Make Your Music Easy to Use
With Metadata and Stems
Part One, TAXI Road Rally, 2018
Panelists (left to right) Steve Barden, Karl Richter, and Tim Bern are joined on stage for a quick post-panel shot with TAXI’s Michael Laskow.
Panelists: Karl Richter, Steve Barden, Tim Bern
Karl Richter is founder of Level Two Music, an Australian music supervision company with offices in Melbourne, Sydney, and Auckland. Level Two's commercial work spans the globe including on-airs in Singapore, Turkey, Asia, Pacific, South America, USA, UK, and Europe for brands such as Corona, Heineken, Vodafone, Virgin, Toyota, and Orange. While working as a music supervisor Karl founded DISCO, a music file sharing and workflow platform built specifically for the music industry. In beta for just over a year, DISCO now manages 8 million files for publishers, record labels, managers, artists, and music supervisors across five continents including Warp, Secretly Group, RocNation, Fox Sports, Concord, and Sony Interactive.
Steve Barden is a production music composer for film and television. His music can be heard on television somewhere in the world on a daily basis. His music has aired on ABC, CBS, NBC, FOX, and dozens of leading cable channels. He’s also scored several animated television series for Saban Entertainment, and recently joined composer Kevin Kiner in writing cues for Season 4 of Jane the Virgin, airing on the CW network. Steve is also the author of the book, Writing Production Music for TV – The Road to Success, available from Hal Leonard.
Tim Bern started his music industry career in his teens as a jazz saxophonist and music teacher in New Zealand. A taste for international travel drew him toward Tour Management and eventually artist management, winning a few NZ Music Awards along the way (two Best Rock Albums for The Checks). After a number of years bouncing between New York and London he joined Level Two Music to run the Auckland office and focus on building ad agency sync relationships. In 2016 Tim moved to Los Angeles to launch DISCO (a cloud music software platform developed by Level Two Music) across the North American music industry, and here we are today.
Thank you for being here and doing this, gentlemen!
I firmly believe that there are two secrets to success in the music industry.
- Making the right kind of music
- Making it easy for people to work with you.
Bill Gates calls that ease of use, or ease of working together, “frictionless,” obviously meaning, no friction… the wheels turn smoothly! Let’s see if we can help our audience members learn how to do that today!
Tim, can you please explain what metadata is, and why it’s so important.
Tim: Certainly. We are asked every day, “What is metadata and how am I supposed to use it?” The most basic idea is that it is information that travels with the files as you send them. So you’ve got an audio file—and there are different types of files like WAV, AIFF, MP3—some of them carry major data and others don’t. So just to show you really simply—you are all familiar with the old iTunes here (on the screen in the ballroom during the panel)—if I open up the info on the particular track and we look at the file… Can you all see the screen clearly enough? This is a WAV file. So all of these fields are what we would refer to as metadata, and it’s all the specific information that you might want to say about the track so that it travels with it.
If I send this file to somebody else across the internet or on a CD and they pull it on to their computer, if they open it up on a system like iTunes or DISCO or Meta, then all of those systems understand that the title field is the title, the artist field is the artist, and the album field is the album. And it’s the same across every different metadata reading system or writing system. Now, the issue with WAV files is that they don’t carry any metadata. So right now I can put in some information here and it’s going to look like I’ve added metadata. So I’ve added this metadata in iTunes to the file, but if I drag this… Am I getting too deep right now (he asks the audience in the ballroom)? But if I dragged this across to Meta, which is another metadata reading platform, it’s only showing me the file name. It’s not showing me the title or any of these other fields. If I come back over here, this is an AIFF. AIFF is also a lossless file type, which means it’s not compressed. It’s a big file. MP3s are small, they have been compressed, which means a lot of the actual information has been stripped out of them. MP3s are just less quality, and you can hear it normally if you put on a low quality MP3. But AIFF is high res, just like a WAV file is, but the beautiful thing is that it carries metadata. Now, see over here (on the screen), I’ve just dragged it across and all these fields are full, which means that the metadata traveled with this track between two different metadata readers. So, MP3s have all this metadata in them, and if I drag that across, it also carries that information.
So metadata is information on files—and I will go into how you use that with different fields soon—but the point is you need it to travel when you’re sending your files to somebody else. Because if I just sent the song “Run Wild” to (fellow panelist) Karl here, who is looking for a song for a sync for Toyota, and he loves the track and he downloads it and he’s got it on his computer, he has no idea where it came from! There is no information on the files that he can go back to; he has to search through his emails to try and find out where it came from, or through many different systems. But if I send him this AIFF, it has all the information in there so he can easily find it when he looks at that information. Does that make sense? It’s data on a file.
It’s basically like a dog tag. You know, if you lose your dog, they know the dog’s name, they know the owner’s name, they know the address or the phone number or email address to return the dog. So if you think of the music files as the dog, the metadata is the dog tag.
Tim: Exactly! [audience laugher]
Karl, as a music supervisor, I know you listen to tons of music for any given spots that you’re working on. Give us some examples of how you use metadata from the music supervisor perspective.
Karl: I guess you look at metadata in a number of different ways, and generally it’s the pointy end of any kind of creative or commercial moment that the metadata being there is going to be very significant for both us (the music supervisors) and for you as the musicians. So from a purely commercial point of view—and Tim touched on this — invariably, if I am liking a song, and there is generally going to be a bit of a time frame or deadline around that, so I need to know where that song comes from. I need to know how to get in contact with the person or the others involved. I also need to know with a song that I only need to make one phone call or one email about who controls the rights on the song. And then, I also need to know, if it isn’t an all-in that is created on both sides (the composition and the master recording), who actually controls the rights across all of the different rights. There is a lot of information there that can be exceptionally useful for me. If there are any gaps or concerns, and I simply don’t have the time for all the concerns as to who has the rights attached to that song, invariably, then it’s time to move on to something else. And that’s because with you guys (musicians in the audience), if that lack of information means that I have to move on to something else, you miss out on that sync—if a song goes through (gets the placement) and there is some sort of problem with the copyright attached to it, there’s a much bigger commercial impact, and it impacts on a number of different people at our end. It becomes a really significant thing we have to look into.
So something as seemingly meaningless or simple as faulty metadata, or lack of metadata, could result in a lawsuit later.
Karl: That has happened. This isn’t make-it-up time, this is stuff that has actually gone down, and the thing is that I only need to have that happen once (to cause a music licensing disaster or potential lawsuit).
The other thing I would say is that you cannot put too much contact information into your metadata. You can put down your family relatives’ phone numbers. [audience laughter] You literally should put down as many different ways that you can get ahold of someone, particularly for us in Australia, where we are working on a different time zone. In L.A., that works well, but if we’re trying to get ahold of someone in New York, or trying to get someone in the U.K., if you are comfortable putting a cell number in there, then you should do that.
The other thing that we look at when we’re looking at the metadata from the rights owner’s perspective, is that it will tell us a little bit about what that copyright is likely to cost and whether or not on a tight budget job it’s something that might be too expensive. Having that information at a glance can also be a time saver for all parties involved.
It might also be that particular publishers might be harder to get ahold of or harder to deal with, or that you might have had some issues with in the past. If I know by looking at the metadata that they might only have 15% of that song, but I need to know about that 15%, because that’s going to be the most painful 15% in my day.
Those are the considerations. If I can see that the song is clearable through one person or through one contact, then that immediately is an advantage for that particular track.
That’s one of the most important things that’s ever been said on that stage in 22 years of doing Road Rallies, it really is. I’m not trying to make a bigger deal out of this than I should, but the fact that you can look at metadata and see that there’s a big publisher on the song, you know instantly whether it’s in budget or out, and you may have had issues with a particular publisher and think to yourself, “I’m not going to go with this song just based on that information.” That’s really good insight.
This is a little off-topic, but I’ve been meaning to ask you this since we met via emails and everything, and I guess there’s no time like the present, because we’re in a room full of people. What was it as a music supervisor that caused you to go, “Somebody’s got to come up with a better way to organize, track, and send music and data files,” and come up with the idea for DISCO?
Karl: Probably about seven years ago, we had a team of six music supervisors and staff, and I saw that the majority of time was being spent on managing data files. So uploading and downloading music from Box, Dropbox, or WeTransfer, and that’s how their time was being spent. It was then being put into iTunes, which is a consumer product, it’s not a business enterprise product. I conveniently lost my iTunes files twice, and there was like 300,000 songs, or whatever it was. I got back the songs, but I did lose my playlists, and that’s really the engine room of the business. If you lose those, it’s an incredibly painful thing.
The other thing is that—and this is important for you guys in the room here—is that that old way of just doing busy work… work that’s not creative, does not add value, and is a repetitive task that gets in the way of the creative work, which is discovering your song and discovering a home for your song. So what it’s just done is really helping to unlock the value in copyrights that people might not know about. And that’s a really significant, important thing to do, because if you make music much more discoverable and you can make it much more frictionless… the competition of the music, the sharing of the music, the finding of the music and where it should go, then that’s when the magic happens (meaning that’s when music gets licensed!).
You are so right, and I’ve got to tell you. One of our members—I don’t know if you’re in the room—but Rebecca Maxwell sent me an email, and probably pestered me with a second or third email, and I’m so grateful that she did—saying, “Are you familiar with DISCO? Have you used it yet? And I said, “Yeah, I’m familiar with the name; I know a little bit about it. I think I’ve been to the website, but right now I’m far too busy to even think about that.” And she said I really should, in so many words. And at some point, we got connected with Tim, and he came over and did the demo at the office, and we could quickly see how much potential DISCO had for TAXI. And now that we have been using it for a couple of months, the benefit that it has created for us is exactly that. It’s gotten us back probably 20% of our man- or woman-hour time, and the A&R team at TAXI is able to use that time doing more outreach to find more opportunities for our members, because of you having that vision of streamlining the process. Thank you. [applause]
And frankly, I feel like we’re just starting to scratch the surface with DISCO. Right now, we’re just getting comfortable in our daily routine of using it. First of all, the most creative way we came up with was using it for the panels this weekend and being able to send the panelists the playlists of the stuff that was curated but not played onstage shortly after the Rally, rather than typically going through a bunch of CDs and… Everything about it really works.
Karl: And look, props to you guys as well… this is my first Road Rally, and I am being absolutely blown away by this positive, collaborate community that you’ve built here. It’s really empowering and really impressive. So thank you to everyone I’ve met and spoken to. It’s really been quite an insightful thing, and props to you Michael and what everybody in this room is building. It’s really quite incredible.
Thank you, Karl. I appreciate that. [applause]
Let’s talk about genre, moods, and the keywords should put in their metadata. Let’s start from the musician side—the composer side—and go to Steve with this. Can musicians get themselves in trouble by using sound-alikes or keywords that don’t really apply, when they’re entering their data?
Steve: Well, there’s a real danger with doing sound-alike names. There was a guy who made a song, and if he put in the keywords Marvin Gaye, you know what happened to him. So there’s an inherent risk. I mean, we are in the business of trying to sound like something else, but if you’re blatant, “Yeah, I sound like Tom Petty,” and it sounds like Tom Petty, there could be a risk of legal action against you.
Primarily, when I’m creating keywords for tracks that I’m turning in, the keywords are basically based on the moods or the emotion that the music is conveying. I don’t like to put, “It sounds like John Williams,” “It sounds like Hans Zimmer” as an instrumental composer. So I’m gonna stick with keywords that represent an emotion.
It is also my observation that many musicians don’t know what emotion or mood they’ve created. Is there a list of moods, genres or emotions that you’re aware of?
Steve: I don’t know, but you could certainly search the internet. I did compile a list, and this is not a definitive list. But what I will certainly do when I’m trying to figure out the ultimate list of emotions for that track, I’ll just put it on a loop and just listen to it constantly. I’ll just go through every single one. For example, I have in the happy/positive category, animated, beautiful, old, blissful, brassy, bright, brilliant, bubbly. So as the song is playing, I’m going, “Does it sound animated?” No. “Does it sound bold?” Yes. And then I’ll just jot those down and add to my list.
I think that’s great advice. And I would guess that adding a bunch of extra (and likely inaccurate) keywords in the file, hoping that your music will show up in more searches is a bad idea.
Steve: Exactly. It’s considered keyword spamming. So if you’re just doing it to attract eyeballs, they’re gonna lose faith in you that you’re representing what you say it represents. So you really have to be careful with that.
You have to put yourself in the place of the end-user when they are searching for a track. What words are they gonna put in? It’s kind of like when you’re on Google searching for something, what is your approach for searching? Are you gonna use the word blissful? Do you actually use that word in your everyday language?
Karl: Yeah, it’s about sort of reading your audience or end used as well.
So, to a large extent, knowing the lexicon of the industry you’re working in is incumbent on musicians if they’d like to increase their chances of being discovered and ultimately used in a TV, film, or commercial. It’s not just about making pretty music.
Don’t miss Part 2 of this panel in next month’s TAXI Transmitter!