this money, accounting for the startup frenzy we’ve seen within the past year. It is advertising’s siren call that drives this industry forward as its main revenue source. Subscription remains in its infancy. (We’ll explore it in part three of this series.)

Podcast advertising expanded at a 48 percent rate last year and is forecast to grow about 25 percent a year through 2020, at which time it would reach $395 million in ad revenue, says Bridge Ratings. Most in the industry consider Bridge an acceptable market sizer.

This year, Bridge pegs the total at $167 million. Though some in the industry think that may be high, most would believe a number of $140 million or more. Ask a digital media exec about that number and he may snort at its puniness. Currently, podcast revenue represents roughly one-third of 1 percent of all digital ad revenue sold in the U.S. But to an podcast optimist — or an audio-centric company that can add podcasting sales to its broadcast/digital packages — that’s just headroom calling.

In fact, as dynamic ad serving technology moves from introduction in the past couple of years to 2017 default, the amount of ad inventory and potential for targeting will multiply ad sales opportunities. Simply allowing fresh ads to be served in weeks- and months-old but still popular podcasts — from Car Talk to Alec Baldwin’s Here’s the Thing — could mean a 400 percent jump in ad capacity.

In retrospect, the old system will soon seem just plain silly.

Panoply’s Weisberg — whose company got into dynamic insertion earlier on with its Megaphone product — describes that old practice.

“In the first season [2014] of Serial, the ads were recorded — the host-read ads were recorded with the show. You make an estimate about how many dollars you’re going to have to get someone to buy those. Then, if you deliver one hundred ads, that’s a bonus. Ultimately, an ad market doesn’t work on that basis. You have to be able to buy finite quantities of things. With dynamic hosting, you can update the host reads with dynamic insertions.”

In other words, Serial — with its dramatic audience success — undersold its advertising.

Art19’s emergence in this space recently indicates the fast adoption of dynamic insertion within the country’s most popular podcasts. The San Francisco-based company began life in 2011, but didn’t see its business really pop until earlier this year.

In June, Art19 announced three new clients (Wondery, Midroll Media, and DGital Media), and last week The New York Times formally signed on for its growing podcasting business.

Dynamic insertion advances money-making the trade in three ways, all in ways that digital media veterans know well:

  • Allows on-the-fly insertion of new ads (host-read or otherwise) into evergreen programs, which today often still carry their original ads. If you listen to a three-month-old Serial episode, you won’t have to listen to a three-month-old ad.
  • Enables the serving of different ads to different listeners, first on the basis of geotargeting, a first application of the kind of digital demographic and geographic targeting that’s becoming standard.
  • Collects audience data, as podcast companies use the Art19, or Panoply, or other custom, audio player.

Panoply’s Megaphone, AdsWizz, PRX’s Dovetail, RawVoice/Blubrry, and Libsyn, among others, offer similar functionality, as do legacy products from long-time radio tech supplier Triton.

Ad targeting — again an echo of what we’ve seen in other parts of digital media — is just taking off in podcasting.

Upfronts mark podcasting’s ad emergence

While companies don’t break out podcast ad revenue, it’s believed that three companies — NPR, Midroll Media, and PodcastOne — each account for more than $10 million in sales.

“Podcasting will be vastly, majority advertising-supported,” asserts Hernan Lopez. Until January, Lopez oversaw 350 21st Century Fox international cable stations around the world, reaching 1.6 billion customers. He chucked that almost 20-year Fox broadcasting career in order to narrowcast, founding L.A.-based podcast startup Wondery in January; Fox is a seed investor.

Lopez sees himself getting back to his earliest media roots — and positioning himself for a windfall of advertising that awaits those who break out from the fat pack of startups now positioning themselves in 2016.

“My very first job was a cable and radio company,” he says. “This is back in Argentina, 27 years ago now. Cable advertising was in its infancy. That’s what audio on demand looks like now.” Wondery, like its peers, loves the direct-response business. But all agree: It represents just phase one of the business.

Last week in Time Inc.’s Henry R. Luce Auditorium in downtown Manhattan, twelve podcast companies further fueled the fire of podcast advertising. There, the Interactive Advertising Bureau (IAB) hosted its second annual Podcast Upfront. Each of the twelve presenters — NPR, ESPN, HowStuffWorks, Time Inc., WNYC Studios, AdLarge, CBS Radio’s, Authentic, Midroll, Panoply, PodcastOne, and Wondery — got 20 minutes to pitch current and new programs to marketers. Podcast talent Malcolm Gladwell, Alix Spiegel, and Michael Rapaport entertained the crowd.

The goal: Get beyond those direct-response advertisers who have so far built in the industry. Get to the big money — brand money — that so far has only stuck a toe into podcast advertising.

Many in the industry will tell you that as that brand money — think Procter & Gamble, General Motors, General Electric — arrives, podcasting’s future as a business is assured. It will then have moved from a pat-on-the-head nice little niche into a serious competitor for dollars, able to reach one hand into traditional radio budgets and another into the digital ad stream, which by next year will become the largest single source of ad revenue in the U.S.

“The thing that’s encouraging to me is [that big Fortune 100 brands are] not just bringing their radio tactics to” their podcast advertising experiments, says Bryan Moffett, head of National Public Media, which sells advertising for NPR across podcasting, broadcasting and digital channels. “They’re willing to listen and learn, and try to take advantage of the uniqueness of the format.”

Marquee brands want marquee top 25 podcasts, which for NPR include The TED Radio Hour and Planet Money.

“We’re getting some mid-six figure, upper-six figure deals around podcasting, so there’s real money being devoted to it,” says Moffett. “It’s not just $10-, $20-, $30,000 orders…I was here when we launched podcasting, 11 years ago, and we did a half-million dollar deal with Acura, and that was like groundbreaking at the time. Then nothing happened for eight years. Now we’re here again, back in the same place.”

Moffett says NPR will double its podcast revenue year over year, and can count a 10× increase over the last three years.

What will keep the momentum going?

Ask anyone about a roadblock to continued hyper-growth and they’ll offer one word: Downloads.

Lots of evidence tells us podcasting engages big audiences, but this is one medium that cannot be tracked as much as its digital cousin. The big issue: Podcast companies can generally get (often imprecise) download counts, but in most listening environments, the publisher has no idea if or for how long a downloader actually listened.

Given that fact, what now sustains the small industry — quite successfully, we must note — is partial data.

“The one thing that the industry lacks is clarity into where the scale is,” says Sarah van Mosel, chief commercial officer for Acast, a Stockholm-based podcast platform innovator. “So right now, you read a lot of trade articles about some of the startups that are getting a lot of press, but nobody really knows true audience levels. If I’m a marketer, and I want to reach millions of people at scale, I don’t know where to go.”

Van Mosel, an alum of WNYC, heads one of two Interactive Advertising Bureau committees, focusing of advertising and technological standardization. Those groups continue to work forward on standards, publishing last week a largely definitional document, which they hope will lead to greater standards. More than three dozen podcast company representatives have participated, marking one more step in the professionalization of podcasting.

At this point, though, there’s no Nielsen or comScore for actual listening. It’s all piecemeal. Stitcher and Google Play get good marks from some in the industry for supplying data for downloads they enable, but their market share are limited. Further, those publishers that offer their own players, or license the new one from Art19 for instance, of course, can know lots about listener habits in certain contexts.

Why hasn’t listening been better quantified? Apple’s dominant role is central, and there hasn’t been enough money in the industry for anyone to invest up to this point. That, though, may well change in 2017. Several sources tell me they expect a more comprehensive listening measuring product to emerge. When it does, says van Mosel, “I think that will be fairly transformative.”

For now, though, it’s all about extrapolation, as podcasting moves beyond what Lex Friedman calls its “artisanal” stage.

That extrapolation takes several forms, as ad sellers fill the data gap in several ways:

  • Show-specific surveys, both of listeners’ show habits and of their recognition of advertising. Ad surveys — done for a limited number of higher-spending advertisers — measure attributes like intent to purchase, brand preference and opinion, and recall of their messaging.
  • Learning from the successful experience of direct-response advertisers. After all, if audiences weren’t listening through to the podcasts, Squarespace and MailChimp wouldn’t be such major repeat customers.
  • Broader surveys of listeners, such as those collected by Edison Research, the primary audience tracker for podcasting.

“What respondents tell us in survey data is that, of the podcasts that they download, at least three quarters of them are listened to, and that number can go well into the high 80s,” says Tom Webster, Edison’s vice president of strategy and marketing.

Further, he says, podcasting listening is more and more resembling streaming. When listeners download, “The dominant paradigm for listening to a podcast is to click on it and listen to it immediately…I think detractors of the medium would posit that people click on a podcast to download it and then never listen to it. [That concern] is really fading into the background.”

Metrics then depend on the observer. Podcasting’s are far from digital’s, but they parallel TV’s. “You have surveys, and studies, and panels, and that’s kind of the same thing we have,” Moffett says.

Sheer scale

Advertisers do want better data, but they also want scale — the ability to reach bigger audiences with bigger buys.

Cable TV veteran Lopez recognizes that holdback. “[Podcasting] doesn’t have scale. One in five people listen to podcasts today. On a monthly basis, that’s 57 million people, but when you go to an advertiser that wants to reach a whole population, that doesn’t sound like a lot.”

One truism most believe: To get the bigger ad money, you have to be a big player. That’s why we see almost all the podcast companies bulking up.

Get Big Fast is the Silicon Valley religion, and so it is in podcasting today. Original content, they’ll tell you, is an essential part of the mix — but most believe that they won’t get big enough fast enough on expensive-to-produce programming by itself. So, they roundup lots of similar programming — programs that aim to bolster their brands, get noticed, and provide ad spots to sell.

In 2016, that means borrowing heavily from the market pushes of Netflix and Amazon. Digital video is the main business model being applied here.

In short, that both/and means doing two things well:

  • Aggregating as many programs under one big tent as quickly as can be done.
  • Creating originating programming that distinguishes a new podcast brand, a Panoply, a Wondery, a Midroll.

Wondery, in eight months of life, has put together a portfolio of 10 shows, and plans to add more rapidly.

Midroll offers a prototypical vertically integrated approach and has a huge appetite. E.W. Scripps bought the company last year and has funded its growth. Midroll produces 45 shows of its own and represents about 200 more in advertising. A year ago, direct response advertisers drove 70 percent of its advertising, with brands representing the rest, Friedman says. Within a year’s time, brands — including Dunkin’ Donuts, American Express, and Amazon’s Kindle — now drive 50 percent.

PodcastOne, signing up pop culture celebrity after celebrity, now offers more than 200 shows under its enlarging tent. It’s the yin to public media’s podcast yang.

Gimlet Media takes a contrarian path. “We want to build at the head of the tail,” says Gimlet CEO Matt Lieber, whose work with cofounder Alex Blumberg has been well documented on their own popular (and meta) StartUp.

“If you’re at the head of the tail, no matter how the ecosystem evolves, you’re more privileged in the way in which you can access distribution. Some people are trying to build the business off the long tail of podcasting. That’s not what we’re trying to do. What we’re saying is we want to build franchise shows.” So joining Gimlet’s six current shows, including Reply All and Science Vs., will be four more by the end of the year.

The Gimlet idea: Focus only on high quality rather than aggregation. Forty of its 55 staffers work on content, and its technology is wholly outsourced. Gimlet intends to be a brand that would open your ears when you hear its name spoken.

That’s how Jenna Weiss-Berman wants her Pineapple Street “podcast production house” to be thought of as well. Pineapple Street just launched its coproduction of the Still Processing podcast with The New York Times.

“We want to be in the top 100 or the top 10 of iTunes. I think in the end that actually is where the money is — it’s in the shows that are doing really well,” she says. And that’s the gospel she preaches to would-be partners. “There are all of these places now, media companies, that want to start 20 podcasts. And we are trying to tell them all is: Just start one really good one.”

Gimlet’s seen the value of branded content in digital media and funds a unit there. In its Gimlet Studios, five staffers work with Ford Motor, Dell, Lenovo, and PepsiCo, among others, and recently kicked off a campaign with HP. In that line of business, it follows Panoply, which has established itself as an early leader. Last fall, it created a major podcast success for GE, The Message, and now is moving more fully into branded programming. Panoply can be seen as an ambitious tweener here, a source of high-quality originals (Gladwell’s Revisionist History, Slate’s podcasts) and a branded content leader.

Brands, brands, brands. That’s what we hear on the commercial side of the house. And in that targeting, the startups have lots of competitions from incumbents NPR and WNYC.

Those organizations have got the ability to sell across podcasting, broadcasting, and digital properties, an advantage they’re now leveraging.

“We talk to these larger brands in a much more cohesive way, and put together plans that can complement each other across radio, digital, and podcast,” says NPM’s Moffett. In fact, they can target by time period, complementing radio’s drive time, for instance, with evening and weekend podcast listening.

Those selling multiple listening formats have noticed a crossover phenomenon.

WNYC CEO Laura Walker sees podcast’s rocketing growth showing up in her revenue numbers. “Fifty percent of our sponsorship revenue now comes from digital as opposed to terrestrial sources.” That’s a huge crossover point. Digital, of course, includes the apps and sites, streaming, and podcasts that ranks second in Podtrac’s latest podcast rankings, just behind NPR.

It’s podcasting sponsorship that is driving that digital takeover of the sponsorship line, she says. Walker still loves, of course, traditional syndication of a number of the station’s top shows (RadioLab, On the Media) through broadcast, but says that the advertising rates are higher — and seemingly more stable — on its 38 podcasts.

“I think our experience has been that it’s harder to get radio advertising sponsorship for shows that are on in different times throughout the country — like RadioLab or Wait, Wait, Don’t Tell Me — or not on consistently versus getting a podcast…It’s easier to monetize on podcast if you’re big than it is on national radio.”

All the podcast ad sellers deal with a buying market in flux. Who books podcast ads? It could be radio buyers or digital buyers. That’s a replay of the confusion around early ad buying, as print, digital, and broadcast budgets — and decision-making — have oh-so-slowly sorted themselves out.

Will advertisers seek audiences beyond the top 200 programs? What about the niches?

Says podcast veteran Eric Nuzum, who now heads Audible’s Channels effort: “My dad has two woodworking podcasts he listens to, and my mom has two quilting podcasts, and they’re in their late 70s.

Bernina puts out a new foot for a sewing machine, and my mom hears about it on this podcast — and this is like big news, right? People who care about this new foot for a sewing machine are listening to that podcast. We’re seeing more and more brands realize that. Bernina can say, ‘Why am I spending all these thousands of dollars to run ads, on magazines or through direct mail. I’m going to use podcasts for the fraction of the cost — and I’m reaching people who all own sewing machines.”

The perils of programmatic

Will podcast programmatic advertising increase the payout — or ruin the business? Programmatic — the automated data-matching between ad and audience — now influences more than a majority of all digital ads. It has become scarcity’s — and, often quality’s — foe.

Today, it is the “premium” sell that differentiates podcasts from the many other ways to reach audiences. Premium — in podcasting as in high-end digital and print media — connotes high-quality content, a usually affluent, educated target audience, and scarcity. Combine those three factors and media companies, podcasters included, hope to maintain higher-than-programmatic rates.

But programmatic may increasingly threaten “premium” — if podcasters end up feeling compelled to use the tech, as their digital publishing cousins did. If advertisers can use lower-cost programmatic advertising to reach the hip listeners of American Public Media’s Dinner Party Download or new moms taking in The Longest Shortest Time, or Recode’s tech sophisticates, will they keep paying relatively high “premium” ad rates?

Midroll’s Friedman got into the game early, selling his own podcast. He knows that programming will enter the picture, but he hopes it won’t dominate, and ruin, it.

“There are shows that are going to do that. There are shows that are going to stick with inserted, injected, prerecorded spots. They’re going to stick with programmatic.

“Programmatic is a lot easier — costs get a lot lower. But that’s exactly the problem, right? Because it’s a race to the bottom on price. The ads are much less effective, so advertisers aren’t nearly as impressed. It becomes just yet another thing, versus something really impactful. And listeners hate it, right? For listeners, it’s really horrible, because it turns podcasts into radio. “It’s the worst part of radio where it’s like stuff that’s jarring and interrupting of your experience, takes you out of the moment and it’s totally unpleasant to listen to.”

Tomorrow in Part Three: Are you ready for a Netflix for podcasts?

The research for this series was commissioned by Public Media Futures Forums, a project funded by the Wyncote Foundation to support in-depth analyses of strategic challenges facing public media. The analysis was done independently and without any foundation review.

Photo of microphone by Alex Indigo used under a Creative Commons license.

POSTED     Sept. 13, 2016, 11:47 a.m.
SEE MORE ON Business Models
PART OF A SERIES     The newsonomics of podcasting
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
What’s in a successful succession? Nonprofit news leaders on handing the reins to the next guard
“Any organization that is dependent on having a founder around is inherently unsustainable.”
Worldwide, news publishers face a “platform reset”
Some findings from RISJ’s 2024 Digital News Report.
The strange history of white journalists trying to “become” Black
“To believe that the richness of Black identity can be understood through a temporary costume trivializes the lifelong trauma of racism. It turns the complexity of Black life into a stunt.”