Today marks the debut of Orfium, a music hosting platform and streaming service intended to empower its user base with a series of promotion, retail, licensing and distribution tools.
At a time when the fate of services like Beatport and SoundCloud are as dubious as ever, Orfium boasts the general functionality of both models. Moreover, its emphasis on licensing options and track recognition software might even give Spotify's upcoming DJ mix hosting service a run for its money. Perhaps most notably, Orfium offers content creators who choose to monetize music through its site a whopping 80% of revenue earned.
Orfium is the brainchild of Chris Mohoney and Drew Delis, both of whom worked on the intellectual property side of the entertainment business prior to their new endeavor. The co-founders made some time last night to speak with Magnetic Magazine at length about what might end up cornering a substantial chunk of the online music market.
Chris, Drew, please tell the story of how Orfius came together from the very beginning.
Chris: I had previously worked for a company that did rights management on YouTube, and also we had a music licensing platform - royalty-free licenses. Some of our customers were licensing music to YouTube videos. We spent a lot of time dealing with YouTube and we represented about 6 million copyrights on YouTube, and at the same time we had several friends who were musicians and users of SoundCloud.
They were talking about some of the problems they were having on SoundCloud, and I saw that there was a wider application in how YouTube was handling rights management. I thought that it was actually very smart - I don’t think that it was perfect, but I think it was very elegant the way that they had rights management set up with user-generated content.
Basically, they used an audio fingerprint, and whenever a match was detected in a video, advertisements would start to show around the video, and advertisement revenue would flow to the original rightsholder. A YouTube uploader could keep the video up, enjoy YouTube services for free, and generate entertainment for their audience. That’s their goal, and then the rightsholder collects money from it. Basically, it doesn’t have to be ad revenue, but being able to abstract the monetization from a digital asset and assign that to a proper rightsholder - I found that to be very interesting, and saw that as a solution to the remix issues that SoundCloud was having.
Also, at the time we had two businesses: one licensing, and one rights management. They were both separate, and sometimes we would have users who would upload to one but not the other, and we would let them know about our other service and see if they wanted to participate in that, but it was very fragmented. I felt like there should be a single upload gateway where an artist has the option to basically opt into any possible digital revenue revenue for their music.
There was an opportunity to create a more elegant, integrated platform that has advantages but also a nice front end for consumers, and lots of tools and options available to the artists in an open format that’s flexible with no legal risks. Every service offered is non exclusive with no contract terms so they can be terminated at any time - and they can individually customize it to their needs.
Every artist, record label and publisher has deals in place with other parties, and they really need something that they can customize and carve in whatever services they wanna opt in and opt out of to accommodate those other deals, so the idea was to create something that works for everybody and has music rights in mind from the beginning. That was one of the issues with SoundCloud; they started out as a sounds database and wasn’t really designed for music, and then they were in this place where everyone was uploading music but SoundCloud wasn’t capturing the right metadata to properly manage rights. Now they’re trying to clean it up, but it’s harder once you’ve already built this massive system.
Drew: I went to law school where I extensively studied entertainment rights and in particular digital music rights on the internet. I first got into the licensing thing, and that sort of spiraled down from there - but also, coming at it as a consumer, I myself was on Spotify and SoundCloud and was buying music from iTunes, and it was not that convenient. Why not have a platform where you can have all levels of artists from amateur to top-level music that accommodates everybody’s needs.
You don’t have to give your music away. You can really customize how your music is consumed, and that’s the core principle that Chris and I have tried to instill into every aspect of Orfium.
Chris: Drew and I both went to undergrad together at the University of San Diego. That’s how we met, and we’ve stayed in touch ever since.
Drew: I brought the legal and Chris brought the technical and music background with his YouTube work, and we decided we should build something that has a business-to-business component for artists but also a promotional side for consumers, and that’s where we are.
Even though you only soft launched in January before officially launching tomorrow, you’ve already got quite a few users, right?
Chris: We actually just checked today, and we just hit 70,000 tracks from about 22,000 artists.
You’re kidding! How do you think that happened, just word of mouth?
Chris: Yeah, we basically just reached out to artists on all different types of platforms and told them what we were doing. We told them that the system we were building would solve many of the problems they have, and we were having an open beta and needed help testing it. All these guys started uploading and gave us a lot of very nice feedback, and with that we made a lot of refinements, and with that we’re excited to launch version 1.0. You’ll see when it goes live tomorrow that it’s enhanced from the beta version, if you’ve taken a look at that.
A recurring theme in the dialogue about music streaming and discovery platforms is SoundCloud’s inadequacies. It’s kind of the MySpace of music streaming services. One thing you guys set out to improve upon with your platform was the search algorithm. Can you explain how it will be better?
Chris: Sure! First of all, in order to create a stronger search engine, we had to build in more structured metadata tags. When somebody uploads a song, not only do they pick what genre the song is - which, by the way, we spent a lot of time organizing as complete as possible of a hierarchy of genres and sub genres - but in addition to that, they can also tag moods, instruments, and vocals. Probably for most consumers, the moods and genres are all that’s relevant, but because we have a licensing component, somebody who’s looking for a song to license that has a particular instrument or vocal to it because they’re trying to synchronize it for a particular scene in some sort of creative work could use that.
So we have all that data from the uploaders, and then we use a special type of data structure that’s actually behind Google’s search engine and enables them to pull back results so quickly when given a search term. It’s actually a fairly simple technology; it’s just a different structure from a normal relational database. It just indexes your relational database and puts your data in a different text format that can be very, very rapidly searched and makes all kinds of very customizable and powerful queries.
Someone could search for a track on our site, choose a mood and a category, vocal, or instrument, and it’s powerful enough to find the tracks that have that and bring them up in a fraction of a second.
Drew: Also, I’m sure you’ve experienced where you go to search for something, and a song comes up and has 20 category tags. That, to us, was unacceptable. We decided to cap it so you only have three categories, one main one and two subcategories. We wanted the search results to be as highly relevant as possible.
Something that comes to mind is that electronic music has been around for over 30 years, and genres have changed meaning over time and from region to region. How will Orfium’s platform be prepared for that?
Chris: In our subcategories there are some regional ones, but if it proliferates and there’s a big distinction in the same category but in different regions, we could easily add region as another filter to put in a separate selection what region of a genre you want to listen to, and then put that in a separate category.
While we’re on the topic of genres, what are each of your favorites?
Chris: Me personally, I listen to a lot of minimalist progressive house. My two favorite artists are Eric Prydz and deadmau5. I just love how their music evolves and how they just keep adding layers, and how another layer makes it more complex as you go throughout the song. I find it very elegant how they tie all of that together. It’s almost like a symphony, in a way.
And then, of course, there are some related, smaller musicians. I’m always looking for similar stuff as well. With that Eric Prydz sound there’s been Fehrplay and Jeremy Olander, for instance.
Drew: I like more deep house, and then on the other end I’m more into rap and hip-hop.
When you say deep house do you mean deep house, or do you mean future house?
Drew: No, well, future house I guess.
There are a lot of comprehensive tools in place to help your user base to better understand licensing, correct?
Chris: Part of it begins when the artist uploads. There are a lot of options, and we worked for a while to try and simplify that selection as much as possible, and we’re now gonna be adding a bunch of tool kits which explain what each thing is. What we’re hoping is that through the platform a user will learn about all the different opportunities available to them. Whether they pursue these opportunities with us or they work through somebody else, I think it’ll be nice for them to learn about it and just know that it exists.
A lot of these guys are mainly focused on making music because that’s what they wanna do, but if we can simplify their understanding just so they’re aware of what sort of possibilities are out there and how they can monetize them, it can help the middle class of artists who are just trying to make a living make that possible so they don’t have to work second jobs and can just do this full time. It’ll explain on the licensing pages what the licenses are for - and that’s actually part of it, too. There’s a huge potential in the licensing business to make smaller, automated licensing details so that if somebody wants to make a YouTube video, school project, or any small project that wouldn’t be worth a publisher going after, it can automate, say, a license that ranges from $60-500.
Instead of having that on a separate website where a lot of people would never even think to look for it, there’s an advantage to having it in a consumer-facing website where they also buy and stream music because it’ll bring a lot more awareness to that concept. If you’re in a creative project where you’ll be using somebody else’s music, this way you’ll be doing it legally.
I’m sure you guys are well aware of Spotify’s recent partnership with Dubset in which the former company will use the latter’s proprietary technology to legally host DJ mixes on their streaming platform. It sounds like a lot of your recognition software will serve the same purposes, so how do you hope to have the upper hand?
Chris: I’m somewhat familiar with Dubset’s service. I’ve never used it, but I’ve looked around their site to see what it was about. It mainly appears to have been about derivative works like remixes and cover songs, and I don’t know what deal Spotify has in place with Dubset, but Dubset has to be making money somehow. I don’t know, but I’m guessing that means that after Spotify takes their cut, Dubset will take their cut, and then the content provider gets what’s left over. On Orfium, we plan to handle this in house and make sure that the same rate we would pay out for an original song is payed out to the original rightsholder.
Do you have any cross-platform integrations in the works for Orfium?
We plan to represent rights on YouTube as well. Artists or uploaders can upload their content on Orfium and check a box, and we’ll inject that content into YouTube fingerprinting where it’ll automatically match and monetize videos on that just for them. There’s that integration, and then we’re integrated right now with two inbound distributors who are uploading content directly to us. They came to us and were really interested in what we’re doing, and said, “Hey, we’d like to test it out and give you feedback.” We weren’t even planning on approaching distributors for quite some time and it actually came about much more quickly than we expected, so we created a pipeline for them and they can basically drag and drop their library, which gets uploaded by their clientele, and it’ll push that content onto Orfium, and on Orfium it’ll generate a profile for those artists if they’re not already on the site. If they are, we have a means for them to claim their profile and merge it so there’s no duplicate content.
Really, we’re trying to create a system that has something for everyone, and so I think the distributors have a lot to offer to artists because it allows them to upload their music to a place where it gets pushed to many other platforms - but if they want additional services as well they can come to Orfium and we’ll tweak their settings, too. So our current integrations are with the distributors, and we’re planning to integrate with more, but we just have some tools for them to finish up that’ll make things easier. We’re trying to create an embeddable claim button which another third party distributor could use in their client’s portal, and that client could just log in, click the button, and in one click be able to manage their Orfium profile and have everything merged together so there’s no duplication.
Drew: That includes duplication of stats. Artists don’t want to lose stat count, so we would merge tracks and profiles and retain all the stats so that way they can get credit for what they had.