The technology had nothing to do with Spotify winning. There were probably hundreds of startups that built cool music tech but they all failed because they couldn't get the licenses. As far as I know the true story about why the RIAA gave Spotify the green light after saying no to so many before them hasn't been told.
> There were probably hundreds of startups that built cool music tech but they all failed because they couldn't get the licenses. As far as I know the true story of why the MPAA gave Spotify the green light after saying no to so many before them.
It's what makes shutting down Grooveshark and going after other music services so ridiculous - it's not a matter of wanting to combat "piracy" as much as it is about wanting to make sure that their favored candidate wins, instead of an independent one.
Again, remember that the RIAA currently has authority not just over artists signed to major labels, but over artists signed to independent labels as well[0], so they really do truly have cartel-like power, and they are clearly not interested in giving that up.
[0] A subsidiary of the RIAA has the sole legal authority to collect royalties on behalf of all artists, and then artists can request remittance of payments from this subsidiary. Perhaps unsurprisingly, independent artists report mixed results with actually getting the RIAA's subsidiary[1] to pay them: https://en.wikipedia.org/wiki/SoundExchange#Authority
[1] Technically SoundExchange is now a 501(c)3 and no longer owned by the RIAA, to make them appear more "independent", but this is a change in name only.
Yeah, this article is written as if the hard part of creating Spotify was figuring out that people wanted cheap streaming music (obvious), and then building it (difficult, but many groups could replicate it). I suspect that it was much harder to obtain the licenses, and that's the story I actually want to hear.
The truth is that Daniel Ek met the major 4/5 labels at the same time and showed them their (at the time pirate) software. They said - yes - if we can own 75-80% of you. The truth is that the major labels own about 80% of Spotify between them - it wasn't the RIAA.
You're right. Having a music player play songs fast is nice but the record label deals in essence gave Spotify monopolistic advantages only other streaming companies could compete with. (Pandora, iHeartRadio, etc)
These same content licensing advantages are given to Netflix and Hulu.
FYI - Hulu is owned by Fox, Disney and NBC.
Spotify is partially owned by Sony, Universal, Warner, and EMI.
You need to use common sense when you enter an industry about your strategy. It can't all revolve around great tech.
Not to mention the fact that besides the basic premise of the product, Spotify manages to be shite at nearly every bit of functionality they offer. Navigation, discovery, organisation, it's all awful. And now that they've dropped the API, there's no third party apps to compensate for it. It has that one killer feature, but sucks at everything else.
I'm sure the back end is technically awesome, but I really hope Spotify gets some competition from a party that actually cares about music and the user experience.
Spotify could be so much, much better. Instead they're just lazily exploiting their licensing advantage and what is effectively a monopoly. (Pandora isn't available where I live, nor in most parts of the world.)
(To be fair, half the mess the UI is is probably caused by pressure from the music industry to peddle their crap regardless of what the user actually wants. At least that's what I hope, otherwise Spotify is even worse than I thought.)
If I created a service that allowed unlimited streaming of the latest movie releases for $10 a month / ad supported , pretty sure it would be pretty popular.
The 200ms latency 'wasn't possible' because of how TCP congestion control works. Spotify used to use 2 channels for downloading - the first chunks would come from their servers, then you would try and download the rest of the song and do read-ahead on the p2p network. Consequently, the TCP connection to their servers would be idle for a bit causing the congestion window size to drop back to 1 segment (1500 bytes) in size. Over higher latency networks (wireless), TCP slow-start could take up to a couple of seconds to get up to speed, as you need a RTT to increment the segment size. Spotify's trick was to build (compile, more likely :)) a version of TCP on their servers that prevented the congestion window from connected clients from dropping back.
There's a difference between generating business value and generating profit. It's common for founders to get liquidity during financing events and this typically involves a sale of equity (i.e. transfer of business value from the founder to another stakeholder).
This is distinct than the company becoming profitable; many companies will increase in value while not being profitable.
The original commenter is making this comment I presume because he/she thinks Spotify is a mature enough business where it should be generating profits instead of operating at a deficit in order to grow and grab market share.
Spotify gets props for being the first player on that market who really managed to grab share, but since Google Play Music exists and allows uploads of one's own music collection as well as dynamic cache-ahead on wifi of playlists on mobile devices², it's had some feature catch-up to play and doesn't seem to even have any willingness to do so.
² this is massively huge for people with bandwidth caps on their mobile contracts
> if the music starts in two hundred milliseconds or less—about half the time it takes, on average, to blink—people don’t seem to perceive a delay
Anyone know how spotify may have achieved this? My instinct is that they pre-download the next song in the queue + heavily cache everything. But in my experience, playing a new song in almost instantaneous too.
200ms seems like a long time to me. From what I know, if your database takes longer than 100ms to return a query, theres something wrong[1] and CDNs have pretty much solved that "I need to get this static file around the world quickly."
With just an song id, with S3/Cloudfront and [db of choice] I don't think it would be hard to deliver a song in less than 200ms (to the continental US atleast).
There is no magic in this. 200ms is an eternity and would actually be considered rather poor performance.
Realistic latency is closer to 100ms for a cache miss (~40ms for your client to reach spotify plus another 40-60ms for their internal fetch).
You have to realize that music files are tiny and consume very little bandwidth by today's standards. All commercial music combined (roughly 30 million songs) fits into half a rack of storage.
I haven't used spotify for ~3 years and last week wanted to listen to some specific music, so i downloaded it. Within 20 minutes, a really really annoying audio ad came on and I went from thinking that this team has built a really great product since i've used it, to having a really negative opinion about them because the ad disrupted my flow as I was in the midst of a great hacking session.
Something that they could do differently that might be interesting, is to special case ad plays for new and returning users and slowly increase the quantity of ads that are played. This way my initial experience is very positive and as I get used to incorporating spotify into my workflow, I get more used to ads, and can decide whether I want to be a paying customer or use the product with ads. I wonder if anyone else has incorporated similar growth hacking strategies.
> I went from thinking that this team has built a really great product since i've used it, to having a really negative opinion about them because the ad disrupted my flow
Just throwing this out there because you mentioned it's been a while since you used Spotify -
Spotify offers a premium service for $10/mo which allows unlimited music without ads playing every few songs [1]. If ads are the deciding factor in whether or not you consider the application "really great" and you decide it's worth it to you, the option is there for ad-free listening.
The main reason I bought Spotify premium (other than access to the full mobile platform) was to ditch the ads. I don't mind ads so much, but Spotify ads are particularly annoying. I seemed to get a lot of ads from fledgling hiphop artists, including extended samples of their songs, which was really disruptive and unpleasant when listening to folk/americana. After one two many focus sessions were ruined by obnoxious ads, I went ahead and bought premium.
Another folk/americana fan here, and I get contemporary pop music ads between my tracks on Spotify, and there's nothing more annoying than autotuned screeching between my Bob Dylan and Gillian Welch.
However instead of buying premium, I just use Spotify for discovery. If I like something I just get the album off iTunes, Amazon or Bandcamp. I find myself spending less per year than if I had subscribed to Spotify, _and_ I get to keep my music.
I remember that this is what blew me away about Spotify ... I'd used streaming services before, rdio, grooveshark etc .. but on my phone the streaming would take a while to load, in bad patches of reception it would cut out etc etc. Spotify was the first mobile streaming experience that made it to where I didn't need my itunes any more, and I never looked back after that.
Its a shame who ever is running the UX for their products is trying really hard to roll back all the awesomeness and good will they've worked for, by making questionable decision after questionable decision, their latest desktop update being the case in point.
- They removed Album view a few years ago now (version 0.8.4) ... I still have that version installed at home because I like/need it so much.
- They removed starring/favoriting of songs, in favor of some other (add song/add album) thing that just doesn't work the same way despite them asking you to try it that way.
- In the latest version of their app they moved things around that had be in locked positions for YEARS ... for no reason.
- Then they removed the ability to search inside of playlists ... again for no reason you can see the fury of the masses in all its beauty here ... https://community.spotify.com/t5/Help-Desktop-Linux-Mac-and/... ... completely ignored by spotify of course.
- Their mobile app is still mostly great but has a couple of no brainer things that they've never got right (playlists sort in reverse order from what you'd expect, now you can't delete a song from a play list without editing the entire play list etc etc etc
Spotify just stopped caring about what the people wanted a long time ago ... they've gotten slow and bloated and I bet they have an army of Product managers poring over all sorts of data and making these weird decisions, instead of engaging their common sense ... either way ... if something comes along that has a bit more soul, and love for its customers ... I'm off in a heart beat, and yes, I've paid the full price for spotify for almost 4-5 years now.
I'm not sure if the different desktop versions have feature parity - from what I've seen of Spotify's development process, each platform is handled by a different team.
Excuse my ignorance, but what is the album view, and how is it different to what's currently in the app (as there seems to be something that I'd consider to be an album view for OS X)?
As for the rest of your points, you are on the money; seems to be a lack of commonsense.
The streaming part of Spotify is what makes the service great. But when it comes to improving their desktop-app they really are doing some weird stuff with their hipster developers. I think they are suffering from something that I'd like to call "the winamp effect".
This may underrate the real difficulty of Spotify: convincing record labels to let them give away music for free on the assumption freemium would have a reasonable conversion rate to paid.
I've never understood the appeal of Spotify, recommendations are shit, the client/ads suck and I'm not whipping out my credit card when I've got a music collection + Google Play + iTunes.
Looks like we are a minority.
Spotify is a lesson about building what people want. Sometimes people just want crap, because free.
Sustainability of the model for the musicians and the shitty payments they get is not even taken into account, obviously.
Eh. Spotify is a great service and that is why I pay for it.
It lets me sample new music instantly and organically without needing to pirate anything, and if I really like something, I can buy it and add it to my Plex too. Yes I know that I can probably find the album on certain torrent sites, download it in less than 1 minute, unpack it and then get the song I wanted to listen up and playing.
But that's a hell of a lot of work, compared to just double-clicking the song and having it playing instantly. And if the song sucks, you dont need to keep seeding albums you actually want to delete just to maintain ratios at private trackers, etc etc.
Using Spotify is just so much less hassle. These days I find I use Plex less and less for music, and Spotify pretty much gets used 99.9% of the time.
So I'm curious: What do you think is crap with Spotify?
The Music experience. It's superficial, and it's imprinting in younger generations that music is just another commodity.
I'm not making a Luddite statement, just saying that listening to music by proxy without the physical or digital experience of owning the music itself is part of a general depreciation of music making.
Edited for duh...