A Newsletter about Big Ideas in Music and Technology, by Cherie Hu
This is issue #55, published on June 27, 2019
Happy Thursday everyone! :) It's finally feeling like summer in New York, in all its glorious humidity.

A few points before diving into this week's topic:

👉 Episode 8 of the Water & Music podcast, featuring MAEKAN's Eugene Kan and Charis Poon, is now out on Spotify, Apple Podcasts, Google Play, Stitcher, TuneIn, RadioPublic, Overcast and many other of your favorite podcast platforms. We dive into many of the themes that Eugene and Charis covered in their essay "The Modern Creator's Paradigm," and what implications they have for the music industry with respect to creative and algorithmic accountability.

👉 This August, I'll be speaking at the Hudson Valley Music Summit in New York and the All About Music conference in Mumbai, India. It's a bit ahead of schedule, but let me know if you'll be around either event or city (particularly Mumbai / elsewhere in India, as I'm starting to plan a few weeks worth of traveling around the country).

👉 A special thank-you to Jeffrey Bruce, Mayur Chikramane, Dele Kadiri, Iredumare Opeyemi and Shara Senderoff for joining the $40/month tier on my Patreon page in recent weeks! Really appreciate your support :)

Now, let's dive in...
Streaming services finally care more about old* music

Last month, I posted on Twitter about a major disconnect I’d been pondering for a while:

The reaction to this tweet was both engaged and mixed. Some followers posited that there remains a significant opportunity for legacy catalog to embrace digital, frontline-leaning marketing strategies more openly, e.g. meme-ing old AC/DC records. Others countered that labels still need to invest first and foremost in marketing and developing frontline artists — otherwise their songs won’t become viable, money-making catalog in the long term.

One argument that I want to focus on today, and by which I still stand, is that streaming services are relatively late to the game when it comes to supporting catalog marketing.

Despite having a reputation for prioritizing new music discovery, the majority of streams on services like Spotify, Apple Music and Amazon Music still come from older rather than newer music. The mechanics of streaming also tend to significantly increase the lifespan of a given song and help produce a long-term monetary windfall for older music that was rarely possible in an era of transactional sales. It was only within the past year or so that these platforms’ internal organizations and external feature sets finally began to attempt addressing these trends, in ways that could potentially benefit both major and independent artists.

Catalog: The (forgotten) gift that keeps on giving

A quick terminology primer, for those who aren’t familiar: in music-industry speak, “catalog” refers to any songs older than 18 months, a somewhat arbitrary threshold that harks back to the peak of the CD format in the 1990s. The opposite — i.e. repertoire younger than 18 months — is referred to as “frontline.” Within the catalog category, there’s also “on-roster” (songs older than 18 months by artists who are still under an active contract with a label, e.g. Ed Sheeran’s ÷ on Atlantic Records) and “off-roster” (songs older than 18 months by legacy artists who are not actively creating music for a label, e.g. In Time: The Best of R.E.M. 1988-2003 on Craft Recordings).

(*I put an asterisk in the title of this newsletter because 18 months is not actually that "old," but nonetheless tends to be categorized in the same bucket as music that has been around for 50+ years.)

Everyday industry discourse tends to center on A&R and artist development as the primary drivers of growth and innovation in recorded music. In pursuit of the next big hit, major record labels are notorious for throwing millions of dollars at a wide range of acts to see what sticks; recent stats from the IFPI reveal that these labels collectively spend an average of $11 million on signing two new artists every single day.

Yet for all this investment, major labels actually make the majority of their annual revenue from catalog, not from frontline. In fact, the proportion of revenue going to catalog is increasing in the streaming age, not declining — nearing 70% of total annual consumption, according to Nielsen Music. Independent labels are no exception to this trend: for instance, Josh Berman, SVP, Digital Marketing at Concord Music, tells me that eight to nine out of the label’s top 10 most-streamed tracks on Spotify every week are consistently taken up by catalog.

Given this dominance, labels have been maintaining their own in-house catalog marketing and sales arms for decades, including Universal Music Enterprises, Sony Music’s Legacy Recordings, Warner Music’s Rhino Entertainment and Concord Music’s Craft Recordings. Most of these designated labels are focused on off-roster catalog; on-roster catalog management tends to be bundled into already-existing sales departments — although they are starting to get similar standalone treatment, such as Atlantic Records’ recently-established catalog marketing division (led by Tom Mullen, who also oversees Atlantic’s podcasting initiative).

In contrast, for streaming services like Spotify that emphasize frontline music discovery in their user experience, investing in any kind of catalog marketing has historically been an afterthought. “I remember [Spotify] wouldn’t return calls when I was at Sony Legacy,” Mullen, who worked at Sony from 2012 to 2017, tells me. “They wanted to break records, and talking about Bob Dylan wasn’t breaking a record.”

The title “Head of Catalog” still remains foreign to most music streaming platforms. Apple has had a much longer commitment to catalog marketing than its competitors, perhaps due to the parent company’s extensive history with iTunes (Steve Wengert leads Apple Music’s current catalog division). But both Spotify and YouTube waited until late 2018 to hire their first-ever Heads of Catalog (Johan Lagerlöf and Darin Soler, respectively); Pandora just hired its first Catalog and Creator Services lead in March 2019 (Jamie Freedman); Amazon Music has catalog-related titles for data management and engineering, but not for marketing.

Pitching catalog to streaming services remains a challenge, especially for artists who don’t have direct relationships with editorial teams. For instance, Spotify’s official playlist-pitching tool — last revamped in July 2018 — allows artists to submit only unreleased music, at least one week in advance; even songs that are just a day old are disqualified from consideration. Even those fortunate enough to have direct relationships with services still resort to pitching catalog tracks in a rather manual manner, i.e. over emails and meetings rather than with a self-serve software tool.

Of course, catalog campaigns are often more extensive, and do require more advance notice and deeper collaboration with labels such that relying only on software would fall short. Software would be helpful for pitching to Spotify’s #ThrowbackThursday and decades-themed playlists (e.g. “All Out 70s”), for instance, but not for putting together a more formal spread like the “Motown Experiences” series of playlists, which takes up the entire top of Spotify’s Soul section (pictured below).

What’s particularly difficult to fit into streaming services’ current layouts and marketing incentives are the cases in which stereotypically catalog-oriented artists decide to release new albums. For instance, artists like Joan Baez, Elvis Costello, Rob Thomas, Steve Perry and Santana have all released new albums within the past 15 months, but receive the majority of their streams from songs dating back decades. This presents a pitching conundrum for labels and streaming platforms alike: “Santana is not going to be programmed next to Tame Impala or J Balvin,” says Berman. “It might fit on a classic-rock playlist, but it’s not really a ‘classic rock’ album. So where does he go? That’s a hard question to answer, and real estate is extremely limited.

Why are streaming services finally taking catalog more seriously?

Despite these challenges, there are three major trends in the music industry that I think will drive more interest from tech companies in catalog:

1. Musical biopics and documentaries

The ongoing surge in music-related biopics and documentaries is one of the most powerful forces for warming up streaming services to catalog marketing today.

For instance, in the six months following the debut of Bohemian Rhapsody, on-demand streams for Queen’s music across platforms increased more than 3x compared to the prior six months, from 588 million to 1.9 billion streams, according to Billboard. Ahead of the feature film’s release, Apple Music participated in this growth narrative by releasing its own exclusive, 16-minute mini-doc “Rami Malek: Becoming Freddie,” which documents the star’s character transformation into Freddie Mercury. Spotify is starting to compete on a similar level; the service maintains a “TV & Movies” section on its platform that includes podcasts and playlists for several biopics and on-demand TV shows, including Elton John-inspired Rocketman.

“I guarantee if you talk to anyone who touched those projects at streaming services, they’ll hang their hat on that and say, ‘look, we can absolutely work on catalog,’” says Berman. “But for every one of those, there are 500 other incredible projects that don’t have a big, mainstream movie to back them up. It’s more of a struggle to figure out how the ‘middle-class’ version of that would work.”

2. On-platform tools for establishing artists’ identities

There are a handful of tools and editorial channels that Spotify and Apple Music give artists to highlight and customize their individual presence on their platforms. One of the most powerful such channels on Spotify is the “This Is” franchise, which showcases individual artists’ top discographies in a personal, branded playlist (examples pictured below).

Spotify itself also actively promotes these playlists on social media. When Taylor Swift first announced the release date for her upcoming album Lover, Spotify’s Twitter account posted a URL to “This Is Taylor Swift,” rather than to specific singles or albums previously released on the platform. Two days earlier, the Spotify Twitter account posted a similar call-to-action for Anderson .Paak, instructing followers to stream the “This Is Anderson .Paak” playlist to revisit the artist’s albums Ventura and Oxnard.

The analogous playlist collection on Apple Music is called “Essentials” (e.g. “Sam Smith Essentials,” “TWICE Essentials”) — and both “Essentials” and “This Is” have become promotional priorities for some labels for driving both catalog and frontline engagement.

“When marketing a new release or moment, we ask artists to post their ‘This Is’ and ‘Essentials’ playlists on socials and run ad campaigns around those playlists, as we see anywhere from 10% to upwards of 30% lift on catalog streams because the user is likely to listen to more than just the new song from within these lists,” says Berman, who adds that Concord regularly makes parallel, discography-focused playlists themselves on YouTube as well. “It’s probably one of the best marketing tools a catalog artist or label has on platform.”

Spotify’s chief economist Will Page offered a similar perspective at the Music Biz conference in May 2019, during which he shared that U.K.-based neo-soul artist Tom Misch had a 5x catalog uplift in the aftermath of the release of his new album Geography, in part because fans were using the release as an opportunity to explore the artist’s previous work.

Mullen says that other on-platform tools like Artist’s Pick on Spotify are also effective in driving catalog streams — particularly around album release anniversaries and other moments in an artist’s life, even if the anniversary in question is just one year old. “We don't need to wait 10 or 15 years before we start talking about an album anniversary or cultural moment again; what about the one-year and two-year increments?” asks Mullen. “With algorithms potentially making you relevant at any time, there’s no going away anymore.”

There are some disparities in how effective these features are for bigger versus smaller artists. For instance, not every artist is granted a “This Is” playlist, and the qualification criteria are somewhat arbitrary and opaque; what’s more, in order to customize one’s “This Is” playlist to highlight catalog in a specific order, one needs a direct relationship with Spotify’s editorial team, which the company cannot sustain for every single artist on its platform. All of Apple Music’s “Essentials” playlists are also curated manually, and handed out sparingly — suggesting that there’s a long way to go across the board in terms of using software to scale catalog marketing in a meaningful way, for artists of any career stage.

3. Smart speakers and voice activation

According to NPR and Edison Research’s latest Smart Audio Report, 77% of smart-speaker owners request their devices to play music in a given week. Moreover, families and older generations are surprisingly fast adopters of the technology, with 30- to 50-years old owning more connected devices than those in their 20s.

In this vein, Amazon Music might be the perfect example of a conflicted company who is pouring much of its marketing budget into advertisements, billboards and activations for frontline, more ‘hip’ artists, yet whose consumer base largely prefers and/or consumes catalog. “You see Amazon Music trying to close the gap on hip-hop because they’re behind with respect to the other services [e.g. see their new weekly hip-hop playlist “Rap Rotation”], but their user demographics are totally different from Apple and Spotify,” says Berman. “They don’t necessarily want to lean into that difference, because they’re looking for younger and newer users.”

Two example voice prompts that Berman brings up as triggering significant spikes for Concord and Craft’s catalog — “play Halloween music,” and “play smooth jazz” — speak volumes to how exactly smart speakers drive catalog consumption, particularly with respect to organic utterances around cultural moments and in-the-moment background music, respectively. In fact, as I wrote in my newsletter back in September 2018, many emerging music-tech trends — including podcasting, sampling and fitness apps, in addition to smart speakers — have catalog revival baked into their DNA, and companies like Amazon would do well to invest more in catalog marketing for the sake of driving more paid subscriptions while retaining their unique user base.

Metadata is to catalog what impressions are to frontline

One thought experiment I’ve been toying around with for a while is what would happen if labels treated catalog marketing campaigns like frontline ones, in terms of tactics for driving consumption — particularly taking into account the power of short-form, user-generated content.

As Lawrence Neil wrote in 2017, “the upended [music] industry doesn’t really control distribution or output in the way it once did, so any song can be accessed — and, therefore, become popular at any time with a couple of clicks.” Deep cuts from the likes of Kanye West and AWOLNATION have been known to chart several years after their initial release thanks to DIY cat videos and dance compilations. I personally think some of the best catalog marketing happens in a decentralized manner on Twitter, through tongue-in-cheek accounts like Keanu Reeves Walking to Music.

Some catalog marketing campaigns are applying this decentralized paradigm to live rather than recording data (i.e. historical showtimes, locations and setlists, rather than single or album release dates), building tools that allow fans to express themselves through reflecting on their favorite shows. “Live music data is way more interesting than streaming will ever be, because live data has a real, emotional connection with fans,” Lee Martin, an independent developer who has collaborated with major labels on live concert playlist campaigns for the likes of Chris Cornell and the Foo Fighters, tells me. “If you really did go to that show, it feels way more important than a quantity of streams connected to a new hit single.”

I think the largest fundamental difference between frontline and catalog campaigns, however, is that the former prioritizes short-term impressions, while the latter prizes long-term storytelling and more meticulous brand management that might not necessarily align with notions of immediate “virality.”

“We could hypothetically take a frontline approach [to catalog] and go big with short-form pieces of content that generate spikes, but then they would be down by the next day,” says Tom Herbert, Director, Digital Marketing and Strategy, Global Catalogue at Warner Music Group. Herbert claims that “the audience doesn’t match up” for Warner’s off-roster catalog on Snapchat and TikTok, and that digital-marketing funds for such catalog are still focused on more traditional platforms like Facebook and Instagram.

Telling these longer-term stories also requires a lot of backend groundwork from labels, though not necessarily from streaming services. For a recent campaign with Motown Records, Martin had to convert a Word document containing 60 years worth of daily historical anecdotes from the label into a scalable metadata framework built on the CMS tool Contentful, which has helped power multiple digital applications including This Day in Motown and an upcoming Alexa skill with daily Motown updates. “Putting in the effort to create our own metadata framework to be able to build the campaign we wanted across multiple services — that’s definitely a catalog approach, not a frontline one,” says Martin.

The unfolding fiasco about the 100,000+ master recordings lost in the Universal Music Group fire of 2008 has woken up many labels to the perils of neglecting proper metadata management and archival work in favor of pursuing short-term, quarterly gains, at the expense even of artists who just got signed to their rosters. In fact, improving the state of catalog marketing isn't just about paying more proper homage to legacy records, nor is it about taking funds away from frontline and emerging acts (which, given the current state of our industry, no one would want). It's also about labels and management companies embracing long-term thinking, and building robust brand and metadata foundations for artists from the very beginning that they can then pull from for future catalog campaigns. In addition, streaming services could build better software tools that any artist can use to drive engagement to their previous discography, regardless of whether they are signed to a legacy label, attract 1M+ monthly listeners or have a direct relationship with editorial staff.

Revisiting my argument that smart speakers and other emerging music-tech trends have catalog revival baked into their DNA, there is also the inevitable question of whose catalog exactly will be revived, and why. Working under paradigms that reward discovery, streaming services could potentially become powerful partners to a wide range of artists in this catalog-first world — or, if things stay the way they are, remain ambivalent barriers.  🌊

✨ If you’d like to support even more thoughts and conversations on music and tech, I encourage you to become a paying member of the Water & Music ecosystem on Patreon. For as little as $1/month or as much as $40+/month, you can access a wide range of perks including:
  • A closed, members-only Discord server, consisting of updates and analysis on the most important music and tech news every weekday
  • Exclusive longform essays
  • Previews and bonus material for my freelance articles
  • Monthly video hangouts with me
  • Live meetups and events (forthcoming!!)
…and much more! Thanks so much for reading! ✨
Follow-up from Issue #54: More notes on music distribution's "middle tier"

Apologies for the messy mobile web layout of the previous issue; I think it had to do with the way I embedded some of the screenshots in the main essay, which were not resized properly. I decided to repost the issue on Medium for easier reading and social sharing — and even got a shoutout from The Long Tail author Chris Anderson himself for including his OG graphic at the top. :)

The y axis on my map of the wider distribution landscape refers to average annual revenue per artist, not to total annual revenue for the company — should have made that clearer. If it were measuring the latter, there would be a lot of reordering (e.g. I'm pretty sure CD Baby is making more revenue from distribution right now than SoundCloud is).

I also want to reiterate that the map I included of the wider distribution landscape is not meant to be exhaustive — but readers brought up several of the same companies that they thought I should have included in my analysis, which I'll address here:

  • Ingrooves would probably fit in the second column because it's owned by a major label (Universal Music) and is selective in its client base.
  • Redeye Distribution and Secretly Distribution should both fit in a new column in between the second and third ones, in that they are not major label-owned but primarily service indie labels and organizations, rather than individual artists, and are selective in choosing their clients. (In contrast, the likes of AWAL and Stem serve a higher proportion of individual artists.)
  • FUGA should also fit somewhere in between the second and third columns, except its focus is much more on technology and software than on marketing — although I hear this is changing.

ONErpm reached out to me to suggest that they should actually fall much further to the left of the spectrum, rather than in the long tail on the right, according to their current business model. While their DIY distribution service (15/85 percentage revenue split in favor of the artist) is open to everyone, that accounts for less than 10% of ONErpm's global revenues — with the other 90+% coming from their hands-on, premium services tier, which operates on up to a 30/70 rev share in favor of the artist. Sources tell me that without charging any upfront fees, distribution services would need to claim at least at 15% share, and ideally at least 20%, to stay afloat; ONErpm's blended margins globally are currently at around 25%.

My writing elsewhere

I have two new Patreon articles out since the last newsletter:

  • A reflection on my Midem interview with Troy Carter, and what it means for the future of music distribution (available to $15/month patrons)
  • A deep-dive into Spotify's "This Is" playlist ecosystem and its various benefits, and inconsistencies (available to $7/month patrons; also a good complement to this week's newsletter issue given the playlists' influence on catalog)

I also wrote a case study on Secretly Distribution for Merlin's annual report, which is a must-read for anyone interested in the latest global trends impacting independent artists, labels and distributors. Thanks so much to Darius Van Arman (Secretly's CEO), Chris Welz (Managing Director), Jacqui Resur (Head of Label Relations), Shelly Westerhausen (Head of North American Physical Sales & Marketing) and Charley Kiefer (Head of Digital Sales & Marketing) for taking the time to share your wisdom! You can download a full PDF copy of the report by clicking here.

What I’m listening to

I had the honor of seeing Little Simz and April + VISTA perform on the same billing in Williamsburg last weekend — what a dream. <3 They also both recorded amazing performances with the COLORS series, which you can watch here and here.

Jordan Rakei's new album Origin (June 14) is one of his clearest, most colorful, most extroverted projects to date — and I was even more thrilled to learn that the neo-soul artist intended to use this colorful production to address more dystopian themes around technology, à la Black Mirror. It does cut straight to the point with the opening track: "It's a mad, mad world / Why can't we keep on living here?"

Eugene Kan and Charis Poon, the latest guests on my podcast, also host their own podcast Making It Up, which offers extremely chill yet super intelligent and thought-provoking discussions on the latest trends in culture. I particularly enjoyed their recent episodes on the experience economy and mega-influencers.


Dang it, even my email hosting platform now has a new music podcast.

Mailchimp launched a new interview podcast The Jump this week, hosted by Garbage frontwoman Shirley Manson. The inaugural episode features Perfume Genius, with upcoming guests include Big Boi and Courtney Love. I would not have found about this podcast at all had Mailchimp not promoted it prominently at the top of its homepage.

Quick takeaways:
👉 In a podcast-first world, brands will undoubtedly be fighting to borrow cultural influence from A-list artists for their original programming.
👉 Podcasts will further normalize brand partnerships in the music industry, because the former still relies on sponsorship as its dominant business model. An artist might not want to be "sponsored by Mailchimp" for their upcoming record; if it's a podcast, that's a totally different story.

Copyright © 2019 Water and Music, All rights reserved.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.