If Only People Cared About ‘Comprehensive Catalogs’ and Millions of Songs…

This industry has been obsessed with signing every catalog, winning every artist, and amassing millions of songs for the perfectly complete service.  And part of the reason is that the people building these services are die-hard music fans themselves.  And therefore, completely different from the mass consumer being chased.

Which raises a critical question:

Do consumers really care about having every last song at their fingertips, millions of songs deep?

The answer is that some do, others think they do, but most don’t. And overall, this matters far less than the industry thinks.

  • Save

Reality tells us this; just look at the biggest subscription services on the planet.  The largest in music is Sirius XM Radio, which boasts 24.4 million subscribers.  That is, four times the subscriber base of Spotify, 8 times the size of Deezer, and 24 times that of Rhapsody.

Sirius has selection, and even Pandora-like stations.  But you’re not picking the songs, playlisting, or otherwise DJing with millions of deep tracks. You’re driving, working, reading, sleeping, or doing something else, while someone else is curating an ultimately limited selection.

Similar patterns are emerging in TV.  Netflix, for example, now boasts 29.2 million subscribers, yet they don’t even have the last season of your favorite show.  Which means people are comfortable navigating huge potholes in content.  And if you’re binging on a series, these are craters in the earth.

Indeed, Netflix is now bigger than HBO, the wise teacher in this game.  Because HBO has acheived its near-29 million on the strength of unmatched, original content, with far less emphasis on ‘licensing everything’.  Yet once upon a time, decades ago, HBO was trying to license every movie that mattered, and paying handsomely to do it.  Just like Netflix.

It was a losing game.  And if you’ve checked out House of Cards, you can see that Netflix is pulling a page out of HBO’s well-strategized playbook.

So, Spotify should be more like Netflix, Sirius, and HBO?  No, it shoudn’t, at all, because every fan is different and music is complicated.  Some, especially early adopters, want deep tracks and obscure remixes, and those people are important.  But most aren’t going down that rabbit hole: they rarely chase obscure artists or react to genome-spliced suggestions.  Even Pandora, which gives lip service to genoming songs, has a very limited catalog and lots of repetition.

Yet Pandora has more than 200 million registered users.

On top of all that, users don’t even know what they want.  Perhaps a reader (@hippydog) said it best in a recent article, quoting ideas from Freakanomics.  Many consumers will say they want everything, but actually don’t.  And all you have to do is look at virtually any chart from any ‘comprehensive’ streaming service.  Because even with the widest selection imaginable, the world’s chosen playlist is amazingly thin.

  • Save

50 Responses

  1. Visitor

    I don’t know many who subscribe to SiriusXM that actually like the limited music provided by SiriusXM. They subscribe because it is better than terrestrial radio or offers some content beyond music they like, such as Howard Stern. Netflix on the other hand doesn’t have everything. But they have more than anyone else. And for that reason, people subscribe.
    When it comes to music subscriptions, I don’t think people really care about which service has the most music. I think they care about which service has the most music they want to listen to.

  2. jw

    The chart argument is oft made but doesn’t hold much weight, imo. What you’re seeing at the top is the middle of a venn diagram. Sure, tons of people are listening to Macklemore. But some people are listening to Macklemore & Britney Spears. Some are listening to Macklemore & Wu Tang Clan. Some are listening to Macklemore & Aesop Rock. Some are listening to Macklemore & the latest indie act. Some are listening to Macklemore & Weird Al. Some are listening to Macklemore &… I dunno, Metallica. It wouldn’t make any sense to anyone if some obscure band ended up in the top 10. But that doesn’t mean that the listening isn’t spread out over a lot of different types of music. This is not the data set that addresses that, & I think you would be really surprised by what gets listened to on Spotify if that data ever emerged.
    What these services have that Spotify doesn’t is a manageable discovery mechanism. HBO’s mechanism is quality control, Netflix’s mechanism is on target personal recommendations, Sirius’ mechanism is genre-specific channels. The Spotify catalog is overwhelming to the average user, they need a discovery mechanism.

  3. Me

    Notice how most of the songs on the list are the same songs/artists that terrestrial radio has been wearing their needles down on. Apparently people hear it on the radio, then go to Spotify to hear it again.

    • jw

      Or they hear it on the radio & share it on facebook.
      Then again, it works both ways… terrestrial radio didn’t break Psy. You’re giving them way too much credit.

      • Me

        Psy isn’t on the list above. All those artists are staples of terrestrial radio.

  4. Yves Villeneuve

    The chart merely shows that discovery is made primarily via radio, not Spotify. I’m sure these songs get leagally downloaded as well.
    Anyone claiming unlimited access is geared toward the average consumer is beating a dead corpse.

    • jw

      Yeah because that’s where everyone discovered Macklemore & Ryan Lewis… the radio. Have you been living under a rock?

      • Me

        Have you been living under a rock? Yeah, people knew about Macklemore & R.L. before, but it wasn’t until Thrift Shop became a radio hit that they jumped to the top of the charts.

        • jw

          Here’s a quote from a Billboard article about thrift shop.
          For some stations, adding the song to rotation went from taking a chance to meeting demand. Rhythmic KEZE Spokane, Wash., PD Zachary “Mayhem” Wellsandt played “Thrift Shop” on Oct. 16 after noticing Macklemore’s online presence and his sold-out show at the local Knitting Factory. Now, the station leads in spins with 638 plays through Dec. 27-a reactionary response to listeners dialing in.
          It’s so easy to say, “Sure, it was buzzing, but we blew it up,” & take credit for the trajectory that the artist themselves created, but if you look at the numbers, & the way Macklemore & Ryan Lewis’ company is set up, that’s just not the case. Their buzz was not slowing down, they weren’t in need of radio. They had 4m youtube plays & a #2 billboard 200 debut back in October before radio even touched the track. (At least that’s my understanding.)
          I’m not saying radio had no effect, obviously it had an effect. But social media unquestionably drove that song. My point is, if you take any one of those songs in the Top 20 & you take social media out of the equation, it’s no longer going to be in the list. I don’t care if it’s Justin Timberlake or Pink or whoever. Radio alone does not cut the mustard anymore.

        • hippydog

          The Thrift Shop video was uploaded to youtube the day after the album was released.. and gained popularity exponentially after that point..
          Youtube (and reddit) ‘broke out’ Macklemore, NOT the radio..

          • hippydog

            JW must have been posting the exact same time as me..
            what he said 😉

          • Me

            Yet, it took 16 weeks for the song to top the billboard hot 100 after it’s release as a single. In fact, it dropped out of the chart the week after its release before re-entering again. Social media got the song attention of major radio, which got the song attention from Middle America.

          • jw

            No shit, Shirlock. The Hot 100 is calculated based on airplay.

          • Me

            OK, Thrift Shop also did not top the charts on iTunes for months after it’s release, either. Not until AFTER it was picked up by radio. So yeah, it was buzzing before radio, but radio took it to the next level.

          • jw

            What I meant to say was, “Sherlock.” I got ahead of myself there.
            Look… who are we talking about here? Are we still talking about Spotify charts? Or are you just taking random stabs at defending radio? Because the iTunes download chart says absolutely nothing about what is driving Macklamore’s Spotify plays, which is the chart in question. “Middle America” isn’t on Spotify. You’re backpedalling into irrelevant territory, & I’m not even sure that your facts are straight.
            I mean, let’s get real here… the Macklemore video had 4m views in the first 4 weeks. And it debuted at #2 on the billboard Top 200 chart with no help from radio. Now it has 300m views. Do you not see the trajectory there? Do you think more people are going to YouTube from radio, or that more people are seeing the video posted on Facebook, Twitter, Tumblr, blogs, YouTube, etc?
            Correlation does not imply causation, as they say. Just because radio picked it up doesn’t mean people stopped finding out about Macklemore on the internet in explosive numbers, subsequently listening to his record on Spotify. In fact, reason would suggest the online trend would continue. Like I said, it’s easy for radio to hop on the bandwagon & claim credit, but the horse was long out of the starting gate, & halfway down the track.
            This happened with Psy, & it happened with Macklemore, & it’s going to happen again as the internet creates a music video renaissance & radio continues to become less & less relevant. It’s just that simple.
            Until you can present some numbers that clearly demonstrate a strong tie between Macklemore’s Spotify plays & radio, I’m just going to assume that you’re wildly grabbing at branches on the way down.

    • GGG

      You really think the avg person under 30 spends more time hearing the radio than browsing the internet? You always make this statement that radio is by far the biggest place to discover music, and while I agree it’s still bigger than one might think, let’s be real here. Justin Bieber broke from the internet. Psy broke from the internet. Macklemore broke from the internet. Hell, you can probably trace more initial interest in Lady GaGa to the internet and certainly her online interaction with fans. Whether it’s social media sharing, Perez Hilton or YouTube, I guarantee more people, at least younger demos, hear the latest song/artist online than on the radio.

      • GGG

        PS-I’ll save you the response time.
        “Disagree” – Yves Villeneuve

      • Casey

        Unfortunately you are right on that. Radio has lost appeal to the younger audience. FM has lost its appeal to younger generations and AM has no appeal. HD Radio was supposed to fix that. It didn’t.

      • Me

        I don’t think the average person scours music blogs looking for the next big thing.

        • jw


          Hey bro, let me clue you in on something. There’s these things called Facebook & Twitter. They’re… pretty much the biggest websites on the internet. lol. People post stuff on them that they like. Like… that Macklemore video. You can watch that shit on your cellular telephone. They call that “going viral.” Because things spread. Like a virus. But in a good way. It’s on your computer, but it’s not a computer virus.

          Kids these days, am I right?

        • GGG

          Hearing something first online doesn’t mean you had to scour for anything. Unless some station or show like Ryan Seacrest gets an exclusive, Justin Bieber’s singles are probably heard more for the first time online than on the radio. Even if there is an exclusive, that shit will be posted online immediately, too.

  5. Visitor

    We intterupt your regularly scheduled news blog to bring you — Digital Television News Redux, two point oh.

  6. Spoken X Digital Media Group

    I think there is a certain market of internet and mobile users that would love to have millions of complete–song listening opportunities as their solution. Albums, songs , video , movies: : Digital media entertainment software is the power of the universe. Actually digital music is a huge bubble on Wall Street they keep in the basement like ‘ Hell Boy ‘ !

  7. Jonathan Bailey

    People don’t care if a music streaming service has all of the music in the known world, but they do care very much if it has all of the music they want to listen to.
    Suppose, for a second, a person only wanted to listen to 100 songs ever. If service A has a catalog of 10 million songs but only has 75 of the ones they want, its useless. If service B has all 100 but only has 1 million tracks, it’s still the better service.
    I see this all the time with Netflix. I never have trouble finding stuff I love in Netflix. It’s library is perfect for me and my taste but a friend of mine swears up and down it’s worthless and is garbage because it doesn’t have whatever latest blockbuster she’s craving that day.
    People only care about library size in as much as it gives them the best chance the service has all of the music they want to listen to. I’ll never listen to 99.xx% of the music in Spotify and that’s fine, it has, with only a few minor exception, everything I do want to hear…

    • jw

      Sure, but the larger the catalog, the more of a chance that a service is going to have more people’s 100 favorites. What if Netflix had all of the movies that you want to watch, PLUS all of the blockbusters your friend wants to watch?
      The answer isn’t to be a niche service, but to service all niches. It’s all about making discovery easy, surfacing the 100 songs that a user wants to listen to with minimal effort, & perhaps surfacing another 100 songs that they didn’t even know they wanted.

      • TheTruth

        This guy gets it. Anyone who says that streaming services need to scale-back to attract and retain more users should just give up.

  8. GGG

    I think people also need to realize there is always going to be a selection of music that is vastly more popular than most other music. Using the fact that Top 40 hits are the top hits on any service or platform isn’t really weird at all. Unless you have a niche service, enough “mainstream” fans will be part of it. And not only that, non-mainstream fans still listen to mainstream music because it just seeps into the culture so much. And the mainstream songs people are going to want to listen to are the most popular ones, so it’s just a cycle for a while.
    I do agree though that your avg listener probably doesn’t give a shit about back catalogue, though. Nobody will care about Katy Perry album tracks in 5 years, they’ll just want to hear the hits. Hip-hop and EDM do still have cultures of mixtapes, remixes, random single releases, etc so those may drive some interest. So I think the overreaching idea of this article is dead on. Nobody can predict anything because there are way too many ways people interact with music now. Trying to get a huge percentage of people to fit some mold is just silly.

    • Mike C.

      True, it’s not the least bit strange to see a Top20 list of artists as the most popular streamed music on Spotify or any other music service with a million-song plus catalog. But it’s faulty reasoning to assume from this graph that most consumers “don’t care” about deep catalog.
      The information needed in the graph is how many streams are Top20, and what percentage of these streams make up Total streams on the service?
      If the Top20 artists make up a large % of all streaming on Spotify, then I would agree with what the article posits, that your average music consumer doesn’t care about catalog. But I would suspect that the percentage is less than 50%, meaning the long tail of music is the real appeal for services like Spotify.
      Btw…the idea Spotify is (or isn’t) a discovery tool, or the # of “active” vs. “passive” subscribers might play into the original question, (do music fans really care about deep catalog) is completely irrelevant.

      • GGG

        I wasn’t assuming all that from the graph, I was using my own anecdotal evidence and speculation based on people I interact with on a regular basis, too.
        I’m not saying I disagree with what you’re saying, I just look at it like this:
        One of the reasons I think top 40 songs/artists still stay so big while being fairly homogenized is because the people that truly like top 40 music allow themselves to be told what’s good. Don’t mean that in the sense of they don’t have personal tastes, I just mean as long as the next batch of pop songs are good to them, they’re happy. So I’m looking at that as “most” people, which might be underestimating the amount of people that do care about deep cuts, so you may be right.
        It’d be an interesting study. You could do some basic research by looking at the play bars of pop records, compare singles to non singles but who knows.

        • jw

          Spotify has play #s for top songs in their latest release. It actually really fucks up the layout, but it’s an interesting feature.
          Carly Rae Jepson (who relied entirely on radio, & didn’t even have an album out for the life of her single) has 114m plays of Call Me Maybe (1.7m of the remix), & 40m of Good Time. Then 3 songs between 2m & 6m, & the rest <1m.
          Conversely, Macklemore & Ryan Lewis (who didn't rely on radio) have a top 10 is all 4.8m+, with Thrift Shop having 102m plays & Can't Hold Us with 53m.
          Not the best examples, or the most exhaustive sample of data, but still insightful.

          • GGG

            Ah, coo, didn’t know that, thanks.
            Yea, different artists will be differnt obviously, and I guess it also depends how you quantify things. Hits will always have more plays than deep cuts so if the hit is 10M and some deep cut is 3M, maybe that’s a very positive thing.

          • jw

            Looks like in between the time I wrote that & now they’ve rolled back the Artist pages in the Spotify app, so no more play figures. But earlier today I calculated some averages… while Carly Rae & Macklemore’s two hit singles were in the same ballpark as one another (114m vs 102m, & 40m vs 53m, respectively), the rest of the Top 10 averaged 2,131,960 for Carly Rae & 7,821,478 for Macklemore. (I guess you can take my word for it.)
            I think that Carly Rae’s dropoff is what one would expect from the “radio effect,” or at least from the top-down push, though the numbers might have played out differently had she had an album available when Justin Beiber first tweeted about the video. Either way, Good Time (2nd single) got 1st rate support during the album cycle & there’s still a massive dropoff.
            On the other hand Macklemore’s album cuts seem to have 4x the accumulation over a much shorter time period, & with plenty of life left in the album… I think this defies the “radio effect,” & is a prime example of what grassroots growth looks like at scale. I would go as far as to say that it’s what success SHOULD look like (without all the short cuts).
            If the artist is properly developed, & the songs are good, & the music is promoted in a “sharable” way that’s true to the artist… maybe I’m stretching it here & maybe a lot of this is actually speculation based on very little data, but I think that fans demonstratably care more about the deep cuts. Maybe 4x more, maybe not. But I see music going in this direction in the long run, more artists owning this own recordings/publishing/merch & managing their own social media, & using record labels for distribution/radio, & breaking on the internet, with youtube+facebook/twitter/tumblr/pinterest/whatever acting the way mtv did in the 80’s/90’s.

  9. steve warren

    so, it is discovered (again) that people actually prefer a limited catalogue of songs. we music radio/broadcast veterans all learned that years, decades ago. shorter playlists produce bigger ratings.
    ask them about it and they’ll always SAY they would prefer a station that plays “lots of different songs, a very wide variety”. but in reality, they wouldn’t listen much to a station like that because “oh, that station plays a lot of crap songs.”
    i (and many others) have been involved in format competitions where our competitor station had a total playlist of maybe 800-1000 songs, we had a playlist of 650 and consumer/perceptual research would show the city’s listeners prefered our station because it “played the widest variety of music”.
    now how/why is that?? because playing the BEST songs over and over is what they truly want to hear. the rub is: how to discover just exactly what the “best” songs are? the ability to do this is what separates the amatuers from the pros.

    • jw

      Oh please.
      When you say “best” you mean “least offensive.” You all but outright said it.
      Sure, people don’t want to hear songs they don’t like. But that’s precisely why radio is in so much trouble. Because there are songs that one person might not care for that someone else is going to love. And Spotify, or even Pandora, allow for this. Thanks to the internet, users can listen to the “limited catalogue of songs” that they themselves enjoy. It’s just about surfacing that catalogue for each user.
      Get over yourself, asshole.

      • Visitor 3

        Whats with your need to be a complete douchebag?

        • jw

          Because proponants of antiquated, one-to-many systems should be publicly chastized.
          There’s a place for radio, & maybe there always will be, whether it’s over FM/AM or the web, whether it’s live or pre-recorded or crowdsourced, equation-driven, or whatever. But the idea that the “professionals” are separated from the “amateurs” based on how much money they can drum up pandering to the lowest common denominator stands in direct opposition to everything positive that the internet has brought about, & as a podcaster it’s personally offensive. My mission is not to water down my playlists to be “just ok” & nonoffensive to the most people possible, it’s to connect with & delight listeners with specific tastes to the greatest degree possible, & if certain people don’t care to listen to what I play, that doesn’t make me an amateur… there are other curators out there to satisfy those consumers.
          The arrogance expressed around curating a one-size-fits-all catalog of least offensive songs reveals a fundamental lack of understanding about the internet, it’s purpose, & the possibilities it creates. And these types of antiquated ideas are going to stifle innovation going forward. And so I feel the need to be a complete douchebag about it.

          • P T Barnum

            “The arrogance expressed around curating a one-size-fits-all catalog of least offensive songs reveals a fundamental lack of understanding about the internet, it’s purpose, & the possibilities it creates.”
            This is partially true. The original comment by Steve Warren has more to do with human nature than technology ” the internet, it’s purpose, & the possibilities it creates.”
            When it comes to selling you give the people what they want. Steve gave a prime example; when given the choice between less and more options people want less choice.

          • jw

            That’s just patently untrue. It’s not human nature to want less.
            Have you ever introduced someone to something & been like, “I think you’ll love this…” & then they’re like “Yeah! I DO! But I’m just listening to so much other stuff right now. I don’t really have room to like a new song.” That’s not human nature.
            The metrics that Steve Warren is basing his arguments on is # of listeners, not the emotional response of listeners. And there was a time when we couldn’t be concerned with the emotional response of listeners because we were all sharing the FM waves, but that’s no longer the case.
            The fact of the matter is that people want MORE of what they like BEST, & they just don’t want to have to hear stuff they DON’T like. And so, in this day of age, thanks to the internet, there’s no justification for nixing a song that someone might love because someone else might hate it except for to line the pockets of executives at Clear Channel, Cumulus, et al. & to feed the ego of asshole program directors at these behind-the-times radio stations.
            That’s why people don’t love radio. That’s why people have to come up with campaigns called “I <3 Radio"... it's fucking brainwashing. Because no one really even LIKES radio. They're indifferent at best. The least offensive wins. And we don't have to live in that world anymore.
            At least, that's my opinion.

    • steveh

      hey “JW” – why is Steve Warren an asshole?

      • jw

        Because of the smug way he suggest that identifying the lowest common denominator is what “separates the amateurs from the professionals.”

        • hippydog

          Smug or not.. Asshole or not..
          I think his basic assumption is correct..
          After all, we do live in a capitalistic society..
          and thats what works..
          and its not always the “lowest common denominator”.. sometimes the public can surprise you..

        • steveh

          Hey I’m not a defender of commercial radio, but the substance of this thread is that while people say they want massive variety they often tend to crystallise around the big successful tracks. As evidenced by reports of listening habits on streaming services.
          Steve Warren was lucidly pointing out that the same phenomenon is observed on commercial radio stations.
          His comment about pro/amateur was purely in the context of fiecely competitive commercial radio where creating playlists that get the big listener numbers is indeed a skill.
          Outside of that I don’t see that he was making the kind of value judgement that you might find offensive.
          I think it was completely unfair and out of order to call him an asshole.
          You should apologize.

          • jw

            Fair enough. I may have been out of line.
            I apologize, Steve Warren. I may very well be the asshole in this instance (as in many others).
            However, I think that he’s totally off the mark in assuming that radio stations with tighter playlists being more successful is a function of human nature, it is in fact a function of the shortcomings of the delivery mechanism, & “not driving listeners away” in a captive situation (i.e. an automobile) is not by any stretch of the imagination a reasonable approach to disiminating music when technology now allows us to delight listeners on a case-by-case basis.
            Based on any reasonable analysis of human nature (that people are going to want to at least be exposed to as many songs that they personally deem to be “good” as possible, regardless of whether that song might cause someone else to change the station), it’s clear that personalized discovery needs to be fixed on the internet, & that tight, “universal” (aka lowest common denominator, least offensive) playlists ought to become a thing of the past. That doesn’t mean that the lowest common denominator stuff won’t still show up as the Top 20… that’s essentially what Top 20 lists are surfacing (It’s not a qualitative list!!!!!! These are not the best songs!!!), it just means that the rest, the stuff that really excites people, shouldn’t be marginalized, which is what I, perhaps wrongly, felt like Steve was doing.
            But I’ll openly admit that I’m biased… almost nothing I really like ends up on radio.
            Anecdotally, a guy I know showed up at Criminal Records at 4:30am on record store day to get the only copy of a Sigur Ros record (I think they only pressed 1,000). I don’t know if Sigur Ros gets played on the radio, but I’ll assume they don’t. I’d change the station if they came on because I don’t care for that sort of stuff. But I was there for the Big Mama Thornton reissue (at 10:30am), & he would probably change the station if that came on. But we’d probably both sit through the new Daft Punk single, even though neither of us would identify ourselves as Daft Punk fans.
            Luckily, it’s all on Spotify, & we’re not under the thumb of smug program directors who rule the airwaves. (Obviously that’s a gross generalization only meant to be taken half seriously.)

          • P T Barnum

            Now your argument makes more sense.
            Technology has made it so that paying an extra percentage for someone to curate or dictate your selection is ridiculous.
            However, we live if a free market system and if no one can make money offering everything to everyone then something else will be offered.

          • jw

            That’s not totally what I’m saying.
            Essentially, my argument is… maybe I love that new Jon Spencer Blues Explosion record. But maybe that would just be “too much” for someone else, & cause them to change the radio station. Should I not be exposed to Jon Spencer because of what someone else’s taste might be? Is a song that’s more pallatable, potentially less offensive… “better?” Obviously not. It has nothing to do with who gets paid for what, it has to do with the fact that there are songs that certain listeners would love if they were just exposed to them, & it’s the shortcomings inherent to any broadcast system (i.e. sharing the airwaves with other listeners) that keeps playlists tight, NOT human nature.
            Just because people want less of what they DON’T like doesn’t mean that they also want less of what they DO like. There’s no argument to me made for that… it just doesn’t make any sense whatsoever.
            That’s why these “most played” lists aren’t qualitative, & are only marginally useful to anyone who doesn’t have something to gain financially (advertisers, program directors, labels, managers, etc). What’s TRULY valuable to the LISTENER is MEANINGFUL, PERSONALIZED DISCOVERY.

          • P T Barnum

            ” Is a song that’s more pallatable, potentially less offensive… “better?” Obviously not.”
            Two things:
            First, making qualitative judgements about music is risky business and usually does not become part of a music distributors calculations. When you begin assigning qualitative value judgements to the product then basic economics dictates that the distributor should charge more for “hey mom” by Jon Spencer Blues Explosion than “Gangnam Style” by Psy.
            Second, maybe not “better” for the consumer, but it is definitely better for the distributor. If you think of it from the distributor side, offer unlimited universal selction without curation and run the risk of confusing and the potential customer OR curate and organize the most popular product offer that only and run the risk of offending the minority of listeners that want music that may be less palatable

          • jw

            I’m not suggesting that I’m the arbiter of quality. I’m just saying that the qualitative value of a specific song can not be abstracted from quantitative data about a collection of songs, & so words like “best” don’t mean what they appear to mean when guys like Steve Warren use them. “Best” is, after all, a qualitative word in this context, whereas he is only talking about the value of a play in terms of ad dollars. Not best for the listener, but best for those with financial stakes in the game.
            When I talk about the quality of a song, I’m talking about an individual’s emotional response to a particular song. Much has been written about how the internet has fragmented culture, how gatekeepers have dwindling influence, how we’re moving from top-down structures to bottom-up structures, & this all plays in to what I’m saying.
            Yes, songs have different qualitative values to each individuals… some would be willing to pay more for a particular song than others. And while there’s no way to price songs on individual scales via the ownership model (and the tip jar model is not scalable), this is, in effect, what you get with the streaming model. Songs that are streamed more net more money for those with financial stakes in the game, perhaps more than a purchase over time. Therefore, in order to maximize revenue, you want to, somehow, present every user with the most songs that they like the best.
            To this point, the goose ought to be prioritized over the gander going forward… this is clearly best for everyone involved (i.e. maximum emotional response to music for the listener + maximum revenue for everyone else).
            Furthermore, what’s clear is that the industry is still trying to shoehorn an old model into a new world (as evidenced by this article and many of the comments). And that the new model is far from perfected is actually good news for the future, because it means that the potential for more revenue is there. And revenue won’t be maximized until music discovery & music listening is optimized for the individual, based on the fact that restrictions inherent to the broadcast system are no longer in place.
            Again, this doesn’t necessarily mean that the “top 20” is going to look any different, because we’re talking about crossover plays, rather than emotional response by users… no qualitative information is being surfaced.

  10. TheTruth

    This comment thread it outrageous.
    Here’s a hypo: You’re listening to Third Eye Blind on Spotify (having an early 00s moment – just jamming with your bros). Then, Bro2 is like, “hey bro1, why don’t you pull up that song by brobandX because if I hear that song by brobandX I will totally sink this cup” (clearly, we are in the midst of a heated game of beer pong). But bro1 can’t find brobandX on Spotify because Spotify coudn’t secure the licensing. This is bad. Bro1 has a few options: (1) go on iTunes, wait for the song to download, and play it upon download-completion long after Bro2 needed the song to make his shot; (2) go on YouTube and stream some poor-quality version of the tune; (3) PUT ON THE RADIO (uh, no – no one wants to hear that sh*t); (4) anarchy.
    Bro2, bro1, and all the other bros in the room are forever angry at Spotify.
    This is real life.
    This is why you need a big catalog.