Andy Carvell, Growth Manager: Mobile, Soundcloud on A Strategic Framework For App Marketing

Andy Carvell works at Soundcloud and gave a talk at App Promotion Summit Berlin on the topic of A Strategic Framework For App Marketing. The talk covered the key components of an app growth engine, and also looked at approaching growth holistically.

You can find the video of the talk below as well the full transcript.


Hi, I’m Andy Carvell. I work at SoundCloud, working on growth topics. I’mcurrently leading a retention team, a team of five people focused purely on user retention. I was here last year, I was actually super sick last year, but I came and I presented some stuff about growth and framework that I’ve been working on and it was really well-received, despite the fact that I was really sick and coughing my way through the talk. So I’m feeling a bit better this year, so hopefully I’ll make myself a little bit better understood. So I’m here to talk about growth, I’ll talk a little bit about the framework that I’ve developed and first present an update to it this year and also give some examples of how we’re approaching growth at SoundCloud and applying some of the principles in that framework. So, this slide here is basically to show some examples of things that we do at SoundCloud and some of the technologies which are relevant in the space, but also to show that growth is a pretty complex topic. Particularly on mobile, there’s a lot of different technologies out there, a lot of different channels and it’s a very fast moving space. So for that reason, last year I presented this, what I call the Mobile Growth Stack, which is my attempt to kind of organize thinking around growth and to help growth marketers to think about growth in a holistic and strategic manner. So very briefly, the stack is split into three main layers, acquisition, engagement and retention, and monetization, underpinned by solid layer of analytics. I really do believe that analytics and measurement is really key to most of the growth efforts and things should be experimental and done with measurement and testing and scaling what works. On the right hand side, there are two vertical columns, which are activities which I believe are themes that cut right through the stack, cut through the different layers, so retargeting can be used to bring back users who like, maybe, didn’t download the app…Well, sorry they must have downloaded the app but
it’s bringing people back in who maybe right on the top of the funnel, not very engaged, bringing back to more engagement, or maybe people who’ve abandoned the cart, for example, and bringing them back to do a conversion step. So it’s really applicable to all areas of the funnel. And international, yeah, it’s a huge topic, but it’s really about a lot more than just translating your app, it’s really about how to grow in new markets. I sort of worked on a few of this areas over the last four years of SoundCloud. So that was last year, and I published a blog post to go with this, which was up on Medium, which has been very well-received throughout he industry. This is an example, the guy from Gram Games, their CMO, came up to me at a conference recently, earlier this year, and showed me this. So, on the left hand side, that’s my Mobile Growth Stack, which they are actually using to drive their development process at Gram Games, which is a games studio in Turkey. Turns out they know what they’re doing with growth as well, they’ve had a very successful title called “1010” which has been up near the top of the games charts for about a year now. Also, Greylock Partners organized a workshop for some of their portfolio companies over in the US, using the Growth Stack as the foundation for that workshop, where they looked deeply at what technical tools and what technologies their portfolio companies were using to drive activities in different areas of the stack. So, basically, it was very well-received in the industry, which is a great feedback for me and very humbling. It was really nice to see. So, this year, over the last year I’ve been thinking a lot about how to evolve the stack and basically, how to improve it from what I presented last year. So this is the 2015 edition. So I apologize in advance, it’s even more complex, but I’ll take you through that, basically, over the next 10 minutes and call out some of the changes and some of the thinking behind what I’ve done to evolve it. And yeah, of course, it’s an evolving framework and I’d really appreciate the input and thoughts from everyone in the room.
Yeah, we can catch up afterwards because growth is a really a team sport and everyone’s trying to figure this stuff out. It’s a complex topic and it’s a fairly new topic, this intersection of product and marketing. Nobody’s got everything figured out. This is my attempt to bring some order to the situation, but I’d really appreciate the input of anyone who is working or thinking about this topic as to how we can work together to to bring a bit more discipline to the to the practice of growth. Yes, so very briefly, I’ve broken out channels and really tried to make a separation between channels and activities. Because what I had in the initial stack, I was sort of conflating sometimes things like welcome emails with first-time user experience. So kind of bringing examples of specific use cases into the stack itself. This time I’ve really tried to keep the core stack about activities rather than channels and breakout channels as a separate layer, because actually, you can often apply multiple different channels to a specific activity. I’ve also built up the analytics layer a bit. But I’ll call out some specific examples of how things have changed now. So, as I mentioned, channels are not the same as tactics, so I built out this layer. This is, of course, not an exhaustive list of channels but I think these are the channels which are probably most relevant to marketers working in mobiletoday. So push notifications, most people know about in-app messaging. I think it’s very underutilized as a channel and in my experience over the last year, in-app messaging is the killer channel for mobile. Of course it can only work if the user is actually engaged in your app, so it’s no good for acquiring uses, and it’s no good for reactivating uses. But if the users in your app and you want to drive them to more conversion or to a specific type of engagement, then in-app messaging is really awesome. The difference between push and in-app messaging, push is something that you send to a user and they receive it when they’re outside of the app and in-app messaging is a dialogue that can pop up, you can generally configure exactly how that looks. You can make it look very native. But the really key difference is that the marketer is in control of when that dialog is displayed, who you’re
So in the acquisition layer, I’ve added content indexing as a brand new box. So this is indexing content from within your app that it’s discoverable in mobile web search that people are searching on mobile, content is now indexable. So Google have…Google App Links I think it’s called, sorry it’s Facebook App Links. Google App Indexing and also Apple with iOS 9 released the extended iOS 9 search. So basically there’s a lot more ways for people to discover your content from outside of your app and get deep linked right into it. So it’sbasically the new SEO. PR, I broke out as a separate box, it used to be part of, I think I had PR and social as one box. I think PR is an activity in its own right and a different discipline to content marketing, for example. And distribution deals, I think it used to be called “Partnerships.” I’ve actually created a third column called “partnerships” because I think partnerships is something that can apply right across the stack. But I’ve called it out specifically in acquisition, because they can be a really big part of an acquisition strategy to do preinstalled deals, or other forms of bulk acquisition through partnering or traffic exchange A couple of examples of things we’re doing at SoundCloud on the acquisition front. Facebook App Invites is something that we’ve been experimenting with. It’s a channel that is not working great for us right now, but it’s bringing in some additional viral traffic, so it’s something that we’ll continue to iterate on. Google App Indexing, I already mentioned. So this is the idea, this is just an example. So you search for “Scrillex” in Google, even on an iOS device, this is not Android specific, then it figures out, does Scrillex have any profiles on other sites or whatever? So SoundCloud is listed there. If you click on that link, and if you already have the app installed, it will take you straight to Scrillex’s profile inside the SoundCloud app. Talk about engagement and retention. Sorry a bit of a crackle there. So retention engagement is something that I’ve been focusing on for the last six months or so, specifically around retention, thinking a lot about churn
production and how to keep users coming back for more sessions, increasing general activity. Engagement and retention are inextricably linked, in my opinion. If you don’t have an engaging app, you can’t retain users over the long time. So I’ve broken this section out a little bit as well. I’ve added a new cell for this idea of like data persistence. User accounts, this idea of giving a user some more permanence within your app and your experience, that as a sort of a lever for bringing them back more often, and maybe building up a barrier to the install, because the user’s more invested in your experience if they create an account or create a profile. So I would say it’s worth considering having some kind of account creation or profile dynamic in your app, even if strictly speaking you don’t need it, because if it’s a purely transactional app, where the user really gets the same unpersonalized experience every time that they log in, then they’re less incentivized to reinstall the app if they switch devices or to come back to the app time and time again. First time user experience is massively important, I’ll talk about that a little more. And I’ve broken out a mobile CRM from last year into lifecycle marketing and activity notifications, which I consider to be quite distinct topics. Just on the retention curve issues, the really top apps are able to flatten out their retention curve higher up. That basically means that more users are retained over the long-term, because they’re basically more effective at onboarding. Yeah, hang on one sec. Okay, sorry about that. Looks like it was my fault, the way I was touching this mic. Yeah, so basically, he best apps are just better at onboarding users in the first few days. As you see, the curve is basically the same. This is how we approach retention at SoundCloud.

It’s really about building a growth machine, getting users into a loop of kind of regular activity and helping to build habits around the product. And we do this in an experimental way. We have control groups for everything that we do, so we understand the effect of what we’re doing versus everything else that’s happening on the platform. We do lifecycle campaigns, so we use in-app messaging. This is an example of in-app messaging actually. Excuse me. Sorry about this. Okay, I think it’s stopped. Yeah so, these are examples of in-app campaigns that we use to encourage users on the onboarding. Sorry, we have some problems here. Yeah, sure. So, I’m just going to speak into the main mic. Activity notifications, I mentioned before. So these are sort of more real-time notifications. We’ve recently rolled out an activity notification system at SoundCloud so we’re delivering a lot more push notifications than we used to on iOS, for example. I’m going to speed up a little now, because I realize I’m almost out of time. So I mentioned already deep linking. Deep linking is super important and it’s going to just become more and more important, at least for apps that have a strong content element to them. And finally, monetization. I have thought about monetization. It’s a big topic for SoundCloud this year, we’re rolling out a new monetization methods as well. So it really just comes down, for me, this idea that most apps are either merchandising in terms of the way that they are selling in-app purchases, possibly with virtual goods, or they’re doing some kind of advertising, like serving ads within their product and so it’s all about inventory management. So specifically around SoundCloud, we have native ad format, such as the promoted track, so artists or labels can promote a specific piece of content within the app. This is completely native to the SoundCloud format, so it feels very natural in the app. And we also have more traditional advertisements we work directly with brands, such as SquareSpace, to have these sort of like takeover, interstitial, visual ads. And we also have audio ads which again, is quite a new format for us.

So that’s how SoundCloud is monetizing. And you probably haven’t seen this stuff because it’s live in the U.S. mostly. Promoted tracks, I think is everywhere but the visual display stuff is currently only in the U.S. market. Thanks a lot, I’m bang on time, I think. So, I’ve covered all of this in a lot more detail in a blog post, which I published last night. There’s a link to it, well a redirect, actually. If you just go to “” you’ll see that blog post in detail. I apologize, it’s even longer than this talk, it’s about 15 minute read, but there’s lots of stuff in there and some good examples hopefully. So thanks a lot for listening. You’ve been very patient. Thanks.