 exciting. Wow, what an entrance. Hello everybody. The topic for the next 30 minutes is community and growth. And I have Holly with me from Kabam, co-founder of Kabam, Gustav, partner at Y Combinator and previously growth at Airbnb, and Donna, now running growth for Greta IO and previously also at TicTail. I thought we would start off, because the topic is really big, growth and community. So maybe start off with what is the difference between growth and community? Or is there a difference? What would you say, Gustav? So I think that the way I view growth, so working at growth at Airbnb for a long time, it's an acceleration that when company gets a certain stage, a lot of things you can do to make the product and the users grow much faster. In the early days of a product, that happens very manually. At Y Combinator, the first advice we give to every company is to go and talk to users, basically spend time with them in person or talk to them on Skype or anything you can do to just get this very intimate communication with them. And it's more than just like surveys. Specifically, the advice we gave is basically when you start off, you can start by just emailing every single new user you have from your own account. So they get a feeling that they can talk to you. How you build a community in your early days, I think community can mean many different things. In the very early days, it just basically means for us or for a new founder, is to do things that don't scale and spend time with users that you can do because you only have a few thousand of them. At some point, you get to a stage when those things won't scale. Literally, you have to scale, but those things don't scale anymore. And that's where a team like a growth team comes into picture. At Airbnb, I was there from, I think we had maybe half a million and a million users, and when I left, it was over a hundred million users. We didn't really measure users, but just give you an idea of when things are at scale, and you have to do things very differently. So at scale, basically, engineering and data science are two very, very important levers on how you grow things. So you basically can't, it's very, very hard to grow to a very massive scale by going out and meeting and talking to every one of the users. You have to find another way to reach them, and that's through technology, marketing, other growth tactics. Great. Difference between community and growth, or are they the same thing? No, I would totally agree with Gustav, but it's also a very different thing when it's like, you're starting off and it's your thousand first users and how you work them, and it's like, as Gustav says, really important when you're in the beginning of doing things that don't scale and trying to find out later on, like, okay, was it here that we could actually take on and now scale it in a different way, and that growth has more to do with product, and then I would say that community is more about communications and how you do traditional marketing. I think for community and growth, I think it's really dependent on the types of businesses that you have, so I only say that because ours is a gaming company. You can kind of think of it as we have players and we have customers, well, everybody's a player, but if you pay money or a customer, and not everybody joins in the community, but for us, particularly in our games, it's almost like a feature or a draw and sometimes it's the actual thing that gets people longer retention and longer engagement, but you absolutely can be a user, a player, or a customer and not be involved in the community at all. So, in some ways, community and growth do impact one another, that's for sure, but they can also be if you're just looking at different things segmented a little bit differently. Exactly, yeah. So, you've all mentioned a little bit like metrics here and especially what it is that you are measuring on the different side of both community and growth. I think one of our goals today is to get as many hands-on, concrete advice as to the entrepreneurs sitting in the audience, so maybe referring to the industries you've been working in, what were the things that you measured on those sites? Yeah, so some of the things that we measure in terms of, so in our games, the way it kind of works is you could join what we call an alliance, and as an alliance, you could go do quests together, you could achieve objectives, so some of the things we would definitely look at because if you were in an alliance, that impacted your retention, and we started caring a lot about these people called regulars, and regulars were people who would log into the game for seven days a week, and we said, if you were a regular, then that meant your contribution to monetization was so much higher, it impacted our LTV so much bigger. So what we started looking at is what makes somebody come back every day over several years, and we've had that in our games for some people, and we found out like it was some people who got involved in the community in terms of the alliances because they wanted to do something together, or at some point, you know, in some ways you run out of spending things on yourself, it's like going to a bar, and there's only so much you can drink as one person, but if you go with a friend, you know, you could go buy something for somebody else as well, so definitely I think for us, measuring how many people were in an alliance, how long they were in an alliance, even one of the more important things, and we'll probably get into a layer is how do you scale the community, so you're not, as what we talked about, like the first 100,000, 5,000 are incredibly important, very hand led, but at some point you kind of have to scale it. Some very tactical advice I would give a new startup is to try to figure out what's the product value that your product is giving users. It's really important to figure that out. Like a representation of a sign up is not really that, it's usually something very specific. In the case of Airbnb, it would be a booking, or travel, I'm literally traveling on Airbnb, and once you figure out what that value is, you want to figure out how often people do it, so we can start measuring retention on how many people are doing this on a repetitive basis, if no one is doing it again, then it probably isn't like a good value, because people aren't doing it again, so if you figure out the product value first, you measure that, and you figure out basically how often are people doing this, and you have some ideal time, the time window that people are supposed to do this, those are the most important things to measure in the early days. If those metrics are good, basically you have a lot of people coming back and doing the things that you want them to do on a regular basis, then you should invest in top of the following growth, get more people to come in and try that for the first time, so Airbnb was in my opinion, and I think most of our users are like a pretty awesome service, so people came in and they booked, and they had a great experience, and our focus on the growth team for a very long portion of the time was basically first time bookers, so people coming in and booking for the first time, and that's, there's many, many metrics in between there, but that was the metric that we focused on. For most startups, I think it's important to figure out what's the one metric that is the representation of the value and representation of new value being created on the platform, and just drive that metric. There's many, many things you can do to do that, and metrics might, could be kind of a technical term to some people, but it's just a representation of users are doing on the site, or on the product, it's just a representation, it is more than just numbers. Yeah, I totally agree, and one thing that I've never done before, but I find really interesting working with DevTools, is the fact that you like metrics with, that is accomplished to like product value, and one thing that we find at Greta that is really valuable is that we actually educate our users in performance, so we can actually see that when users use us and get figures in performance in Asia, for example, which is interesting to some, then they learn more and they feel like, oh, now I'm going to go next, and then you can like reveal different parts of our dashboard, and when they feel like, oh, this value is really important to us, we can actually customize it, and I think that one thing that's really interesting is how driving product value through driving insight for the, like educating the user, that has really been something that I've been looking at, like for the last year at Greta, that I have not been looking at when I've been working in e-commerce, but it's really important when it comes to like developer tools, because then they get more engaged, when they feel like, okay, I'm actually learning something, and then I'm going to put this higher up in my priorities at my work. Yeah, super interesting, when I worked at Slush, we actually used the percentage of international attendees, and I think that's actually maybe a very value-driven metric to follow as well for growth, so we got a question from the audience here, that was actually something that we were thinking about covering regarding online and offline communities and that growth. I think we have an interesting combination here to cover that topic as well. The question is, is the only way to start an online community by starting with offline events, are there any other approaches? In the gaming, we mentioned they would be online first, right? Yeah, so in the gaming world, what's really unique about us is our whole community is built online, because we started with original IP, even if I were to have an event, nobody would really care, and so it was really hard to kind of, like, it just didn't make as much sense. It made much more sense for us to go into online communities, people of some type of interest, and go tell them, so in some ways like an equivalent unscalable kind of way to grow the online community. The one thing that I realized that it's, in general, I personally think it's very hard to convert offline into online, so for us, we also, and this probably is, we'll talk about more later, but we grew a lot through paid acquisition, so what that was is we paid for traffic to come online, and that really was how we ended up growing, taking, not growing, but really scaling our communities, well not our communities, but our games and our users, and when you do do paid acquisition, you have to be, and I'm pretty sure Gustav has a lot of stories around this, you have to be very careful about knowing kind of what we call your unit metrics, or basically your LTV, which is your lifetime value of your customer, because otherwise you can waste a lot of money trying to figure it out, so you have to have already kind of knowing a little bit of the revenue of your customer, so that I think is incredibly important when you're thinking about paid acquisition, I also think very much, it really is a business model thing in the sense of for games, I, how I make money is the players are going to pay me, so I need to provide enough value for that player, I don't care about ads, I don't care about anything else, but if I was Facebook, I wouldn't care as much about, I probably wouldn't do as much paid acquisition, I'd be like free, free, free, please tell other people, here's you know how you're going to grow in terms of some virality, send some invites to some friends, or you can beta some things, so that's really kind of how some, how we did some of the online communities, and how we grew for, for us just purely online, and what's been really interesting for us is many of our online communities have turned into offline events, so I think I was telling you guys earlier, we've had people who like have met and married through our games, I actually met somebody who like, somebody from Kansas, and somebody from Holland who met and married in our Hobbit game, so and sometimes they'll do like, bowling events together with their alliance, or camping and such, so ours has gone the other way, long-winded answer to that. No, no, no, great. Online and offline, so I think, I don't think it really, I think offline is a good way to meet your customers in the early days, but I don't think it's a way to grow really big, and I think with offline events, like community events, that's one specific thing, but you can have offline products, but if you think of something like Uber or something like Lyft or Airbnb, they are growing either through word of mouth, but they're also growing massively online, like basically the way a lot of these companies are acquiring, they're both the riders and the drivers or the host and the guest is online in many ways, so travel specifically is very, very tied to Google, most of us don't travel very often, so when we think of travel we go to Google and we type something in, and that's why some of the largest travel companies in the world, besides Airbnb, like booking in a comm and TripAdvisor, like getting all the traffic from Google, so I think offline events is not a way to grow, but I think it's a great way to meet your first customers and that's how I would start, like I would really, if you want to meet the customers, no matter what you do, you should find a way to meet them on a regular basis in person. We spoke about the kind of word of mouth example earlier as well, and growing a community and growth overall through that, with the travel bag example, right? Yes, yeah. Do you want to tell more about that? Yeah, sure. I have a really great example, so I used to work in New York last year, so I bought a bag called Away, they're a startup, I think they're like two years old, and my bag broke a little, little bit, and I emailed them and I said like, I can't find my receipt, but my inner lining broke, and I covered by the lifetime warranty, I'm based in Sweden, but I bought it in New York, and I was like, I'm not going to get this back, and then they emailed me like three, five hours later and said, what's your address? Here is your, we will send you a new bag from New York immediately, this should not have happened, and then I got the bag from New York seven days later, and during that time I could actually follow the bag with the UPS and they had done it so nicely, and at that time I went from like being angry, because this bag had been with me for a year and it already broke, and to being like, this is the best bag I've ever had, and they also, at that time, which is when it comes to referral, people like, oh 10% discount here and there, but they did the timing so well, I have not, not gotten a discount code before, but now when I was like very pleased, they were like, share this with your friends, spread the words, here's $20 off, and I think within like eight days I've told this to more than a hundred people, this is like, this is a really good growth method, I think, like focusing on your customers. I think what's so cool about that method is like, usually afterwards, I just get a survey, like how did you like this, or how did you feel, and sometimes I can't even remember how I felt or whatnot, it's very in the moment, but this one you could clearly tell how they felt by their action, that's one thing I noticed like with consumer, people will say anything, but it's more important what they do, and I thought that that was such a clever thing that they did, because they can measure like exactly how far this code wins, which is pretty cool, you might think they're top refer. Yeah, and I mean, I wasn't like talking about community and growth, and that I wasn't engaged before, but now I'm like, this is the best bag ever, everyone should buy it, and I'm also considering buying more, I was not thinking that before. Don't I sharing the code to everybody? Yeah, what's the code? You'll get it all sent out. How did you do growth through word of mouth, and maybe also customer service at Airbnb? Yeah, so one tactical advice to startups is that if you're trying to figure out what to do for growth, is you can just start asking your users how they first heard about the product. Now there's two ways to do that, one is to look at the Google Analytics traffic sources, I'll give you some insights, they won't give you all of it, so you can literally have a survey at sign up or at landing that will ask you the most common reasons why you heard about the product. That just gives you a start, it'll start giving you ideas. If you have a great product, it's very likely, like in Airbnb's case, that a lot of users will hear about you from word of mouth, so they'll hear about you from friends. Word of mouth is a strong driver, it's hard to accelerate it because it often happened in a very natural way, so you could try to figure out and be like, how does it happen? So we did that Airbnb, people generally talk about Airbnb when they talk about travel, you talk about travel when you're about to travel somewhere, you're telling you where I'm going, when you came back, I show you all the photos and like I tell you where I went, and then when one of my friends are traveling, I tell you, oh you should try Airbnb, that's awesome. Those are some of the times that people do that. For other services you might be a lot more often, so let's say you try Uber and Lyft, well that's going to be every week, so they have a lot more chances to tell people about the product, so the cycle time of using something is directly impacting the word of mouth, but yeah, I would just start by just asking people, that will give you an idea of how like it is to happen. Word of mouth can happen like I'll tell Holly about Airbnb, but it's not always that you go straight to Airbnb, you might go to Google or the App Store, so a lot of traffic that you see on online sources might just be a result of word of mouth, like very often if you have branded traffic, the number one source is going to be someone telling you about that product. I think for gaming it's a little bit unique because it's not like, you know, travel is like an event that happens all the time, but it's like gaming is a little bit more like taste, because it's entertainment, right? So even, don't get me wrong, people will recommend the game, but you want to be super targeted, and that's probably in general, like I love Scandal, the TV show, and there's probably, I probably wouldn't recommend it to you, but I don't know, maybe you like those kind of shows, like drama soapy, soapy dramas, right? Yeah, so I was like, oh, okay, if you love that, if you love Grace of Nataville, then you would love this. So what's really kind of fascinating about games, the moment looks very different because it's based off of interest, or entertainment in general, it's not just games, but entertainment. And then what happens with word of mouth is it looks a little bit different, so instead of recommendations, it'll be reviews, can be a really big deal, particularly like in the traditional console gaming world, like Metacritic was really big, movie reviews like A Rotten Tomatoes, but now kind of in a different way for us, like, our game Marvel Contest of Champions, what we think about is word of mouth is like, let's go to Comic-Con, let's get it into the hands of the fans, and let's have them tell their other fans, because the best thing about fans is, A, they're short for fanatical, so they're very excited, and then they know other fans, because that's who they want to kind of get together on kind of some of this interest space. So I think word of mouth looks a little different in something that's more on the entertainment side, but still extremely important in this age. Cool. So we have a few more minutes left, and I think one question that a lot of startups and founders here probably would like to listen to more is how do you build a growth team, and where does it sit in the organization, when do you recruit, when do you overall start building a community and the growth, but maybe specifically first, how does that team look like? It should be in growth. In product. In product. Yeah, I really think it should be in product, because it should affect how the, that's obviously it's people that are engaged in trying to affect that, that's in product, and you should make sure that you have like a product manager, and that you have developers, and that you have designers, specifically so you can be able to do like quick changes and be like autonomous. I think that's really important for a growth team to be really tied to product. Yeah, I agree with you. I would just start by the first people on the team should be people that can actually build it. So you need a science engineer, like you said. You need someone who pay attention to data and metrics, and that can be either those people, or they can be a product manager, or you can be a data scientist or analyst. And from there you can probably get going, like that's enough to get started. As you grow bigger, the most important leverage is going to come from engineering data science. So you need a lot of those people involved, and they're going to be very important for it. You also need design and use of research, because they will give you the ideas of where to run experiments. You can't, it's very hard to, well, looking at data to come up with ideas of what, like looking at experiments to come up with new experiments, it's usually not enough. You're going to have to do user studies and talk to you directly about the specific things you're trying to change to get ideas for what you're doing. And then the other thing about building a growth team is you want to have some specific unique way that they're accountable. So give them a metric and say, this is the thing you're driving, let's say first time bookings, that's the thing that we were driving, and you are measured on the incremental, incrementality and like the incremental value you're providing to the product of first time bookings or whatever new metrics that you're driving. One thing I have seen not working is giving that metric to the entire company, saying everyone's responsible for growth. That basically doesn't work. Like it works if you kind of like have had really awesome growth team that have kind of evangelized that to the entire company, and then maybe your Facebook at then at that point it might work, but if you're just starting off that won't work. So you have to put a specific team and a specific individual in charge of specific metrics. That's how you get started. I think we're different in gaming. I think in some ways it's a little different. So before we moved into gaming we did fan communities on Facebook and we were very much an ad revenue generated company. So at that point and we were smaller, we were about 30 people and everybody could think about growth because that's all we cared about. And if we got good growth then that meant like we could bring in the ad dollars, but when we moved to gaming it was it was a bit different. I'm a huge fan of what both of you are saying in terms of like it first needs to be in some ways I want growth-minded product people, growth-minded engineers. I need them to care about thinking about it from almost a business perspective. Oftentimes we would sit there and say, okay, you know, the kind of like I guess analytic we'd care about is growth and what are the things inside of this that's the lowest level of effort with the highest amount of impact. And then we would kind of prioritize from there. And sometimes that's not always your focus because it's growth, retention, engagement, monetization. It's always like a balance between these but because it's on growth you could really be highly strategic around how do you measure that and that's like or not measure that but how do you prioritize that and then that's in the early days because you're like doing these experiments back and forth. But as you do get larger it does become and you're kind of for us as a game and even as a business I'll say for any business the ultimate kind of metric you're measured by is money revenue unless you're a non-profit or what not like that thing it has to sustain itself somehow. So for us as we grew larger even if it was when we were just one game it did become more important for somebody just to focus on what we call user acquisition. So with that person like there was our head of marketing I guess you call marketing her entire job was to figure out how to spend smartly like she just we just gave her a water cash and we're like I need you to deploy this as great as like as well as possible I was like wow your job is just to spend which is phenomenal. And so then it's hard though that's hard. Yeah. Oh it is not easy at all. Yeah exactly it's funny how I thought like oh give me a water and some ways you can say my job is just to buy a lot of stuff and I'm like that's an easy job but it's not it's not easy and that's why we started having kind of a team focused on just that because we needed like a lot of specialization on just kind of some of these user acquisitions but things like they definitely needed to talk together things within the product of like how do I send kind of thinking about those moments of how do I ask them to send something to their friends how do I ask them to accept a trip how do I ask them to really send get an alliance gift or any new kind of player you bring in what do you get what kind of tokens that really does need to live kind of with the product because ultimately all that stuff impacts what they we call the LTV which is the unit economic of revenue per customer in our games exactly yeah but also when it comes to paid acquisition one thing that I've really noticed is that you do it too early and then you spend too much money and then it's like oh this is not working but we've spent like a hundred thousand dollars on this so like you need it's really really hard meaning that you need to do all those small things first to figure out okay where and when how are we going to start working with paid acquisition because it's really hard and what part of it that we're going to start working with because otherwise you can spend so much money it's easy to spend all that money and get nothing out of it it's super tempting to to get on the ad platforms and start acquiring users but if you're not if you're not making enough revenue to cover the cost then you really haven't learned much because like that's not going to use be useful learning because you are not going to be able to spend more money because you won't make the money back so it just like you will learn you'll learn how to acquire users that you can pay for basically yeah as the last question I I figured it would be interesting to hear I'm not sure if this is like a swear word for for growth but the the amazing growth hack doesn't have to be of your own company but one one kind of inspiring growth story that you've you would like to share today as a as a wrap up for for this session okay so for us we found I'm going to take an example from our fan communities because it was on Facebook and so we had these fan communities that were broken up by TV shows so you could be like I love Grey's Anatomy or I love Scandal you could put like quotes on there and then you could be a part of all these other TV fans the best growth hack that we had was basically we would once we said hey now you've downloaded immediately there was an invite but there were all these friends here that were like oh these are your friends that love Grey's Anatomy and once like you know that the intent is high and you're not going to spam your friend because they want like oh you like Scandal too or you like this too it was it just didn't feel that way now you're making a recommendation a wanted recommendation so that's the thing hacks kind of sometimes is kind of a word that feels a little dirty because it doesn't care about the interests and alignment but that's the thing is it's not a hack when you align the interests together right at that right time I think so I don't like the term growth hack but I would say the most important hack at Airbnb was actually building the process of experimentation because that meant we got so many different experiments that turned out to be really positive but we couldn't predict them in advance so I would rather have companies take that hack than to actually take any of the specific ideas that we came up with as a process of that but if there'd be one thing that I would say if you have a website don't log people out it's just very simple just don't do that I put people if you have something sensitive let's say payments just put a login while there but then do like Amazon and Airbnb just don't log people out it'll be very beneficial it's true I can never remember my passwords anymore yeah I get so frustrated it's like what happened there's no I think we're running out of time sorry last quick story just everyone get to share that I'm also not a fan of hacks but I think that this is more a strategy than it is like growth hack I really really like Slack's adaptive pricing model I think that that's really interesting and it's really strategic because what they do also is they don't leave you they don't give you money back when you have users that are inactive they give you credit meaning no money needs to leave them and I think that building that kind of customer trust is really good for attention and I think that that's one thing we'll see much more about in the future coming up in like startups how do you actually do pricing and not just here's a monthly cost being like this is the product value you're paying for the value and feeling that trust with the product is really important super cool thanks everyone this was amazing good discussion it's getting way warm here close to the fire so time to get out to the cold thank you thank you thank you everyone thank you so much thank you