Hacker News new | past | comments | ask | show | jobs | submit login
SoundCloud cuts jobs, closes SF and London offices (techcrunch.com)
462 points by janober on July 6, 2017 | hide | past | favorite | 425 comments

It's been a downhill trend ever since trending likes were replaced by reposts. Trending likes actually showed which tunes my friends found most interesting. Reposts allowed a single user to spam a feed with their personal taste. Then repost bots were allowed to require repost to unlock a free download and now discovery on SoundCloud is almost dead. My friends who are successful producers but refuse to play the bot and pay-for-repost game are completely drowned in the noise.

My bet is that since trending likes are a Graph DB problem, SoundCloud just chose not to solve it once it became too hard on their database. They famously had issues with early Mongo adoption so this would fit right in. Their most recent frontend HTML5 rewrite always winds up my CPU and hasn't changed in a couple years.

SoundCloud to me has always been a great idea with some great design that got ruined by some poor engineering choices from the top. And at the end by a desperate grab for stats and cash. I think producers would have paid more, better subsidizing free plans, if discovery hadn't gone downhill. I ran the numbers once on how CDbaby and Bandcamp became successful (based on stats trickled out over the years) while SoundCloud could never turn a profit. There's still plenty of room for someone to do this right.

Agree on the reposts issue. I used to spend 30 minutes plus a day finding new music, then stopped after that change.

I made this if it helps anyone: https://github.com/kyranb/SoundCloud-Feed-Cleaner

Maybe there's an incompatibility with another extension or they have changed the site, but I can't get the Uploads Only tab to show up.

What browser are you using? Care to take a look in Chrome's console for any errors and submit your findings as an issue?


I investigated your issues page and the previous comment may be related to the following issue.


After loading a fresh copy of the page, the tab appeared.

With that said I opened an additional issue. Chrome console log did not appear to provide any relevant information.


I used to go to SC every day. When reposts started becoming a thing I loved it, since the people I followed would repost stuff I would have never found. But I guess I quit in time, as I never saw bots or what you are describing. That's too bad :(

I only started using sc once reposts were already out and never had an issue with them, but there's this new trend I've started noticing on SC, where artists add their new songs to a playlist and then repost the playlist every day. If a playlist has 20 songs it can be impossible to listen to anything else on your stream because the whole playlist gets added to the queue. I find it extremely selfish since it just makes it that much harder for other artists on the stream to get their fair share of air time.

What caused you to quit SC?

Cultural and product focuses that emphasized engagement quantity rather than engagement quality.

Comments would be littered with things like "hey, great, check out my stuff!"

That type of shallow "like for like" culture is absolutely cancerous for what was previously a very maker-heavy audience.

> Comments would be littered with things like "hey, great, check out my stuff!"

> That type of shallow "like for like" culture is absolutely cancerous for what was previously a very maker-heavy audience.

Man, I've been in there since day one practically and it was always like that, as has been every service of that kind I've ever used (hands up anyone who ever maintained an MP3.com playlist). Music runs on shameless self promotion. Players gonna play. Etc.

Ideally, promoters would always be trusted experts. Question is: How do those experts, without engaging other promoters, discover potential tracks for promotion?

The slush pile or spending an ungodly amount of time in low end venues.

What would be your solution to this?

The amount of music in my feed and the trouble of curating it all. I found new songs I loved every day, but being a programmer, and having to switch to the SC browser window/tab to hit shift-right-arrow to skip a song, or space to pause it, became too troublesome.

I eventually switched to Spotify, since I could have global keyboard shortcuts for the app. If SoundCloud had a desktop app I would use that instead of Spotify in a heartbeat (assuming I could play my own music as well).

http://www.streamkeys.com is a chrome extension that gives global hotkeys for many streaming media sites (including SC).

I wish there was a firefox port.

I use this desktop app, opensource and cross platform, constantly


I was aware of that, but if a company fails to understand it's users' needs, I don't have much faith in them, so I just move on to something else.

Not to mention that the limitation in the API usage made the soundnode app useless, as it hits the API limit in a few hours of use.

They are failing not to provide an alternative and to prevent you from using an alternative.

Is the API dependent on popularity of songs you listen to? Because I find I dont hit limit until a whole day of use (4 to 6 hours) but guessing this could be because I don't have much popular music in my queue. I ask because I notice that now they introduced advertisements but these also only come up on songs with over 60k plays

That’s the reason I’m using Hype Machine to listen to those "trending likes" tracks on SoundCloud. I almost find a nice new song to listen every single day.

Yeah me too, Hype Machine is fantastic for that. Such an undervalued service nowadays. If you haven't subscribed as a supporter I'd recommend it http://hypem.com/supporters

Got my supporter stickers in the mail today, they even printed a special postcard with a bunch of titles I have loved over the years. Love the site.

One out of how many? I'm just curious about the signal to noise ratio. I could use a new discovery tool, with not too much noise.

Between 1/10 and 5/20 or sometimes more from http://hypem.com/popular/noremix

Sounds tryable. Thank you.

I wasn't a fan of the UI refresh for the same reason, so I made my own alternate homepage which sorts the feed by artist and de-emphasizes reposts (http://gurlitz.org/soundcloud-digest/). Unfortunately like you said, content also dried up around the same time.

I haven't been using it long enough to have an opinion, but I did notice that the hosts of podcasts I enjoy would repost when they guested on another podcast, which I otherwise would not have noticed, and did appreciate.

I only use it for podcasts though, never music.

Trending likes solved this, because even an upvote brigade required the user to have multiple friends in the brigade. All I had to do was follow a few key influencers on a topic and the post would show in my feed.

Trending Friend Likes is a Trivial GraphDB problem. We could help them offload that to a separate service if they'd ask.

MATCH (u:User {Id:123})-[:FRIENDS]->(f)-[:LIKED_ON_2017_07_08]->(t:Track) RETURN t, COUNT(f) ORDER BY COUNT(f) DESC LIMIT 25

I signed up years ago, but only became a user in the past year or so. It's never really appealed to me for music discovery but it's been great for checking out new podcasts.

> …it's been great for checking out new podcasts.

For anyone confused by this: SoundCloud can't play podcasts, but a small percentage of podcasters host their show on SoundCloud, and the app can (naturally) play those.

However, SoundCloud has probably seen the peak of its podcast-driven business. Podcasters have been wary of SoundCloud's fortunes for some time, and most have a Plan B in place. Today's news has likely expedited plans to find alternatives among at least some podcasters.

Any reason you don't make use of iTunes or the charts in other podcasting applications?

I use pocket casts on mobile and soundcloud on the desktop. Soundcloud's better for discovery and it lets me listen without going through a subscription workflow.

How successful are CBbaby (nowadays) and Bandcamp? I know people who use Bandcamp but I don't know how much of a bubble I'm in.

Insignificant in the grand scheme of things.

The music industry is a shade of what it once was, and both CDBaby and Bandcamp are tiny slithers off of that.

Bandcamp is fine, but it's not on the level that Soundcloud was for browsing music (or for hosting your own). It's still really limited on the social side - the only way you get "likes" is when people download your stuff, the discovery tools are limited to "what's popular in this tag today" and random curated content. There's no "feed" like Soundcloud had. Also, possibly the most irritating part of the site is that fan profiles and artist profiles are distinct entities. So if you want to buy someone else's music but you're signed in, you have to sign out, get on your other account, purchase, and then log out/in again. Super clunky. Last, there's no way to communicate with other people, other than the email that you send when you buy things. I used to legitimately befriend like minded musicians through SC messaging and we'd collab or remix each others stuff. Not a thing on Bandcamp, or not as much. Everything lives outside of their network - in Facebook chats or Reddit threads or some random forum.

However, despite its shortcomings, Bandcamp is where cutting edge music lives in an age without Soundcloud. Labels like PC Music, Mall Music and Dream Catalogue have all pushed independent music to bizarre new limits in the last 4 years alone. Orchid tapes, an indie outfit, made their name off of Bandcamp releases. Alex G, Katie Dey, Elvis Depressedy, and a slew of other musicians have been recognized for their indie releases and now enjoy moderate indie popularity. As long as no one else is beating them, Bandcamp will continue to run on its skeleton frame and empower artists who use it.

I think bandcamp didn't take money - so they can operate as they want. I totally agree with you with respect to bandcamp being where cutting edge music lives.

Doesn't True Ventures have a stake in BandCamp?

Bandcamp is where most of the bands I care about seem to release their music.

But, like you said, most of my discovery occurs on YouTube and a small forum thread.

Sounds like BandCamp is leaving a lot of opportunity on the table.

That's exactly what's going on. It could be, however, that the cost of that opportunity outweighs what they're working with right now. Building a social network is a ton of work and could be a lot of risk. Allowing users to upload homemade music and post it to other social media is apparently tried and true, as they've been doing it for 10 years. While Soundcloud went "full social media" and then ended up with losses and layoffs, it could be that Bandcamp has opted for a slower and more measured approached to growth.

Yes. But discovery and repeat visits are key. Bandcamp != #HookedBook :)

They don't have to go crazy. But something less than the current clinical experience would be a step in the right direction.

Def yeah they are insignificant, but OP said they ran numbers and used the word "successful". I was wondering if that meant something pretty impressive like in the range of $1M a week in rev or something closer to $1M a month in rev, which is still good, but really small.

Iterface not changing in years makes me think it's ripe for an acquisition from Barry Diller's IAC. They specialize in sites that don't update their interfaces for years at a time.

(Apologies to those that work for one of those companies.)


Unfortunately for SC, IAC tends to price acquisitions based on historical revenues...

Thanks. Fixed.

Meanwhile Bandcamp is killing it with discovery, something they used to be notoriously bad at.

I originally went to SC to find new music. If that's even possible, I haven't been able to crack that code. I've given up, and typically only end up from a link. Rarely by choice.

I was a director of an engineering department for SoundCloud for a time and left several years ago. Frankly none of this should be surprising: nobody there could figure out a product definition nor a monetization plan. What is more: the engineering organization at SoundCloud at large was completely fractured and suffered from in-fighting. Nobody trusted anyone across organizations. There was no shortage of bullying even. Taking that in mind, SoundCloud's downfall was to be expected. Because of caustic environment even before the financial crunch, they couldn't execute. Blame the incestuous Brazilian boys club there for it (I'm male, by the way); they bullied just about everyone.

As for Berlin, I am sorry to say: I had high-hopes for the tech culture there, but to call it Mickey Mouse would be a charitable statement. It was laden with confused hipsters who couldn't differentiate between language du jour and its monads and delivering a product.

When I applied, the warning signs were strong. Nevertheless I ignored them — to my own peril. Needless to say, I won't make those mistakes again!

Right before I resigned, it was revealed in a private leads meeting that 18 percent of the engineering force had resigned in that given quarter. Was I surprised? Not in the slightest. That knowledge gave me resolve to get out, which I hadn't yet announced.

I was a long-time SC engineer and can confirm, the influx of Brazilians and especially former ThoughtWorks folks was the inflection point in the destruction of the engineering organization. Conniving and intensely political, they absolutely ruined what was a compassionate, innovative, and productive culture.

Could you go into a bit more detail ? What did the Brazilians do / change ? As for ThoughtWorks, I've been at an interview with them and got a strong cultish vibe from the whole thing. Especially when they made it clear that their 'social responsibility pillar' is cool and all, but "we're still a commercial company that needs to make a profit so don't imagine you'll be doing charity here"... I'm curious how the people from that environment affected SC's culture.

I'm sure the company had a much nicer culture when they started, given that their product attracted so much original content.. So what went wrong ?

I happen to be one of the "incestuous" Brazilian boys. Also, I came from ThoughtWorks. I doubt thowaway999 considers me one of the "bad ones", but I do need to give my two cents here.

1. I fully agree about the fractured organisation with lots of in-fighting.

2. I relate with the statement about "confused hipsters who couldn't differentiate between language du jour and its monads and delivering a product". I think it was more a matter of very clever yet immature kids that loved to play around with eccentric language features. That was not the root of any problem, though. The problem was lack of leadership to curb the in-fighting and give the engineers some direction so that they don't get lost on their drive to experiment around with whatever they feel like it.

3. It's ironical to blame the ex-TWers Brazilians _and_ the monad-loving hipsters. From my perspective, these were distinct groups. The Brazilians were not the stronger advocates of monads and other Scala typing tricks. Quite the opposite. The few Brazilians that were more fond of Scala were not ex-TWers. In short, simplifying the blame to one nationality is very short-sighted.

4. I'm very curious about who we supposedly bully. Or who was bullied at all, for that matter. Perhaps, I was too far from the director ranks to witness that. I'd imagine a director could do something about lowly engineers bullying people.

5. About ThoughtWorks, there's another misconception here. There's a strong cult-like culture there, yes. It happens that the former ThoughtWorkers that joined SoundCloud were exactly the ones that we dissatisfied with the cult, and joined SC in search for a better work culture. The ones that I knew personally have some quite strong feelings _against_ the "social justice" hypocrisy that is rampant at TW.

6. How were we "incestuous"?

I apologize about calling out the entire Brazilian group by name. That was unfair. Not all of you were bad. I think if you asked folks at large anonymously they would say that their was a whiff of nepotistic networking privilege in whom was hired, why certain immature acts were tolerated from members of the group, and why certain favored outcomes occurred to the Brazilian network disproportionally.

no need to apologise. By no means I read your post as an attack to Brazilians in general. It was at a particular group. That's fine. I just think it's an incorrect assessment.

I concur with the nepotistic networking in part. Everyone tends to refer people they know personally. Our network was strong enough to become some sort of inner joke. I doubt that this was what brought SC down. Immature acts were not restricted to us.

Can you go more into detail about the Thoughtworks culture? I applied there while in Chicago and got a strong sense of that from the culture portion of the interview (after the programming challenge).

Yeah, that's pretty much dead give-away that you're in some kind of stealth cult environment:

"We're all big believers in X here. But actually, we don't really believe in X"

My favorite example for the above:

"Great news, bro - we offer unlimited vacation! But actually you aren't expected to take more than 10 days of vacation per year."

I've met many decent people who work at ThoughtWorks, so it pains me to say this, but based off the interactions I've had with the organization professionally I've started to ask in interviews if the company does any work with TW (and a couple other firms, Pivotal has a poor track record as well). If so, the conversation usually ends there unless it's followed by a qualifier where the interviewer understands my reticence. I've been at two companies where TW involvement torpedoed projects, the culture, and even a sound business model. Not again.

This is interesting, as both these companies have high reputations and at times low reputations depending on your perspective. All consulting companies have their mix of failures and successes. Is this a correlation/causation thing, or do you truly think Pivotal or TW do something uniquely horrible to companies?

( I work for Pivotal but not Labs and am curious. )

I know the type. It's unfortunate that you have got a taste of the worst we have to offer. There are entire companies that are composed of nothing but these types. Most do not survive abroad and self select so you tend not to get the bad bunch – unless you reach a critical mass, then it becomes self-sustaining.

beware that this is only one side of the story.

I joined after the re-organisations that Phil mentions in the linked post here were well underway. From all I could tell, no one regrets it. At the position I head, I only heard rumours about the struggle that he went through trying to improve the engineering problems there.

Everyone (with perhaps a couple of exceptions) I know personally there (Brazilians or otherwise) was trying very hard to improve things. Their efforts could have been misdirected. Their strategy could have been even counter-productive. The intentions were good, though. And I did see some ill-intentioned people in other places.

I even think that thowaway999 was probably well-intentioned. However, an anonymous director calling people that he disagrees with "conniving and intensely political" and claiming a moral high ground is almost comical.

Note that I never used the term "conniving" myself nor called anyone by name. Someone else did, and that is on them.

For me what was most difficult and disappointing was not being able to rally peers or their teams to get the necessary work done. Rather they were too busy sniveling amongst themselves — too jaded from burnout, I suspect. I felt incapable in the role at first until I recognized that it was really a maturity problem that pervaded much of the engineering organization.

I knew it was time to call it quits pretty early when seeing that nailing myself to the cross wouldn't affect change but hung on for two years. I was able to help a few people along the way by shielding them from the bullpen and the drama and giving some good career development opportunities. Let's face it: my peers in the leadership circle were long burned out, and I was quickly becoming so myself.

I have nothing but respect for the founders. They had their work cut out for them. Never asked a whole lot from them but they gave it their sincere all when I did. Maybe a learning opportunity for them would be having someone in their steed who was less conflict averse who could have helped referee the discord.

If anything, I wanted to caution folks to not be under any illusions with respect to SoundCloud's organizational and personnel context and work environment in Berlin. Everybody held it up as a panacea, which it sadly wasn't. Irrational exuberance is dangerous.

> Note that I never used the term "conniving" myself

That's true. I only realised it later. My bad.

beware that this is only one side of the story.

Hi, former SoundCloud VP here. First I think it is highly inappropriate to use this sad event to blame certain nationalities/groups. I hope everybody agrees that SoundCloud is a much needed product and we all hope for its success. Some general points: Was there a lot of chaos at SC and were some people treated unfairly during that chaos? Yes, most certainly... I am not happy about it. But we were growing like crazy in headcount and every day we were afraid that the site was going down. Everybody tried to do their best to cope with it, but it was a huge challenge... People had different opinions how to deal with it, and that led to heated discussions and some fights. Then to come in from a established organisation and complain about it, is at the very least naive.

Growing engineering organizations is simply not a reproducible science. Successful companies are so rare and the conditions which led to their success so exclusive that it's much easier to point fingers at the people you disagree with. I think the parent captures this.

The backlash which is emerging now are really a products of pent up frustration. SoundCloud was never able to develop a culture which resolved differences. Lots of acquiescence, lots of political assassinations but no mechanism for reconciliation. That's where I point my finger at least.

ding ding ding. we have a winner!

when conflict cannot be meaningfully reconciled the whole system tumbles apart.

fwiw my critique focuses less on specific disagreements but the overall culture. i personally observed more innocent people eaten by this angry machine than necessary. that machine can't be the sole reason the company finds itself in this position today but friction from discord can't help an already uphill battle.

lesson number one: the no asshole rule is inviolable.

Is growing like crazy the issue? You lower the bar for hire and not enough time is given for the new employee to integrate into the culture. Instead new teams make their own cultures, and fight hard for their own success at the peril of company's success.

s/political assassinations/political maneuvering.

Maneuvering is a far better less inflammatory word here. Nobody died from disagreement working at soundcloud.

As they say in the German, der Fisch stinkt vom Kopf her (a fish rots from the head down). This man bears a tremendous amount of responsibility for everything described above.

Can you say why? I don't understand the link.

It's the fish?

I'm really curious to understand why a single person was responsible for a big failure. That sounded a bit personal.

I'm surprised that anyone architect or manager would blog that they're leaving a senior role after just 20 months. However when I've seen architects leaving a trail of destruction that looks familiar.

Everything mentioned in the linked article and his soundcloud post seems quite good. They greatly reduced new feature cycle time by bringing the right people together and having them work collaboratively rather than sequentially.

What were the unintended negative side-effects?

Hi, my name is Jano González, I work for SoundCloud and I'm from Chile, so it hits me when I see latino americans been stereotyped. I don't know if we ever met, but blaming things on people with a specific nationality it's super fucked up dude, please don't do that.

i apologize if i offended you or anyone else. the mention of "brazilian" is a relevant as a secondary detail. there was a group of problem engineers who were ring led. that group was brazilian. that they were latin american or brazilian is unimportant but a significant identifier of the social group. if they were there when you were you would know exactly who. otherwise it may have been confusing. prejudice is not at play. i was pointing out an organic association.

as i said they don't bear sole responsibility but rather signified the problem for the impunity of being able to break the no asshole rule. other immature apples spoiled the basket as well. plenty of innocent people suffered. none deserved that. burnout is evil. i hope we have all learned our lessons and become better people.

I worked for SoundCloud for about a year, I can attest that SoundCloud was a huge mess, mostly because the two founders (both Swedish) had no clue how to manage the business.

I think every nationality represented in the company gave a good contribution to make the place both good and bad.

I can't say the Brazilians there were an issue, but I can see how some folks, especially in the infrastructure team, wouldn't like them.

I personally had a lot more issues with German people, both in the company and outside of it, to be honest.

I hope you understand that you sound fairly angry and biased (no pun intended).

By being a director and having this sort of blaming outlook on other people, you most likely have a good share of responsibility for the downfall of the company.

I did my best to behave professionally every day of the job and help those around me. This is to say act better than the fray. Doing so for so long led to cognitive dissonance and personal burnout. My resignation was promptly tendered when I realized the burnout became so severe that I couldn't hold back from dispensing the most restrained of zinger, which was nothing compared to what was happening around me.

If I sound angry, it was because I sacrificed a lot of time, energy, and giving up a good life I had in other city for taking the job. Worse was realizing how much of a lost opportunity all of it was. We could have achieved more had we more trust and decency.

If you're looking for examples, consider scouring the annals of the web for "shit Soundclouders say" - a Tumblr feed of leaked statements from our colleagues. I don't know who put that together, but it certainly didn't help the trust or maturity aspects of day to day work. To give an example of the shit posted on their, English was a second language for the majority of us at the company. There was a fair bit of mockery for peculiarities about how people wrote in their second tongue - among memorable immature tantrums recorded on the Tumblr. Do you see where I am going?

I bear responsibility for not being more persuasive. I tried to get this damn Tumblr deleted but couldn't. Nobody cared - especially because it predated me and them. Everybody accepted that immaturity as a given.

Digital Ocean was smart to cut this short (the flock of brazilians from SoundCloud migrated there). Not all brazilians are like this tho - only the agilists.

Why were they smart to cut that short? What was wrong with those specific group?

Agilists? What has this to do with a preferred way of working?

Mostly a slang in brazil meaning people that use agile methods in ways to avoid real work. Not related to the agile manifest.

This all hits really close to home for me. I'm wondering if anyone else shares your opinion on the Berlin scene being "Mickey Mouse" and why?

I don't thing the Berlin scene per se is "Mickey Mouse" but there are a lot of companies that grew to fast with a insane amount of money. A lot of times the founders were installed as CEO and had little to none experience regarding leading a business. The were just happy to "make it" but had no clear vision what happens after the big money came in. Same for SC: They got a lot of money for a good idea (democratising music I would call it) but there was no clear vision or business model for the future just like OP calls it out. Some goes for a lot of other companies we've seen come and go. In the end Berlin is just not as mature as the valley but this is changing lately at least from what I see

Pretty much spot on.

Being a former Director at SC (Mobile), I highly doubt this a former director at SoundCloud (unless they joined after me). The Directors at SC were at least decent people that did not speak of the company or people like this. Please do not take this post seriously unless the person reveals themselves.

I stopped to read when you mentioned the origin of the guys... Sorry buddy there you lost the points of your argument


Heard people would disassemble his bed in hotels as retaliation

They also laid off the entire New York team and at least some individuals in Berlin:



> Well, SoundCloud just laid off all of its New York engineering

> Literally the entire payments and subscriptions team, ads-eng, monetization engineering, everybody

> Not really clear to me how the execs think this company will be able to make money from now on

That says a lot. German labor laws protect their citizens strongly. I'm pretty sure they would always choose to get rid of other country staff first.

Not only citizens, laws protect all workers

More likely the German engineers cost half as much as the New York ones. No health insurance and much lower CoL and market rates.

I think you're probably right about overall cost, but just a note: German employers do pay into the health insurance pools, too.

Yes but operational reasons such as we can't afford to pay them falls under a fair reason. Only issue could be they would have to a certain amount of notice, so it could be more costly.

It doesn't sound like they plan on running the company for very much longer.

This seems likely to me. Apparently the office staff and HR teams are down to 1 person each. Previously it was about 10-15 people.

I just quit the payments team with Pje a month ago and I can say no one saw this coming. The entire payments and subscriptions infrastructure is now unmanned. Pje was a veteran but most of the team was <6 months and we just moved into a new space. The numbers seemed good to me.

The silver lining here is that recruiters are in a blood frenzy; I've had ~30 messages come in today.

I did the back of the napkin math, and saw this coming a year ago. The numbers just didn't work out.

When the whole finance team quit late last year, I knew it wasn't good.

I got out of there earlier this year. I feel bad for the new hires that were just coming in. I heard that a atleast one was just moving from the US to Germany, their things are still in transit in a shipping container and they just got laid off.

How was it going with the Go launch? I've been dealing with it on the b2b end and it has been pretty painful.

thanks for your kind words, "bm1362". we'll meet again.

have someone lined up to buy them and removing duplicated jobs to make themselves more appealing?

Normally such layoffs are made when an acquisition closes.

They're done in preparation as well, in order to juke the burn rate numbers by reducing overhead. Make no mistake, an acquisition would likely result in further cuts (acq is always bad for the little guy) if not a complete wind-down, but cutting staff historically does make a company more attractive to acquirers.

I'm also hiring in New York for Time Out. I would like to think that engineers from Sound Cloud may be excited by the music, theater, film, and event domain. Drop me a line at [hn_username]@timeout.com

19 days ago an employee of SoundCloud posted an Ask HN about deferred salary reviews: https://news.ycombinator.com/item?id=14575627

It's on the list!

Ask HN: As an employee of a company, how do you assess its health? | https://news.ycombinator.com/item?id=14653564


Other indicators: delayed salary reviews

Things I hate about Soundcloud:

1. They haven't updated their iOS app seemingly in years. It lacks basic controls and makes it impossible to listen to more than a single song at a time.

2. The really shitty "artist engagement" sites that force you to follow 20 other social media accounts in order to get a download. I know this isn't Soundcloud directly, but they're the ones letting artists link off to these basically spam services through a button that promises a direct download. They could have easily required you to follow the artist before downloading directly to solve at least part of this problem.

3. Music discovery is absolute trash. Discovery on a platform like SC is even more important than a place like Spotify because there is so much underground talent that should be listened to. This was the entire point of SC and they largely failed at that.

4. I have seen countless artists on twitter calling out SC for taking down their own songs from their own accounts for copyright strikes. It also seems like their ability to work with artists in general is just utter garbage.

Why would you want to listen to multiple songs simultaneously?

I think OP means that there's no "queue" like Spotify has.


I'll add (5) - They haven't fixed a bug in their 'Spotlight' editing widget. Trying to drag and drop to re-sort your 5 highlighted Spotlight tracks in your profile always crashes Chrome, and has done so for years, despite me (and many others) filing multiple bug reports.

> Things I hate about Soundcloud

I actually liked SoundCloud a lot, before they introduced SoundCloud Go. It wasn't perfect, but since it was free I didn't have an issue to live with the flaws.

But they really burned all the goodwill they had with me and others when they introduced SoundCloud Go. Here's a list of the "features" of SoundCloud Go [0] and why it's difficult to accept them as a user:

1. Access the world’s largest music streaming catalog, a constantly expanding mix from established and emerging artists

It's not clear to me that any of this monthly subscription is going to the artists I like (e.g. like Patreon). So, I'm paying 10 Euros per month for you to run some servers while having a much smaller library than other streaming services?

I do not work in the legal or music industry, but it's my understanding that since most content on SoundCloud is indie remixes, they shouldn't have to pay royalties to the record labels because the content is not copyrighted by the record labels (legally remixes would fall under fair use). So unless I'm very wrong in this impression, SoundCloud bent over to the record labels because they didn't want to be sued out of existence like Grooveshark?

2. Full access to all 150M+ tracks

Versus 120M on a free account. I highly doubt this number, since after introducing SoundCloud Go, a lot of tracks from artists I listened to became "Previews" of 20-30 seconds long and you had to subscribe to SoundCloud Go to listen to the full song. It was pitched as a subscription to access additional/premium content, but this is not what it felt like as a user, seeing songs go from free to pay walled.

3. Offline listening

You updated the Android app to remove the caching option, where previously you could select 100% and have the app entirely cache the song offline, assuming you had listened to it at least once before. I used this feature to cache gigabytes of music on my phone so that once it was listened to on WiFi, I could go out and not consume my 3G data while listening to the playlist. That was a huge "fuck you" to users.

4. No ads

Install an ad blocker in your browser and never hear ads anyway.

5. Millions of premium SoundCloud Go+ tracks

Here's how Engadget covered the announcement:

"the subscription plan costs $10 a month and includes a library of additional content as well as the usual remixes, emerging artists and podcasts." [1]

Given what I've said in #2, this was not the user experience.


I'm sorry to the talented people who lost their jobs today, but I don't feel any sadness for SoundCloud. This hurt is entirely self inflicted. For any PM's reading, here's how to burn your reputation with users in 3 easy steps:

1. Introduce an expensive subscription service at the same price as your competitors, but don't offer as much content, and don't publicize that any of the money is going toward your content creators

2. Hobble your existing free product to force people to the paid service

3. Don't push any meaningful new features or notable bugfixes to your website or app in years (e.g. needing Flash to play a song, HTML5 which eats up 100% CPU so you can display... something)

Now I use Mixcloud. Not better for individual songs, but great for music discovery given all the podcasts/shows available. Their app is a bit meh, but there aren't any ads, and their website doesn't require Flash and doesn't use lots of CPU. Best of all: they've yet to screw over users by introducing a paid option.


[0] https://soundcloud.com/go

[1] https://www.engadget.com/2016/03/29/soundcloud-go-hands-on/

Mixcloud Founder here, thanks for the support and if you ever have issues reach out. We're working hard on the app, but we're a super small team, so bear with us!

> Mixcloud Founder here, thanks for the support and if you ever have issues reach out.

Wow, awesome! Never thought anyone from either SoundCloud or MixCloud would read this.

Short list of improvements I would love to see:

1. If an artist uploads a tracklist for their mix, support seeking directly to a track within the mix (similar to a cue file) via the app or website.

2. Introduce a better way to seek in the app. When the mix is 1-2 hours long and all I have is a seek bar the width of my phone screen and a fat thumb, it's very difficult to seek to somewhere specific in the podcast. Something like 10 sec forward and 30 sec back buttons would already be a huge win.

3. Sometimes the mobile app plays the mix but all you get is silence. This seems to happen most frequently in the middle of a mix when there are connectivity issues. Closing the app and opening it again when you have better connectivity still results in blank sections of the mix. Super weird and very annoying!

But otherwise, yeah, please keep it up! Mixcloud has basically replaced the other music discovery services for me because it's got the content I want and it just works.

> legally remixes would fall under fair use

No. Remixes are a derivative work. They are a copyright violation without permission from the owner. Fair use would be satire or an excerpt for the purpose of commentary. Remixing is just more involved sampling and Vanilla Ice can tell you how well that works out without paying the piper.

Remixing is not covered by fair use :/

> Remixing is not covered by fair use :/

Well, that's good to know and also super shitty. I guess like many other services, once they got the attention of the recording industry it was only a matter of time until they were forced into a licensing agreement.

Fair use is strange and doesn't have a very strong standing when it comes to IP

Your points are understandable. But what would your plan be to keep the doors open? They need to make money somehow. Not being sarcastic, am genuinely interested.

I'm a heavy SC user, listen to my "manually curated" (for lack of better words) timeline in background for hours a day (using global shortcuts in chrome to skip what I don't like), but what if I want to buy a song I like? I have to give my money to Bandcamp!

Also, 10 bucks to remove ads is ridiculous, their catalogue is tiny and with overwhelmingly obscure/indie music (at least for me), AND I have no idea if the artists will actually get any of it! Cut it in half, have some sort of revenue share and I'm in.

Why would you buy a song you like if you can listen to it for free?

> Why would you buy a song you like if you can listen to it for free?

Because you want to support the indie artist/label so they keep making music you like?

> Also, 10 bucks to remove ads is ridiculous, their catalogue is tiny and with overwhelmingly obscure/indie music (at least for me), AND I have no idea if the artists will actually get any of it! Cut it in half, have some sort of revenue share and I'm in.

100% this. I'm not going to pay that much just to remove ads and especially if that money is going toward big record labels instead of the indie artists/labels I actually listen to.

This is a perfect example of how strong software engineering from competitors can result in you getting completely left in the dust. YouTube and Spotify have great algorithms for suggesting new songs - I feel like when my current song is done, I have a really high chance of hearing another song that I've never heard, but will enjoy.

This is not the case on SoundCloud, where I'm almost always jolted back to reality when the next song comes on.

I cannot comment on Spotify, but I pretty much dislike all Youtube recommendations.

I get sometimes different version of the same song again followed by some other song that is also popular but has nothing to do with what I'm listening to followed by something else I "liked before" but obviously don't want to listen to since I already heard it many times. That is followed by a different version of the same song again finally eventually converging on pop music that I generally dislike.

SoundCloud on the other hand sometimes recommends interesting things that I like but have never heard of. Other times, it recommends things that are completely unrelated and that I don't particularly like. However, they're still new/interesting.

My biggest gripe with YT recommendations is that songs I hit dislike on and always browse away from are constantly placed in my autoplay queue. There are some songs out there that I just hate and there's just no way for me to avoid them.

I've never had this issue with Spotify - mostly because of how differently its recommendations work - and it's probably one of the biggest reasons I've abandoned YT for Spotify.

YT recommendations in general is pretty rubbish

Can vouch for Spotify. Moved from Soundcloud subscription to spotify 10/month. High quality music, great collection and great recommendation.

I kind of feel too much money and not a strong org structure destroyed sound cloud.

> YouTube and Spotify have great algorithms for suggesting new songs

Spotify recommendations are far from great. They crowd source based on popularity, Pandora's recommendation engine is far superior.

They do a lot more than that. I just looked at my Discover Weekly playlist and the second track is from someone with 5,500 monthly plays. That's not showing up anywhere based on popularity. I'd say about half the playlist is similarly sized artists.

My anecdata indicates the opposite, as far as quality of recommendations is concerned. Curious how matters of taste might mean that you can't accurately make sweeping generalizations like this.

Music preference is not something you can generalize to a group of people. For example, I may like one song from Taylor Swift but not anything else. You may be right for popular songs but where Pandora's genome project shines is recommending music you would never otherwise discover.

If your suggestions are based on crowd sourced data like play count, your recommendations will be further encouraged by recommendations and thus become obsolete over time.

Neither Youtube or Spotify make any real money. Recommendation algorithms are the least of Soundcloud's worries as the platform just doesn't have any real business plan and just can't play the "growth game" anymore.

There is just no real money in selling access to or hosting digital music.

Soundcloud used to be a place for DJs. They died when they tried to go mainstream instead of focusing on DJs and selling them services.

Spotify looses money because they have to pay massive royalties to the music industry. Hosting content that is not part of that oligopoly should in theory be feasible, but of course it doesn't attract the same amount of attention. I wouldn't say it is a foregone conclusion though.

> Youtube make any real money

I hope you mean specifically on music, because that comment is laughable.

> Recommendation algorithms are the least of Soundcloud's worries as the platform just doesn't have any real business plan and just can't play the "growth game" anymore.

This 1000%

YouTube has considerable revenue, but as far as I can tell they're still unprofitable.

YouTube as of present day is profitable. To me, and I'm sure Google thinks the same too, the data they've amassed and are exponentially amassing far outweigh in value any notion of profits. This is about playing the long game, not focusing on quarterly earnings.

Exactly this point came up in a conversation with non-technical friends recently, everyone agreed. The nice thing soundcloud has is the ability to host mixes, which can last hours.

And that‘s exactly what they should have focused on. I never saw them as a service for listening to tracks and albums or storing my collection. It‘s not a Spotify or Apple Music competitor. It’s a great place for DJ mixes, sets and live recording, though I think Mixcloud is the better product - unfortunately I see way less engagement over there. That might change when Soundcloud goes down.

Mixcloud founder here, thanks for the vote of confidence. One key difference is we internally measure success or engagement in terms of minutes listened not play counts, and actively optimise for this. So often the visible play metrics don't actually respresent true human engagement. It's something we're thinking about a lot internally right now, and we might start exposing it shortly. I suspect most people will be very surprised for the better when we do.

Nice to see you here.

Don’t get me wrong - I agree that minutes played is a much better metric than play count - in fact I always wondered why Soundcloud wouldn‘t go a few steps further and not only show minutes listened, but also where those minutes were spent within your mix, like a heatmap over your upload‘s waveform to see which sections were most popular or where listeners stopped listening.

However, regardless of the fact that you use different metrics, for me it feels much harder to find an audience at Mixcloud. It‘s not that I‘m very popular at Soundcloud, but the small circle of like-minded followers I found there doesn‘t seem to translate very well to Mixcloud. Again, that may change with Soundcloud‘s demise.

That's fair and we're highly focussed on improving that this year. The groundwork is now laid, time to crunch out the features :)

That is the same internal goal that SoundCloud has. Play counts is just what users want to see. All of SoundCloud's internal optimization is around listening time.

The nice thing SoundCloud has is white-label releases

Yet they still don't even scratch what last.fm used to be when it came to recommendation.

How do you get that to work? I'm not on Spotify, but I use YouTube (Red subscriber in fact) and I find the suggestions to be pretty bad. It just seems to suggest random songs in the same genre mixed with stuff I've listened to before.

I generally don't subscribe to channels or upvote/downvote songs. Is that critical to getting good suggestions? I would have thought just listening history would be enough.

this is really the dumbest possible take on the whole situation.

Soundcloud started dying when they decided to be more like Spotify and less like the independent music discovery service they started out as. Now they have an app and experience that is not particularly good at either.

I've worked at a few places that died this way. CEO: "We're #1 in $NICHE. But growth is the only way, so we need to compete in $BIG_MARKET." [months of de-focus later] CEO: "We're only #10 in $BIG_MARKET and now we totally lost $NICHE!!" Resume writing ensues...

Expanding to $BIG_MARKET is not a bad thing, in fact this is the textbook example of "crossing the chasm", where you start with a niche and expand--if you try to go mainstream from the beginning you will always fail.

The problem when you're expanding is you need to have a clear idea of what you are. Because knowing what you are means you can scale by taking advantage of your strengths instead of dumbing it down. Amazon is a great example.

But if you just expand without knowing what you are, you'll end up like Soundcloud.

Also another thing is this idea of "what you are" (aka vision) shouldn't come from some media coverage or VCs or pundits. You should have already had the vision before these people started talking about you. See Snapchat for example, media pundits started calling them "a camera company", and Snapchat itself started believing the hype, and even renamed themselves "Snap". Snapchat is not a camera company. It's a social network that revolutionized private sharing. But they don't seem to think so anymore because they have to match the stratospheric expectation set by their IPO. Instead of expanding their own playing field they jumped into other giants' playground (competing with Apple in AR, competing with Instagram in public photo sharing, etc.), this pattern never ends well historically.

Soundcloud fell into the same trap. Their investor called them "The Youtube of Audio", instead of seeing it for what it is. If they had realized they were popular because they were a great service for indie musicians and focused on that aspect it is possible that by now they are more influential than any existing music tech companies. Instead they tried to become "the youtube of audio", which is the most uninteresting thing I've ever heard, it's the definition of "dumbing down" in order to expand.

> Their investor called them "The Youtube of Audio"

And now, of course, (and for quite a while) YouTube is the YouTube of audio.

It should have been apparent that it was foolish to try and beat google on that turf. Google beat them easily with a simpler interface and raw performance. The biggest difference, though: The audio quality on YouTube is actually better. Nail after nail in the coffin.

Ain't no way I'm listening to mixes on YT when they put brickwall-compressed ads cutting in on un-normalized audio every 10 minutes. It's a non-starter; when I see those little yellow dots I hit the back button.

Have you used YouTube Red / the YouTube music app?

God no. The only way I'll consider paying YT a single red cent for anything they make is if I can permanently turn off annotations on videos, and I haven't been able to find anything that indicates a yes or no to that question, so I have to assume their stance is "GFY."

You can though. I've had them turned off for too long a time to remember what the setting was though.

That's a good point, actually - I've been wondering what I can do with my music that's hosted on SC and putting them on YouTube (with some generated visuals, I guess) might work.


I approve of the dollar prefixed variable names.

Exactly - I'm a hobbyist musician and have all my music there. But, the last year or so they've been focusing on podcasting, etc. and, as you say, it's not good at any of that. For people like me, they haven't offered any new features in a long time.

Oddly enough, when I first heard of SoundCloud, it was in a completely different context from either. I first saw it used as a pastebin for voice recordings.

Specifically, members of some online transgender communities use it to solicit community feedback on their voices. They'll record themselves talking, post it to SoundCloud, and ask the community for input on what they could do better.

I don't think there's enough of a market for "pastebin for voice recordings" to be a successful business though.

Pastebin for voice recordings is basically what vocaroo.com is, and I discovered that in the same context you discovered SoundCloud. And also because popular use by 4chan.

When I first heard of SoundCloud it was mostly used by self-releasing EDM artists as a place to upload their music, before YouTube was a popular alternative. Nowadays I see a lot of artists buying visualizer templates and uploading their music with that - or sending it to an aggregator on YouTube who displays a certain style of music (many of these aggregators don't seek artist permission and just upload popular songs and such, but many of the ones I follow get things sent to them by indie artists because a channel with 60,000+ subs will beat their own personal channel for recognition).

Wasn't the reason for the shift to try and make enough money to support themselves? Free Independent music is not really sustainable

ITT no one admitting that the problem with the recorded music industry is now the fans who expect bits that make sound to be free while other bits are paid.

Spotify, Pandora, and Soundcloud fixed just about everything that people said was wrong with the industry from the consumer point of view. And don't give me that line about the labels screwing artists because thanks to fans who won't even pay for a premium subscription, Spotify literally can't afford to pay artists more without going under entirely.

TL;DR - music doesn't want to be free. Selfish people want it to be free.

It's not the artists who are struggling, it's the platform. By your logic, I'm oppressing Youtube creators because I didn't buy Youtube Red.

In terms of artists posting on Soundcloud, the fact that there's any demand for Soundcloud at all pretty much torpedoes this entire worldview you're espousing.

It is extremely common for artists to simply give away mixtapes on Soundcloud. When/if Soundcloud is gone, those same artists will continue to release on p2p networks. It is bizarre and delusional to believe that some valley person has to be making money off of it for me to listen to the music I like.

People sell things to make money. People make things for all kinds of reasons. Music is fundamentally an artistic venture, and everyone who tries to turn it into this other thing can frankly get completely bent imho.

You might enjoy this Twitter account:


It doesn't seem like a controversial position for those of us who enjoy music to wish there were some professionals making it.

If we are going to have professionals they must make money some how & the current status quo requires them to be constantly touring.

Talk to them and this is taking its toll both in longevity & album quality. I listen to albums a lot more than live concerts so I find this concerning.

> I listen to albums a lot more than live concerts so I find this concerning.

Or .. disconcerting?

How dare you.

I take it you've never had to buy pro-audio equipment?

The capital investment required to make music these days essentially negligible unless your heart is set on setting up a studio so you can have a band like it's the 1980s.

You are right, people are cheap. However, some of the blame lies with these companies. They have taken a growth at all costs > figure out monetization later path to their own doom. You can't go from free unlimited streaming to paid limited streaming once you realize your costs are killing you. The recording industry does keep raising the prices (rightfully so?) making it harder to stay afloat. There is still a win here for the person with the right vision.

I have no idea how you got that notion, but as a previous paying spotify customer I quit because they consistently screwed up their UI, made pointless changes and kept forcing "social" into something that shouldn't be social in the first place.

I now use google music. It is still no way near as good as groveshark was (especially when it comes to having all the music and a simple interface) but it works pretty okay.

And of course they can afford to pay the artists: I get to pay way more to the music industry now, than I ever did before.

I love google music because their "Radio" feature is incredibly useful. So if I like a song, there is a "radio" option that plays similar songs, which has worked surprisingly well (imo). Maybe their AI is just better or something, but I'm a very happy customer so far.

The only thing I fear is that they will shut the service one day for no good reason (because they have done this before). But since its a core part of their android offering I'm hoping it won't happen.

Also any Google iOS devs reading this: PLEASE ADD WATCH INTEGRATION. I love to listen to music while running, but currently there is no watch integration, so all I can do is play the next/previous song through my watch. Definitely more control would help :).

I use Google Music to listen to music, and when I find an album I like I buy it on Vinyl, and also attend concerts when they are nearby. More than I have spent on music than ever before.

I keep thinking about this problem, and music isn't the only industry affected.

I wonder if this is us hitting some of the limits of capitalism? It's not that I mind paying, in fact, I want to support artists I like. But the entire ecosystem of media is set up in a way that it is just more efficient for everyone to get stuff for free.

I feel like financial transactions get in the way more than they help, but this is obviously from the consumer point of view.

I get it. You feel like maybe we're hitting some fundamental limitation of capitalism, whereby deserving artists aren't properly supported by the economic ecosystem around them. You're right in the media ecosystem is weird here. Music is relatively unique in that it is something that's easy to consume more or less piecemeal. Nobody reads only 1/12th of a book or watches 3/10ths of a movie, but these are common in how people relate to music.

But what if there's another possibility?

I think we're expecting the future and present of music to not look like the past of music. Music, like pretty much all forms of art, has a small number of professionals and a large number of very talented non-professionals.

What's changed is how professionals attain that status. Once upon a time it was impress something really rich enough to be hired on permanently. Then this complex system involving record labels came into being. Now we have a couple of different viable models at hand, and we have yet to settle on one as the dominant new model.

Couple this with a lowered barrier to entry and generally increased competition for entertainment dollars (which compete with mobile games, ebooks, etc.), and users have good reason to favor subscription models. Artists have yet to puzzle out how best to interact with this, but I have faith people will get there.

I wouldn't call it a limit of capitalism. I would call it a limit of the old music business model.

Take this with a grain of salt as economics is not my field but...

>You feel like maybe we're hitting some fundamental limitation of capitalism, whereby deserving artists aren't properly supported by the economic ecosystem around them.

It's actually the opposite. I would like the artists to get paid, but it's more that the payment barrier seems like a waste of time and energy.

> Couple this with a lowered barrier to entry and generally increased competition for entertainment dollars (which compete with mobile games, ebooks, etc.), and users have good reason to favor subscription models. Artists have yet to puzzle out how best to interact with this, but I have faith people will get there.

This is what I mean by seeing the limits. The subscription model makes no sense to me. It's inefficient at best, and it's clearly not working out.

It's not that I mind paying, it's that I don't want the work-flow for accessing new content to be limited by a payment barrier. But capitalism encourages the growth of artificial barriers for payment. It's just easier and a better use of time to bypass the payment method, even a subscription based one. It's more efficient for everyone if we could figure out a payment method and avoid that artificial barrier.

So far, a subscription model is as close as we've come. It allows users to deal with a minimum of payment friction a minimal number of times while still paying.

Any time your business is exchanging information for money - music, video games, movies, etc - there's always an incentive to circumvent it. The approach thus far has been to make it convenient, easy, and cheap enough that users are willing to pay in some fashion.

I am deeply curious what you imagine this artificial-barrier-free experience that still coerces payments to be like, however.

A subscription model is the best that I've see so far, but it's still a shitty system, and it seems fundamentally so, IMO.

> Any time your business is exchanging information for money - music, video games, movies, etc - there's always an incentive to circumvent it.

Sure, but the new incentive is that it is currently a better product/service in almost every aspect. It used to be that circumvention was to save money, but that's not why a large number of people do it now. The service is better, the quality is better, the selection is better, etc. Maybe I'm missing something, but other than "the artist gets paid" I can't think of a single thing that going through an official route makes better.

> I am deeply curious what you imagine this artificial-barrier-free experience that still coerces payments to be like, however.

I have no idea, that's why I'm wondering if we've hit a limit to our current system. And it's not just music, basically anything that's digitally distributed is running into the same limit.

I believe you may be underplaying the role time plays in these matters.

Curation, collection, access, and reliability are all value-adds. In my student days as a profligate music pirate, I had to put a non-zero amount of time into making sure the stuff I'd torrented was of a reasonable quality, tagged in compliance with my schema, and accessible where I wanted it to be. I was willing to do that because $10/mo was a lot of money to me at the time and I had time to spare.

Today, a new album from a band I like is one click and two seconds away from being actively in my eardrums. $10/mo is no longer a lot of money to me, and worth paying to avoid an hour of work a month.

> Curation, collection, access, and reliability are all value-adds.

That's even more of an argument that our current system isn't working. You might want to look into the current non-official music scene. Non-official sources of music come way better organized and faster than the legit sources that I've bought from. Tags are done correctly, quality is much better, and it's online before a lot of sites put it up. I've also had music that I paid for disappear because the company I "bought" it from went away.

Having a middle-man for this transaction has mostly been a waste of time for me.

Why must the model of terrestrial radio always be left out of "customers are cheapskates" harangues?

Because it's a an irrelevant, circumstantial argument that is largely made-up to begin with.

No, it isn't. Billboard released their first charts in 1912 for Top 10 sheet music, but people have been using radio to listen to music since the early 1920s. The foundation of the contemporary music industry is based in free radio.

Their Github org seems to have gone from 45 to 37 members. Not sure what that means, do they only have 37 devs?



edit: now it's at 35... ouch

I wonder why they decided to remove them from GitHub already [1] when they were given a 90 days notice [2].

[1] https://twitter.com/pje_txt/status/882983954059595776 [2] https://twitter.com/pje_txt/status/882981765064871937

Often times (in the US) the 90 days notice doesn't mean anything badge/credentials wise. You are walked out that evening (sometimes supervised) and your credentials are revoked asap.

WARN layoffs in NY require 90 days protection.


Maybe to limit the damage notified (and thus possibly upset) employees can cause.

That must've been an interesting calculation. Now they're paying dozens of employees that literally aren't able to work three months worth of salary.

Severance is pretty common, and some companies will just keep you technically employed (and thus paid, and even given health insurance) with no expectation of you doing work or coming into the office. Three months is definitely on the generous side, but not that unusual from what I've seen and heard.

Yep. Severance is pretty common when employees are "laid off" i.e. their employment is terminated not because of employees fault. And generally, you are informed of it, and your credentials and such are revoked almost immediately.

Quite different from when you voluntarily leave, when you give notice and do knowledge transfer and such. I imagine that is because the employee is unlikely to do anything bad, since they likely have a better future.

They're going to place work second anyway (to finding a new job). It's usually to give them a severance, not screw folks on a visa, etc.

Spot on. You never know who might have taken the news in a bad way and decides to lash out.

The default settings for my org were private membership and I hadn't realized it until later on. There may be more developers who just are private

those are just the people who publically show they're part of that org.

yeah. You have to chose to be publicly visible as part of an org. I can assure you they have more than 35 devs.

It was never more than 100 at a time tho.

With these layoffs, I'm guessing it's hovering around 60

They are very thin in engineering as is my understanding.

I heard it was disabled during/before an all hands this morning.

people have to choose to display publicly that they are part of a github org. I can assure you they have more than 35 devs.

SoundCloud is my go-to for music while programming. I've spent countless hours curating my likes, playlists, follows, etc. I'd be quite upset if the service shuts down, 1) for the artists that have gained large followings through the service and 2) for all the personal time spent that will in the end mean nothing.

I think I'm reiterating what has been said before, but the reposting is horrendous. It's made the listening experience quite poor from just using the activity stream. I've also been really unimpressed with the lack of track uniqueness - if 2 artists repost the same song, it'll show up in my stream twice. Even more frustrating was the lack of uniqueness between tracks & playlists, where one could conceivably listen to the same song multiple times in a row because artists would post the track and then a single-track playlist with that track inside it.

The UI is also lacking for quickly adding to playlists, etc. The simplicity was a feature, not a bug, and the power of SoundCloud has been their artist community.

Same here. I pay for the subscription so I can cache mixes I find, if I lose those it's gonna be pretty devastating.

Wouldn't be impossible to go rip them off the internet somewhere else but would require a lot of extra work etc.

Soundcloud should have stuck with boosting amateurs rather than play for the big guys and funded with subscriptions from the bottom up in my opinion. Leave the professional track streaming to spotify, this was youtube for sounds, a wild west just short of Grooveshark bootleg territory. Deep down, we know the ride is coming to an end and when soundcloud retires it will be missed.

command line youtube-dl will download anything from soundcloud you point it at, tracks, playlists or users

I would unfollow an artist that was reposting too much.

If your playlists are public, please share :)

Carbon-Based Lifeforms, Solar Fields, Aes Dana, Miktek, HUVA Network (this is actually Solar Fields and Aes Dana), Sync24 (a side project of Daniel from CBL)... I could go on, but there's a good start!

here's my go-to playlist - I added a few tracks every week. Mostly softer songs without lyrics:


For me as a music producer and consumer, one of the most frustrating changes recently made my Soundcloud has been their cancellation of groups, which used to be a decent means of discovering new music related to kinds of music you already like.

Once groups were cancelled, the listens on the songs I post to Soundcloud plummeted to virtually nothing, and I really don't bother to upload my music there anymore, nor to go to Soundcloud to discover new music.

I'm not a SoundCloud user, but as someone who just started using groups on Flickr i noticed that the views on my uploaded photos went from barely anything to at least a thousand per image (and with just a handful of groups). I can definitely see how this can break your audience. Seems like an incredibly dumb move on SC's part.

This shows why the modern trend of not buying music and just subscribing to streaming services is so short-sighted. The companies are all basically fly-by-night, and can change their product in a heartbeat, leaving you with something substantially different from what you're used to.

I use streaming services all the time, but this is why I buy albums from artists I like and keep a self-hosted plex library.

Should anything I like ever be taken down, I can still listen to it.

The good thing about music is that it is ultra portable so mixing music from many sources, services and providers is hardly a problem.

And you can clearly see that in how the market for online music works. Unlike video, which is so dysfunctional I'd be hard pressed to even say there is a market.

End result: despite streaming services I still spend $100s of dollars on music every year. 0 on video.

I think it's great you have a two-pronged approach (buying stuff you really like, so you're not out of luck if the streaming provider folds like this), but I have to question your assertion about online video: Netflix is clearly a success story I think. Maybe the only real standout, but still it exists. Lots of people are signed up for it. There's also Amazon video; I don't know how successful it is, but people do rent videos on there.

But those are closed services, closed silos letting me stream content through closed source apps running on a extremely tiny subset of the digital devices out there capable of playing digital video.

Hardly any service lets me buy and curate my own videos, without DRM, from several sources and build my own library which I can play on all digital devices out there.

It's just a few big giant services, each with their own DRM and limited platform support.

It's nowhere near a real market where I can pick and choose and providers compete on merit.

yup, my last.fm data is my insurance policy. Still enjoy Spotify quite a lot.

The problem with groups was it was used by 0.5% of users, and most of the playback was actually fake spam plays. The soundcloud spammers had all the gaming figured out on that part of the system and it was basically worthless.

I know, because I worked on getting it shutdown. Not because I disliked the idea myself, but with no developers working on fixing it, the feature needed to be removed in order to cleanup the database spam and a huge chunk of a legacy code base.

We wanted to replace it, but nobody could convince management to fund it with headcount. But that's the story of SoundCloud, fantastic levels of miss-management.

> Some rough news today for SoundCloud, the audio streaming site whose content is largely based around uploads from its 175 million users in 190 countries.

Does SoundCloud see itself as an audio streaming service now, as opposed to an audio hosting service? That seems like a focus shift away from where they started, and puts them up against behemoths like Spotify. Maybe that's related to why they're struggling.

Tangentially: does anyone have any idea of how Bandcamp is doing?

Love using Bandcamp.

Apparently great[1] and profitable since 2012[2].

[1] https://daily.bandcamp.com/2017/01/24/everything-is-terrific...

[2] http://5chicago.com/technology/bandcamp-profitable/

I am so pleased with Bandcamp, every time I buy music it is such a joy to see a link to Bandcamp.

I'm too very happy with Bandcamp, and it's usually the first place I check to see if an album I want is available, if it's not directly for sale on an artist's website.

That said, music is extremely portable, so it's easy to buy from several online shops and combine it all in your own library later.

Based on that I sometimes remind myself that, unlike streaming services, where you are forced to put all your eggs in one basket and can only have one source, with music sales there might be other options too. There's actually a free market out there!

That's extremely rare on today's internet, and I want to support that. Therefore I consciously try to buy stuff from other sources too, like Ninjashop, 7 digital, Waste, etc. But it's just very easy to come back to bandcamp.

For me as a producer is also a joy to sell my music through Bandcamp. The tools to release and monetize my content are far superior than Soundcloud, which in our label we use just to showcase singles without download option.

Yes, SoundCloud (IMO wrongfully) saw streaming services cut their market share (I believe it was actually product decisions.). So to get paid end user subscribers they offered streaming fully released albums and downloading them to play later.

This is completely at odds with securely streaming unreleased music, as a preview for the final album. They always have charged producers, and now are charging both sides for (in many cases) the same content.

This shifted their product focus and shafted producers who didn't have a major label deal that could afford the streaming SoundCloud partnership deal. Basically giving the finger to the early community that made them popular.

I would have focused on the producers and dissemination of their tracks that led to more and better discovery. I believe producers would pay a premium for that -- they already do with other forms of media delivery.

File hosting is just not a profitable enterprise. This pattern should be familiar by now: gain prominence by giving away tons of stuff to draw in users, and then clamp down when the bill comes due, resulting in a hard pivot that alienates the users.

The truth is that users are not loyal. If you are offering something for free, they will be happy to take it, and they will have no qualms about moving on when you stop giving them the free stuff that they came for. Selling eyeballs is a very tough business. People who want a realistic chance to make money should have plans besides "sell ads".

>File hosting is just not a profitable enterprise.

Yep. There's simply no money in just "pushing bits" anymore. That business has been consolidated by people like Google and Amazon.

You have to both provide a valuable service, as well as create a "network effect" around your product. Otherwise you're easily susceptible to being out competed.

Isn't facebook basically file hosting with a bunch of links?

Facebook is infrastructure. It provides the connections out to your personal network. Facebook's staying power comes from its ability to say "This is the best way to communicate with your personal network at any time". "File hosting" is just one subelement of that, because one of the things that you may want to share is a file.

Couldn't the whole internet be generalized as "file hosting with a bunch of links"? A "file host" is a company whose primary purpose is for you to upload a file and give the link to someone else; they're a middleman that exists only because a more convenient/direct means of distribution to the intended network isn't available. This does not make the company nearly loud enough to establish its own identity/user base.

Facebook has always been its own repository of people intelligence. It was never just an "upload your photos here". Such platforms, like Photobucket or Imgur, sometimes prosper for a while, until the communication channels that reach the intended audience directly offer something easier. (imgur is going down now as reddit introduced its own image host last year)

Yes, though it's 1) the largest such service with 2) some very sticky features.

(Not that I use it.)

Realise that the flipside of such services is their real customers -- the adverts side. For online advertising, the two largest agencies claim over 60% of the market, and are named "Google" and "Facebook".

Everyone else is an also-ran. Which for any ads purchaser means "a lot more work for a lot less reach".

Was actually just trying to create a counterpoint to the claim that file hosting services don't make money. Facebook is a hosting service for pictures and comments, and it makes money.

As far as I know Bandcamp is doing better than ever - they are also a completely different service, not very similar to Soundcloud at all. At Bandcamp you buy and truly own your music. While they offer streaming of those purchases, I‘d surprised if the vast majority of Bandcamp users doesn’t maintain and prefer their local music collection, which Bandcamp probably has contributed to a lot.

It‘s my impression that the Bandcamp userbase is very loyal and full of music enthusiasts - at. least I buy there weekly, about 5-10 releases.

I'm a musician who's planning on releasing an album through bandcamp, and I've been a fan of soundcloud since I discovered it.

bandcamp has a very different business model from soundcloud. it's set up to cater to artists selling music, and they take a cut of any music sold through bandcamp. This is a valuable service and they charge a straightforward fee for providing it. the downside of this model is that it assumes people will buy music. Which, to be clear, they very much do today. Streaming is popular but there are still tons of people actually buying music.

soundcloud's early business model was for artists to pay for more storage, or something, and in recent years, I'm honestly not sure what their business model is.

So, not surprising that bandcamp is profitable and doing well and has apparently not raised tons of money yet gotten to profitability, and also not surprising that soundcloud, which has taken a lot of VC capital, is not going to be betting on a business model that seems to be going away (but in the meantime is presumably quite profitable).

Bandcamp, Spotify, and Pandora are for the most part an afterthought to my social circle compared to SoundCloud. I'm genuinely surprised they aren't doing well; they have great engineers and have a fearless attitude that helped them reach a large audience quickly

That is so sad for Berlin as well. SoundCloud is one of the companies that made the startup scene an actual scene.

I know that SoundCloud's interview process is tough, so I guess it's a good day for companies looking for talent around Berlin.

very true. However, most of the layoffs seem to be in the US, for now...

I wrote an AWS Lambda thing that automatically pirates soundcloud tracks you like:


EDIT: I think it's broken right now for some reason but this should give you an idea of what to do.

Very sorry to hear! Apparently additionally also the engineers in New York: https://twitter.com/pje_txt/status/882977097232338947

SC had a unique product and at one point I really enjoyed discovering music. In fact, I had a music blog where I reviewed indie and folk music I found there. I found some incredible talents. For example, this dude Packwood from Australia. I ended up buying his albums on Bandcamp, but never would've found him without SC. Looking back it's surprising that SC never tried to get into selling music. I would've bought download for lots of the indie music I found; if for no other reason than I wanted to be able to listen to it in the event that SC folded.

I switched to Amazon music, which I pay ~ $6/mo. I couldn't be happier. I'm actually happy to pay money to have access to almost anything I ever want to listen to. they even have obscure Drum n bass albums (which surprised me). As great as AM is, it doesn't fully replace SC for me. No podcasts or indie music by random Australian dude without record deals.

One thing that they did right was their use of HTLM5 to keep what you're listening to playing while you browse the site. Especially for long running podcasts it's nice not to feel like you would drop the thing you were listening to if you weren't careful.

The question isn't 'Will SoundCloud close?' but instead 'When will SoundCloud close?'.

Just a very steady decline over the past two years.

Their talented team will easily land of their feet anywhere else, it's a shame that someone's so previously unique is fading away due to trying to compete with Spotify.

Removing Likes & Groups have really pissed off their community.

I don't think it will cease to exist. It might be acquired at a ridiculously low price even if it's just for the users and the content.

Awful news. But I'm glad to see that it's part of an attempt to stay independent - I hope Soundcloud sticks around, in control of its own destiny, for a long time.

Well, I mean, they're independent because no one was willing to buy them. Spotify and Google passed.


Yes, from the content of the article it seems more like an attempt to just stay alive.

As somebody who listens to SoundCloud almost daily I really hope they can turn things around. They should completely ditch the free plan, and just have a $2.99 or $3.99 plan. I am actually signing up for SoundCloud Go+ (offline listening) now.

I love SoundCloud, but the premium plan options are surprisingly confusing:

$4.99 - SoundCloud Go $9.99 - SoundCloud Go+

According to the SoundCloud Blog circa 2016 (and most external reviews), "SoundCloud Go" is the plan with the expanded catalogue: https://blog.soundcloud.com/2016/03/29/introducing-soundclou...

However, according to the SoundCloud Blog circa 2017, "SoundCloud Go+" is now the plan with the expanded catalogue: https://blog.soundcloud.com/2017/02/28/introducing-new-sound...

Is there some sort of "bait and switch" going on here?

I can only think that SoundCloud purposely shuffled the product names around to make us think that the $4.99 plan includes the expanded catalogue, comparable to Apple Music / Spotify. Or am I mistaken, and the $4.99 plan does include a catalogue comparable to Apple Music / Spotify?

Furthermore, both plans advertise more tracks (120M+ tracks) compared to Apple Music or Spotify (30M tracks), making it even more difficult to reconcile what you're getting. And the $4.99 price point is further complicated by Apple Music, which also offers a $4.99 price point (for students).

The product branding, pricing and positioning here is bonkers for anyone comparison shopping, reading external reviews or simply trying to understand what you get. It's easier to do nothing, and simply continue using the free service (and switch to Spotify from time to time to fill in the blanks).

Other sources say total employees are 420, so it's roughly a 40% cut. Ouch.

That sounds about right. When I left there 1-2 years ago what I heard was, that SC would be profitable if they were running on 1/3 of the engineers they had at the time.

420 still seems like a rather unreasonably large number (to me) for something that is basically a file hosting service with a couple barely maintained mobile apps and a web site.

Quite a few artists I follow are on Soundcloud primarily for new material. I would happily pay for Soundcloud if they'd bother to update their app for Android Auto and Android TV.

What's up with reductionist posts like this every time a company announces layoffs? Scaling out a global business takes a lot more work than it does to run a no-SLA side project. And you hire people to build out businesses that don't yet exist. When you've overreached, or targets are consistently not being met, you re-org, or, if that isn't possible, you scale back.

> What's up with reductionist posts like this every time a company announces layoffs?

I hate people like this. "Oh I could build that in like a weekend with Go and React no need for 100s of devs..." Those statements only come from people who never built something even remotely in this size and have no idea how complex this stuff becomes when done proper.

Glad you hate people like me, that's very inciteful :-)

Bandcamp (a competing music platform) only has 47 employees: https://bandcamp.com/about

Well it comes from having been part of companies that were going through this explosive growth phase, and seeing the corrosive effect it can have on productivity if not managed properly. I've seen formerly efficient and productive and happy teams get completely demoralized as truck loads of new people are added but nothing new gets done.

I'm not saying this is exactly what happened there, but as a user of their product I'm disappointed because I simply did not see a growth in features or improvement quality that I would associate with a company that was scaling out like it apparently was?

420 is total employee count, only a fraction of that (i believe substantially below 100) are devs, which for a site of that size is adequate i guess. They also have quite a few engineers working with the data instead of product development.

What is the max dev (+ engineer) to other office worker ratio?

Even in tech companies with no physical product to get out the door it is easy to get a 1:10 ratio. You just need an army of sales, press, marketing, social media managing, customer service, HR, management, legal, design, print design, accounting, finance, office management and other random people with important jobs to do.

How about a 1:75 ratio?!?

With three offices you are going to need nine receptionists at a bare minimum if you are to cover reception for an hour either side of when everyone turns up. At a minimum the IT guys that plug people in and get them setup on 'Exchange' are of a similar number if all the offices are to be covered.

So there must be a 'dev to total headcount' ratio, a maximum sustainable upper number, any higher and the company is doomed to 'miss the web' and get totally stuck in 1990's ways of working. I have seen a 1:75 ratio 'tech company' and saw great opportunity in changing it, although a long way off the desired 1:10 or lower.

There must be companies where the ratio is the other way around. A programming concern with little of the normal sales/marketing requirements could be a 3:1 company.

AMA I've worked at SC and AWS (another service that gets criticized for its warts) so I'll help you understand.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact