About Geoff Koops and Mike Towber
People are consuming music at a higher rate than ever before. With the advent of streaming, people around the globe have access to giant catalogues of songs in all genres, available instantaneously on their phones, in their cars, through their laptop speakers and beyond. The competition for these users is fierce, and Rdio has carved out a special place for itself by combining innovative features with incredible design. Mike Towber and Geoff Koops, leaders of Rdio’s design team, are masters of creating interfaces that are functional, interesting yet incredibly simple.
Mike and Geoff hopped on their Soapboxes to share how music influences their design, the challenges of designing for different kinds of people and the best ways to gain insights about your users.
Designing for a Rapidly Evolving Industry
The music industry as a whole was turned on its head in the lately 90’s with the MP3 revolution. For years, the entire industry struggled to find new revenue streams and ways to monetize what essentially became a free product. Major players were dethroned and new forces entered the space, changing the ecosystem and writing new rules. Within the last few years, the introduction of streaming radio has further disrupted the status quo. This state of constant change and evolution compounded by the incredible diversity of music listeners presents a complicated set of problems for any designer working in the space.
Geoff referred to music listeners as falling on spectrum, with casual listeners on one end and meticulous music aficionados on the other, and dozens of other groups in between. Mike agreed that designing a single interface and product for such wildly different people was a challenge. He went on to explain that beyond the differences in preference and habits, all groups shared a common passion, a love for music that was undeniable.
To satisfy the needs of these varied groups, Mike and Geoff have developed a set of easily accessible features that cater to the listening styles of the user. No matter their place on the spectrum, there is a feature or features that are designed with them in mind, giving them a degree of customization to their experience.
One of the features that allows for this personalization is “Home,” a unique, dynamic landing area that evolves along with the user. Home shares a user’s activity, the activity of the user’s network and the best of Rdio.
Designing With Data
In order to deliver this tailored experience, the Rdio team pays careful attention to data collected about the user. Geoff likened this to the information we take in about people while we have conversation with them:
It's like getting to know somebody in real life. I'm asking direct questions in conversations, and certain facts will come up just in the natural flow of conversation, but also I'm listening to nonverbal cues and I'm taking context clues from things around us. Similarly, we're picking up on explicit signals.
Rdio pays attention to things like what a user is listening to, what songs they choose to skip and so on to develop a clear picture of the kind of listener you are. The goal of this attention to detail? To create a music listening experience that perfectly matches your mood, the perfect music for every moment.
In the case of the Home area, Rdio created a special tool to harvest data. This tool gave them the information they needed to fulfill the needs of their diverse audience. Home uses small units called Stories, that can be adapted to a variety of users based on a variety of criteria.
Learning From Feedback
The evolution of Rdio’s design has not been without critics. With every change or feature introduction (in some cases feature removal), there are always those that are quick to make their displeasure known. It was refreshing to see that instead of bristling at this, Mike and Geoff welcomed it, even appreciative of it as Mike expressed:
... It always feels great to have people care enough to say something.
Geoff shared that Rdio has a passionate user base, many of whom are long time users, that provide constructive criticism that they take into account for every decision. This willingness to accept even negative feedback is balanced with the realization that it is impossible to please everyone, and that critics often represent only a small portion of users.
Finding Harmony Between Music and Design
At Rdio, all efforts and resources exist to support one thing — the music. Geoff shared a quote from his predecessor Ryan Sims about the role design was to play at Rdio:
Rdio is the canvas and the music is the paint.
This serves as the design team’s mantra, a rallying cry as they move forward with clean, minimal design. When Rdio first hit the scene, it made a huge splash due to its sparse, minimal white design. This simplicity was purposeful, an intentional choice that reveals more about Rdio and their values than one may think. Geoff called the choice of white a “philosophical” decision. So what is Rdio’s philosophy? In simple terms, they believe in complete freedom, freedom for users to enjoy their experience the way they want to.
For Geoff, the dark, moody colors so popular in many popular media apps are subtly confining, stifling users. With Rdio’s latest redesign, album art plays a very prominent role. This is in complete harmony with Rdio’s goal of putting music first. It was impressive to learn just how much design was helping Rdio tell their brand’s story, a story that is continuing to evolve and grow along with the music they deliver.
Our chat with Mike and Geoff ended with questions from the audience. We want to thank them for coming down and sharing his insights with ZURB and the ZURB Soapbox audience!
Ryan: Let me sit here. You're listening to KROQ in the afternoon, and I will get a little settled here. But speaking of music I just want to start off right out of the gates with music since we got the afternoon radio show vibe going on this afternoon. What is some of the music that influences you and what are you currently listening to or what are you excited about in the industry?
Mike: Wow. It's hard I think working in a music company is difficult because we love music so much and we're constantly listening to it and discovering stuff and all kinds of things. The things that I've been most excited about are the reemergence of Indie bands being able to release stuff sometimes without even a label and have these digital distribution channels for people to discover their music and some of the smaller labels flourishing and finding new audiences because of these platforms. That being said there's a lot of cool stuff that has come out this year. The most notable stuff, there's this band has been Viet Cong from Canada that released an EPA that I love. I don't know Geoff, what have you been listening to?
Geoff: Wow. If you looked at my heavy rotation lately it would involve a lot of Elmo's Greatest Hits and Sesame Street. I have a two-year-old daughter and we have a lot of dance parties in the kitchen.
Ryan: Elmo Dance Party, right?
Ryan: It's very popular these days.
Geoff: Yes. Me Left Me Cookie at the Disco. That's a good one.
Ryan: That's a good one. That's a good clubbing song.
Geoff: Yes, which leads us, eventually made it to a feature that we need to work on which is play history editing. But it's funny, we're all really passionate about music at Rdio. You might expect to walk into the office and hear music blaring and it's like an MTV set or something, but it's actually pretty quiet and subdued, and thoughtful environment. I think there's a party going on in people's headphones. But I think we also consume a lot of music. So we have this voracious appetite for music discovery, and we work at the perfect place for that, but we can talk later about it, but it also sets up a challenge for us because we are atypical music listeners. We have a prodigious appetite for new music and pretty wide-ranging tastes. I was saying earlier, I kind of came of age in terms of my musical taste in the 90s and so, there's a resurgence of Indie bands that have this jangly guitar sound that hearkens back to 90s college rock a little bit that it's kind of fun to hear come around again. But I also listen to a lot of hip-hop and I tend to like stuff that's a little bit of off-kilter, a little darker, weird.
Ryan: Like Elmo's Dance Party?
Geoff: Yes, yes. I'm a complicated guy. It's like Elmo's Dance Party and then there is like. . .
Ryan: I preferred Grover's last album over Elmo. But speaking of music, because it's obviously, it's just a huge part of what you guys do on a daily basis, and that must present an unusual design challenge to say the least. So what are some of those challenges in designing an industry that really has drastically changed in the last decade or so with iTunes and streaming music and all of these other options out there?
Geoff: I'd say we have lots and lots on that. I think. . .
Ryan: Please share.
Geoff: Yes. Well first of all I mean it's just really fun to work on music. It's something that it's not about us doing our. . .we're not helping people do their job more efficiently or streamlining their workflow or anything. We're helping people get incredible enjoyment. It's like kind of a quality of life product I feel brings a lot of joy to people's life. So it's unique in that way. We could be working on a real-estate app or something and we're definitely not. But in terms of some of the unique challenges I think there are, it's a crowded field right now and you see all the people tackling the problems of building and streaming music service. You see them wrestling with I think a few spectrums that the listeners embody, or they fall somewhere on a few spectrums and, the spectrums that we think about a lot are there's a style of internet radio listening, the Pandora style listener on one end of that spectrum and then on the other end of that spectrum there's the more lean forward albums and playlists engaged listener. There is just in general kind of a casual music listener who might listen in their car a little bit in the background here or there while they're working.
Mike: That's been the hardest thing for me I think is going back to what we are talking about previously. We all love music. Everyone who works at Rdio is this massive music fan, and initially there is that challenge of just empathizing with these different types of users, trying to understand that love for music for someone who turns on the radio and is like, "Man, this Nicki Minaj is hot." That's a kind of love that is different but no lesser than the kind of love of someone who carefully curates a history of Protopunk playlist that meticulously accrue the chronology of this genre that wasn't a genre. So, those are different types of love and they're expressed really differently but one is not lesser than the other, and that's sometimes difficult because we love music so much in a very particular way and express it in a very particular way. So that spectrum can present its own challenges of just like trying to embody the user and understand their problems.
Ryan: How do you, speaking of that spectrum, how do you determine what exactly is that spectrum and how do you kind of match what you're working on to meet the needs of each of those individual spectrums?
Mike: That's a good question. Its features sometimes lend themselves to the spectrum. So for instance, sometimes a feature around improving stations predicts the kind of user that you're anticipating. And some other features are we just have to look at these genericized personas of these different types of users and appreciate that it's not that black and white. Right that most people live somewhere in that gray area in the middle of the spectrum. They're not album listeners. They're not station listeners. They're a hybrid of both. And in some ways although it's a fuzzier answer it makes the problem easier to solve because it gives you, I think, a realistic appreciation of the set of expectations that people have when they are coming to a music service, right? Like is it station time? Is this playlist time? Is this album time? Is this music discovery time? And those are pretty well-defined sections of the app to facilitate all those different types of listening.
Geoff: I mean we, the spectrums I was talking about from like casual to more serious, there's also individual listeners and then more social listeners who want a network, who want to share and have music shared with them. There's people who really want a human curator presenting a selection of music to them versus a machine algorithm generating a smart genre based playlist or something. So those are spectrums but we don't necessarily think that those are challenges. Those are the parameters of the landscape and of the users, but we don't think that you're either this or that. Or that even a single listener is fixed anywhere on. . .it's a continuum, right? Those things from individual listener to social listener, lean forward lean back.
Ryan: Depending on what day or mood, all sorts of.
Geoff: Exactly. We think that all of us kind of have within our musical horizons, whether it's each day or each week or just sometime during our lives, the way we relate to music that we will embody different places on that continuum. And that could be like I'm cooking dinner really fast and I just want to hit the button and have the machines do something magical and play me just what I want to hear. I don't want to put a lot of thought into it. Or, but that same person, me later might be meticulously preparing some kind of New Year's Eve party mix or something and I want tight control over all the details.
Ryan: Just the right thing.
Geoff: Yes, so all that to say we're working on features that speak to all of those different users at different times, but we believe that's the same user. We're not switching modes that discretely. We're trying to serve them all.
Ryan: It's not as if each user is siloed[SP] in a particular way. They're just going to move across that at any particular given time.
Ryan: And how does that kind of influence your design process? Take us through, what is it in the Rdio design process that is influenced by music? And what are kind of like the tools and the relationships and the things that you think about in terms of that and how does that help you push out a feature or a redesign per se?
Mike: It can be really difficult to account for all of these different behaviors and all these different listening types. And it feels like the way that it impacts our process the most is having to answer a lot of questions within a design and a lot of times that's helped by prototypes. So a prototype will help us move through a journey and see what's this like for the album listener? What's this like for the stations listener? Or what's this like for someone in between? And understand what those extremes are for a specific problem. But it's tricky. It's a lot of conversations and a lot of what ifs. I think listening the way that we're describing and all these different types of listening, you can see a lot of them as edge cases. So there's a lot of that, what if a user tries to do this or tries to do that? And just vetting those against not something static but something moving. So trying to get a prototype of the thing quickly helps us answer those questions quickly.
Geoff: Trying to serve that wide range of listener types, it's a lot to ask. It's hard to be everything to everyone. And to still have an experience that feels focused and elegant and enjoyable for our users, we have to personalize. So we have to find ways to personalize whatever features we're building to those users wherever they are on those spectrums. So, for example, we recently redesigned the app to include the section that we call Home. As the name would suggest, it's where you land. And it's a feed of activity that is recommendations, it's your own activity reflected back to you. It's activity in your network. It's meant to be the best of Rdio. It's meant to be kind of like a music magazine that evolves with you, that's personalized to you. But that means not just responding to what our users have explicitly told us.
There's a lot of signal, a lot of data going through our app and we need to personalize. We need to listen to those signals. And it's like getting to know somebody in real life. I'm asking direct questions in conversations, and certain facts will come up just in the natural flow of conversation, but also I'm listening to nonverbal cues and I'm taking context clues from things around us. Similarly, we're picking up on explicit signals. So when we first on board you onto the app, we're going to ask you a question or two about what you want to listen to. That's going to give us just a germ, just a seed to build on. But then we're also going to be paying a lot of attention to implicit cues from you about what you play repeatedly. If you're listening to a station, what you choose to skip. What you give a thumbs up versus thumbs down. Whether or not you follow people or communicate with them, and also you're style of listening. So do you identify yourself by your behavior as a stations listener or more of a playlist builder, etc.
And some of those get into more implicit and then eventually we have the aspiration to get even smarter and almost be able to read your mind by predicting like. . .thinking about recording what time of day is it when you're listening to this kind of music. What's the weather doing outside? What is your past behavior in this context suggest for what you might want right now? We're not doing a ton of that right now, but that's where we hope to get. And that would be more of sort of contextual signal that we'd be listening to. But anyway, so there's all these signals and some if it is more direct, some of it we have to read between the lines. But we try to compile all that signal and use that signal in the case of like Home that I was talking about to create this feed that is relevant for you not just in terms of presenting suggestions of let's say the style of hip-hop that you're interested in. Not just music style stuff but also music listening style stuff, so if you're a stations listener, we need to be showing people stations suggestions more than albums. And social if you have expressed zero interest in a social experience on Rdio, we don't want to keep bombarding you because it would feel really unpersonal, out of touch with you. So those are some of the ways that we try to approach.
Ryan: Right, and when you say how you approach it, that's a mountain or so of data that you have to sift through. How do you go in there and kind of triage and say this is what this is telling us, this is what we need to look at. Because that mean you could have tons of it and you should be overwhelmed by it.
Mike: No, it was a hard problem to crack. What we ended up doing in the case of Home was building a tool that let us give any username and say, okay, well, let's try to define a kind of pseudo algorithm that would define the layout of what their Home experience would be like and we tried to bucket people into some kind of behavioral clusters. Put them through this tool and the tool was not beautiful. The tool spit out lines of data that we then had to deconstruct and reconstruct, but it was massively helpful in terms of taking this really complicated problem and giving us a simple input and some reasonable output that we could then massage into get an idea of whether we're answering the questions, the things that we're trying to solve here? Has this person who listens to stations most of the time and albums some of the time getting the right kind of information? Are they only following two people, but they're very active people okay, versus this person who follows 300 people. It let us just see the variety of situations that we would find ourselves in. So we just tested tons of users through this. It was a crude prototyping tool. It was basically a glorified admin tool. But it helped us massively and it was just like trying to identify the answer we wanted and then finding a simple solution that helped us test all of these varied scenarios and it's not, it wasn't quick. It took a long time and a lot of effort, but we got the answers that we needed.
Ryan: Go ahead.
Geoff: And just thinking a little bit more about Home and how the design execution of that in order to serve so many different types of listeners, we had to come up with a system, like a templated system, where these. . . we call them stories, this kind of units in the feed. A kit of parts that was really small and versatile and repurposable so that the complexity wouldn't balloon out of control in trying to personalize too much visually to the different user types. A system that could adapt to all those to the social versus individual listeners, stations versus albums and stuff.
Ryan: Just to follow up a little bit on how you guys tested this. I think it's, how did you guy actually test with users? Do you like push out a live prototype because, for example like Google is notorious for this, right? They'll push out something and then my Google Docs page looks different every time I log into it so I don't know which version I'm going to get. But how do you guys actually test out those new things for the Home?
Mike: We've done a combination of things with Home where we had this big very different experience. It felt like it needed a set of users interacting with the real thing. So we looked and found friends and family users. We found, they ultimately about 100 users that we could invite to use a thing in it's crude form and apologized for it up front. We were like, "Hey, you know this thing is going to break. It's going to be ugly at times, but we would love your feedback on it. It's a new thing for us." So we kind of set the story for them, and then just we're not very prescriptive about the feedback that we wanted. Once people accepted, we said, "Great, use this. We'll turn it on for you. Here's an open text field. Ran at us." And we got a lot of really great feedback about the feature that way. For smaller scale features, we'll absolutely do the kinds of things that you were talking about where we will turn a feature on for 10% of users, watch the numbers very carefully, and see what happens. And then sometimes we'll take a feature and put it in front of a group of users to test. So it's kind of a variety depending usually on the size of the problem, the technical complexity, can we actually get this thing in the app easily and get some feedback about it.
Ryan: And in terms of getting feedback, because I'm finding this very fascinating is, obviously people are very passionate about their music to say the least, and they can be a little aggressive maybe with their feedback. How do you sift through that feedback and see what is important and what's not in what you are trying to accomplish there?
Mike: It's always stressful making changes. Usually when we make big changes, we get this very vocal feedback about things that we've taken away. Never take anything out of your product ever. It's really, really hard to take features away as I'm sure people have heard time and time again. God, that's true, especially with a really passionate user base. So it's really difficult. We recently went through an exercise, what about six,-nine months ago or something like that, of re-doing our navigation. And we knew it was going to be painful going into it. And we messed up in some places. And we had to just be honest, and there were a lot of very stressed meetings that were like, "Should we roll this thing back? Should we change this?" And we did make concessions. We realized that we'd gone too far in places and a lot of that was just like not being too stubborn about the ideas that we were trying to communicate about being realistic about what users were willing to enjoy and what they were not. I try to stay off. . . I don't search for Rdio on Twitter too often.
Ryan: Yes. Especially the day after a big launch.
Geoff: Yes, I did actually read a review about the redesign and having moved the listening history out of the sidebar, and then there was just this whole flurry about, "Where's my listening history?" Which you want to obliterate.
Mike: Yes, and the thing is it took a long time, not a long time, it felt like a long time because of how passionate the response is and that's one thing where I think we stuck to our guns a bit on that and keep listening. It's like the first part of Home. It's the first thing you see in the experience and it replaces a lot of that, but gosh, yeah.
Geoff: We're blessed with passionate users.
Mike: Yes, absolutely. No, I mean it always feels great to have people care enough to say something.
Ryan: Right. Sure. And how does that deluge of criticism, how do you guys deal with that? Because that can also be a little trying at times or a little stressful because you have such a vocal group. But obviously they care enough about the product to be vocal. So if they didn't care they wouldn't have been vocal. And how does it actually propel you in terms of how you approach the next iteration and the decisions that you have to make?
Mike: We've introduced a really wonderful process that I love. We have this great support team and they tag this incoming issues in a way that we can sit in a meeting every other week and say, "Okay, what are users asking for? What are they complaining about?" And it just has this direct impact on the features that we work on and the way that we evaluate changes, and the way that we address any technical problems that may be happening. So yes, there's a process that we've put in place to hopefully get at some of that.
Ryan: Great. Do you guys ever just select customer and users just to get on the phone with them and talk it out with them, or get out in the front lines just to see what they are actually saying?
Mike: We haven't done so much of that. Usually, we try to have a strong hypothesis around any testing that we're doing. So there's less of like the free-form interviews. Some of that has to do with the size of the company that we are. We're about 200 people and so we don't have a dedicated research department. We usually outsource all that stuff, so we try and be efficient about it. We try to come into a problem with a really clear, "I think that this method of onboarding is going to better than that method of onboarding," and then have an answer that we can get out of that test.
Geoff: But we do have a group of super users that have the ear of a few people in the company just because they've been so loyal for so long and they have kind of grown up with Rdio. And we listen. I mean we listen to all of our users, but we have some super involved, super vocal users. And it's not annoying. And it's not like the flaming tweets that come at you, it's more measured critique, and we pay a lot of attention to it. But we also have to take everything with a grain of salt because they represent a slice of our usership and are biased toward a certain style of listening and a certain set of feature concerns. So we listen but we also kind of modulate and edit and decide carefully what we're going to act on.
Ryan: Awesome. I kind of want to switch gears just a little bit away from it because I do really want to ask this question because music and Rdio has a very clean and vibrant design and album art seems to really be a fixture of that. It really seems to be centered around that. And I wanted to know why is that? And how does that figure into you guys' designs when it comes to things as discoverability and that personalization.
Geoff: Yes, so my predecessor, Ryan Sims, who led the product design team for years at Rdio and did some amazing work, had a saying that I really like which is, "Rdio is the canvas and the music is the paint." And it was a nice mantra. It was his rallying cry when he stripped a lot of what now seems superfluous texture, color, dimensionality out of the app a couple of years ago and went with a really white, clean, sparse, minimal chrome for the app. And it shocked people at the time. It shocked me. I was already a user long before I worked there and it was jarring shift for me. Time definitely proved him right, and that metaphor canvas and paint resonates for me for a lot of reasons as a designer. The stripping away, like do as much as possible with as little design as possible is obviously a truth to aspire to.
I think it also reflects something about our approach that I was talking about earlier. There's a certain democracy to that minimalism. We're trying not to prescribe how you should use the product or cater more to a certain style of listener than another. Maybe it's a stretch. It gets a little bit into art critique territory but if you color your app purple or you color your app predominantly charcoal gray with pops of red or whatever, what are you saying subtly to your users about how your app should be used? I think that you are saying some things. Maybe you're saying something with white too, but I think it's the least prescriptive visual route to take, and that's intentional. It's not just an aesthetic decision. It's a philosophical one.
Ryan: You're communicating an overall story for Rdio in that way then.
Geoff: Yeah, or a lack thereof, an openness for you to have your own story with Rdio instead of this is your dark, moody media environment console thingy. That feels kind of dated to me. And I think for a long time our media apps have black and dark and dim the lights and let it wash over you kind of thing. I don't know. It has its place maybe like on your TV, but it's a metaphor that feels a little tired. Make it black because it's music or video. And so I appreciate some of those hard decisions that have been made to strip a lot of stuff away. Yes, and it almost goes without saying the music is the star of the show, so maybe that should be the most prominent thing you're looking at. And I think that has informed a lot of the decisions around big beautiful album blur in the player, and picking a predominant color out of a piece of artwork and using that to wash over a page or wash over that album artwork to create a backdrop over which you could retype or UI. But always kind of respecting the artwork as our most precious ingredient on the pallet. I think that it's consistent with the product vision, respect for the music.
Ryan: It's almost as if you're letting the player actually get out of the way of the listener in that way. And I have one last question before I throw it out to our wonderful audience out here. Finally, for each of you, Mike and Geoff. How is Rdio changing the world of music?
Mike: It's a difficult question.
Ryan: That's why I asked it.
Mike: It's hard because, it's specifically hard I think to answer. We are a niche product for a niche type of user. So we love to be an example of this shift to streaming music and as much as it's happened for a lot of us in this room, it hasn't happened for a lot of people. So for me it is kind of a non-answer but that's still a shift. That's still a change that we want to encourage and inhibit. We love it so much and see so much value in letting go of mp3s, and tagging on my ID3 tags and all the rest, and keeping everything beautifully manicured like letting that go and having this library of millions and millions and millions of songs to discover as the next evolution of listening. And maybe that's evident to some of us but it's not a really popular concept in general yet. So we're still making that shift.
Geoff: Yes, I think we forget probably most of the people in this room use a music streaming service or are familiar with the concept but that's not a global understanding or use pattern. We are a tiny sliver of the pie. Not Rdio, but music streaming a tiny sliver of the music consumption pie. So just by being in the market we're changing people's relationship with music. And then, I don't know. As for what Rdio uniquely does, I feel like maybe I'm repeating myself a little bit but I think it is unique that we believe we can be the best album on demand, album playlist on demand listener service, and the best internet radio service. We can replace Pandora, and we can replace your ripped CDs, or whatever other streaming music service you use. That we can create one home for all listening types. I think not everybody's trying to do that. A lot of people are trying to tackle one part of that puzzle. Yes.
Ryan: Cool. Well, thank you very much, Mike and Geoff, for answering my questions. And I want to throw it out to you guys and have your questions answered. Yes.
Man 1: [inaudible 00:32:03] I didn't hear you mentioning about [inaudible 00:32:09] all of us are US-centric [inaudible 00:32:13] we are US-centric, and we are [inaudible 00:32:15] a given demographic. How much does internationalization have you talked about? I've just noticed, you're very, very difficult because [inaudible 00:32:29] you were talking colors and palettes. What color means to one is different for the other. Light colors, dark colors, please comment to that.
Mike: Yes, absolutely. So internationalization is hard/impossible. And the way that we've approached it is by having small local teams. So we'll work really closely with our teams in Brazil, or our teams in India to assess even things down to like, "What should the station artwork actually be?" Because it has such a massive impact on people's behavior and consumption. So, yeah, we try to lean on our employees in a region. And that being said like that makes it sound like we're doing a great job at it. And I don't think we are because we can be out in every territory. We have to pick and choose.
Male 1: And that depends on who you choose.
Male 1: You're bound by their preferences rather than the national preferences.
Mike: Totally, so we do so much of them. . .
Male 1: Actually, a country like China, you'll have a serious problem because they want it to limited. So how do you handle that?
Mike: Well it's difficult. The closest we've gone is leaning on local experts, and it only get us part of the way there. And, yeah, I mean it's an evolving process for us as we expand into more and more territories.
Male 1: That is the reason [inaudible 00:33:51 has become rich.
Geoff: Yes, that said internationalization is a huge priority for us right now. I think we're in 85 countries at last check. So it's huge. But, yeah, to scale to serve them properly, we've got our work cut out.
Male 1: Okay.
Ryan: Paulina in the back.
Paulina: I'm wondering about [inaudible 00:34:13] together and work with your team?
Mike: The question is how do we work together, and how we work with our teams? We work really closely together. Geoff and I have been at the company over two years each, and there's a lot of built in collaboration. There's a lot of ad hock collaboration. And so we work really fluidly as a team and part of that is having the right tools to allow us visibility into what everyone's doing. So we've been using Slack like I'm sure a lot of people are using Slack and we've built some hooks into the app that post whenever design changes are made that facilitate conversations around any changes. Geoff can chime in, and I can chime in and we see each other's comments or we can have a conversation and then replicate that feedback in these tools and other than that it's we have a great relationship with our engineering teams as well, and we're fairly hands on, so we're doing less of just handing something over and be, "All right, see you in a few weeks when it's built," and a lot more real-time feedback. And some of that's fed by prototypes and some of it is just like looking at the alpha and beta builds of things as they come out and feeding back in real time.
Geoff: We've had increasing success with smaller teams to where engineering and design and product management is all represented within one team that may range from five people to 15 or so, but it feels like everybody has better visibility on the work. There's more just buy in. There's more collaborative spirit and better communication. In that sense, there's not an entrenched feeling of just teams kind of that can almost create friction and slow the work down. There's a feeling of we are this team working on this feature. We all have different skills sets and we'll bring them bear at different times, but the communication is much more fluid and the process is obviously less waterfall and things can move more fluidly back and forth between engineering. Sometimes they need to pull back and go back through a cycle of design before they get handed back again to engineer.
Male 2: Following up on the comment about streaming, prevalence streaming versus [inaudible 00:36:46] It seems to me that streaming is just a radio station and radio stations have been around forever. So it seems like that is still a prevalent. It's like so you really, I mean, how do you see that plainly? It's a market. You should be able. . . that's basically waiting out there. You can say it's over [inaudible 00:37:03] from a radio station perspective but, I mean you're trying to, or just comment on that. How do you see that fitting? Because it's actually kind of an old market.
Mike: No, absolutely. A lot of it is around habits, but there's some uniqueness in that you can do in a streaming service that certainly you can't on radio.
Male 2: Yes. I mean you get to pick your playlist which is the cool part.
Mike: That and be able to listen to specific albums, and we're trying to bridge that gap a little bit and see how do you move someone from being a radio listener and introduce elements of on-demand listening into their life. But it's a challenge. It's one of those things. We want to replace that radio experience.
Geoff: Yeah, the concept of terrestrial radio is super old, and there has always been a relationship between hearing music on the radio being pumped to you and using that as a discovery channel and then going to the record store or wherever and buying that record. So there's always been, not always, but for decades and decades there's been this peaceful coexistence between on-demand music listening and terrestrial radio. And that is something we recognize and that is one of the reasons why we're not trying to just be radio service, or just be an on-demand service. There's a relationship there that has existed for a long time and we can serve all online. Is that kind of what you were. . .
Male 2: Yes. Just sort of wondering how your tapping [SP] cuts. Yes, sort of answered. How you are tapping into that because in one sense you could say the way has been paved for you but because it's so common you're going to be overlooked because they already have it. You really have to target something so prevalent [inaudible 00:39:09] unique way I guess is perhaps maybe. I'm hearing a little bit of that, but that's what you have to do even though it seems like it could be so similar.
Mike: But it's difficult. We struggle with a clear answer because I think the clear answer is a really crude one for us which is partnerships with a lot of these radio providers. So we work with Cumulus a lot and they own close to 600 stations in the US. So there's a lot of these pretty clear hand-offs from these radio stations into our streaming service.
Geoff: And also trying to get Rdio into the places where people traditionally listen to the radio too is another thing that we work on a lot. Meaning cars, living rooms, places other than just your phone. So we're busy on a lot of car and home partnerships as well.
Ryan: Cool, we have a question here and then we have a question over there.
Male 3: I'm just curious how your team works when you design across devices of like one person taking it across all devices or split up [inaudible 00:40:21].
Geoff: The designers we have now are able to work across platform and we kind of encourage that. I think it hasn't always been that way and people are always bringing certain expertise to the table. We're not always equally good at everything, at designing for every platform but that's definitely been the reality lately and the desire is that people design cross platform.
Mike: On the UX side, it can be a little bit easier but also not quite as diligent because it's easier to say, look, what here's a wireframe or here's a prototype that works across platform but then we're ignoring a little bit of the peculiarities and the flavor of individual platforms that you wouldn't be able to in a full design.
Ryan: Cool. And then we had a question over here.
Male 4: So how do you balance designs for like users' wants versus design for business needs and the actual need to monetize and have profit?
Mike: Yeah, designing around ads was a new challenge for me, and there's that side of things and just having a free experience that is subsidized with ads not trying to make that. . .like the ad doesn't have to be a horrible experience, and we tried to make that fairly seamless. It's hard. A lot of these problems, you try to make as good as possible of an experience and still have these concessions. I think we are fortunate that people accept the burden of ads for the benefit of having a free product. And there's a obvious upgrade and subscription path. Yes, it's really difficult. We're trying to introduce partnerships that aren't as straightforward as just like here's an Rdio ad. What if it's like Home Depot creating a playlist? And maybe that sounds a little bit awkward on the surface but the content is really good. I mean this is an actual example of a station that we did and it got a lot of plays and a lot of engagement. So there's just trying to turn these advertising experiences into something that's meaningful within our products which is a little bit more beneficial to the user than just hearing a few minutes of music and then hearing an ad and then hearing a few more minutes of music.
Geoff: But yeah, there's the tension. There's no easy answer to that question. We're committed to making an experience that is free and still awesome and some of that involves ads and sponsorship and we're trying to do it right. But we're wrestling with it constantly and we have people who work at Rdio now for whom that's their sole job and we didn't always have that so, yeah, it's constantly on our mind and it's a real tough challenge.
Ryan: Pretty good. One last question.
Male 5: So when you guys are designing features for your products, do you find that design for the extreme users benefits the casual user? Or is it that you design for the extreme user and then you have to temper it to something that everybody can like?
Geoff: When you say extreme, do you mean the music enthusiast or do you mean like an educate.
Male 5: I mean, it depends on how you define extreme. How do you guys define extreme? You guys came through and talked about like the album listener and the radio listener or somebody that manicures perfectly, the perfect playlist.
Mike: Yeah. It's hard because I think often times we are the extreme listeners. So the first reaction a lot of times on answering our problems is to answer it for yourself. And that answer is often not satisfactory for these vast number of users in the middle of the spectrum. And I think one of the ways that we get out of that is just vetting it against various journeys and usage partners. We're really lucky, I think, to have identified people even within our team that have really different style of listening. So one of our product managers, she loves listening to songs. It's like pretty exclusively. She's basically a radio listener in her behavior, and I love bouncing ideas off of her because she has such a different view into the product than I do. And to extrapolate that out many times and talking to many people we are benefited by the fact that everyone loves music and interacts with music in our product differently. You get such a variety of answers deepening on who you talk to. By the time you're done telling the story in a simple enough way that people understand it, you've hopefully addressed not only those extremes but the people in the middle as well.
Geoff: But one person might like to discover music by hearing music by hearing a few seconds of it, and another person might like to say, "Where is this band from? What label are they on? What are their affiliated bands? Oh, this the side project of that guy?" Like super music nerdy stuff. Those are really different ways of discovering music and designing for one extreme, let's say the latter is the extreme and the first is the casual listener, designing for the extreme isn't going to solve for the casual listener in that case. So it's not a matter of solve for the extreme and you've automatically solved for the casual listener. They are pretty different. Depending on, I guess, what feature we're talking about, they can be really different needs sets.
Ryan: Very good. Well, thank you, Mike and Geoff for joining us here for ZURB Soapbox. Thank everyone for coming out. Give them a round of applause.
Mike and Geoff: Thank you.