Will Wu on Product Innovation at Snap and Match Group

Foreign I am really, really excited to have my dear friend Will U. Will and I worked at Snap. He was former chief technology officer of Match Group and joined Snap even before me. Very, very early January 2014 I believe and all the way ultimate when left he was a VP of product there. One of the greatest product mind that I have spoken to. So I really, really always like hearing Will. So I thought it'll be great to have Will on this podcast and hear his perspective. So Will, I think will be very helpful. Maybe you can give people a little bit about your background. You know, I know you went to UCLA undergrad and then Stanford for your masters, but walk people through your journey. How did you became a product person and technology? How did you get into it?

Sure, yeah. And thanks for having me, Imran. It's so good to see you again. My journey into tech started from when I was really, really young. I was lucky to grow up in a household where I was just surrounded by the latest technologies and the latest consumer products. My mom was a PhD in chemistry. My dad was an electrical engineer turned serial entrepreneur. So growing up as a child, I always got to play with the latest consumer computer gadgets. Essentially my dad would go to CES every year and bring back some new gadget. And for me as a child, I would take a look at these computers and electronics and just be so fascinated and almost like mystified with how they worked. And to my dad's kind of frustration, I would take my tool set as a kid and like rip apart these electronics. I would unscrew the PCB boards and everything just with a desire to really understand how everything fit together. So my love for technology and products started at a very young age. I was very into computers growing up. When I was around 12 years old, I on my computer late at night, every single night during the school week, I would log on to these Internet chat rooms on what's called irc. And I basically started participating in this community of computer scientists online. And it was a little weird because I was 12 years old, 11 or 12, I was talking to these Internet strangers essentially. But I was in these really kind of geeky nerdy chat rooms where all these computer scientists who are much older than I was would hang out and just like talk about, you know, the latest pieces of software, the latest beta versions of Microsoft Windows that were coming out, and these very kind and generous strangers on the Internet would mentor me. They knew I was like a kid. And I know it sounds kind of weird, but I learned an incredible amount from them. At a time when I was just so curious to understand how technology worked. They taught me about like computer networking and how to set up web servers, how to code. They taught me about cryptography and video encoding and just a whole spectrum of different technologies. So that was very formative for me because it gave me a sort of like early access to understanding the way that really technology fit together at the time. So I didn't really know growing up what I wanted to be. My role model was always my dad, who was an electrical engineer. So like you said, I majored in electrical engineering in undergrad and then I started going to grad school for computer science. I ended up dropping out because a professor that I was working with asked me if I wanted to co found a company with her and some of her PhD students, which felt like a really exciting and rare opportunity. So I did that for a few years and then ended up meeting Evan Spiegel in 2013. And Evan and I really hit it off. He gave me a job offer for what was called a product designer, which I had no idea what that was. And I had always thought about myself as an engineer up until that point. But Evan, I'm very, very grateful to him because he really saw me as a product designer and that really helps to kind of reshape my own identity and my own feelings about who I could be. So I ended up staying at Snap for around a decade or so and then ended up going to Match Group. So it's been a sort of like, I guess, a circuitous journey, but it started at a very young age getting into tech.

But also it's really interesting you talked about, Evan, that you saw yourself as an engineer and you hired you in the product design team. And that's actually, I think his superpower, like identifying people who are not obvious and putting them in the right place. So let's talk about, you know, 10 years. That's a very formidable years of Snap because when you joined, I think January 2024, did you have stories at that.

Time, in January 2014, stories was already in existence. Yeah, it was, it was, it was before chat, before chat.

So the before chat, before group chat, before, you know, lenses, before, before all these things. So what are the 10 years, you know, working in a probably one of the coolest, you know, consumer Internet apps. What are the two or three things that you learned that, you know, other designer or other product people who are building products should know about?

I think one of the biggest things that I really learned both directly from Eben and I mean First I, I guess I just start off by saying like, almost everything that I know about product and design I learned either from Evan directly or from just some incredibly talented designers and creatives on what Evan called the design team. One of the biggest things that I learned during my time at Snap was how every little design detail really matters. The tiniest of design decisions can just have massive fan out effects and trickle down effects on your product and your business. And I remember early in 2014, after I joined Snap, we were working on the first ever version of chat in Snapchat where you could actually text your friends. And I remember Evan was sort of in frustration reporting these design bugs to our team. And he would, he would basically be like, oh, like this, this pixel on this icon is slightly off, like this single pixel. There's like a little bit of aliasing around this corner or something like that. And he would basically be such a stickler for every single design detail and at first was like, oh, wow. I didn't really understand the importance of why all these design details really mattered, but over time I started to get more of an intuition and respect for the fact that they really do. An example of that there was at one point we were basically trying to get people in chat to send stickers, you know, like kind of like Bitmojis. And even before Bitmojis, we had these other like beautiful illustrated stickers. The way that we would inform users that there were stickers they could send was basically by putting a little red notification dot on screen in chat while they were typing. And interestingly, what we found is that the diameter of that little red circle, it's like the difference of just like 1 or 2 pixels. I think we went from like 5 pixels to 10 pixels. The diameter of that red circle, that little red notification dot would have massive impact on how many people actually saw the, the notification dot and clicked into it. So I think we found that, you know, when we actually doubled the size of the red dot, we would have like some huge incremental lift of say like 25% conversion or something like that. And that would have massive effects on how chat as a product felt because it meant that in your everyday conversations with your friends on Snapchat, chat would feel more visual because they'd have more stickers, chat would feel more creative. So it was just an interesting case study for me and how like the tiniest of design details can have such profound effects on how a product feels, on how people actually use a product. Another example of that might be like the Snap Map, which I wasn't, you know, directly leading, but sort of was working with a team as we were building and trying to scale it. And I don't know if you remember that the first version of the Snap Map, you would have to navigate to it by pinching out on the camera. Do you remember that? So that was the first way that we launched it. And we thought it was a really cool, novel interaction that you could pinch out on the camera and it would kind of like zoom out to the map. But obviously it's really hard to discover the Snap map. Well, maybe not that obviously, but, you know, at the time we thought it was cool, but not that not enough people were discovering it. So all we did is we, you know, made this little design change to access the Snap map by pulling down on the camera, just by dragging your finger down. And that, like, in my mind at least, totally changed the fate of the Snap map and drove, you know, tens of millions of more people to it every single day. So I just. I think one of the biggest things that I learned was just this appreciation for how the smallest of design decisions can just have massive effects on the user experience and user engagement.

So I think one of the thing that early days criticism of Snap was that it was very hard to use. Right. A lot of older people like me would probably complain, oh, you know, the MAP example you gave was a perfect example. But it was, you know, but I think it was also, at the same time, it was very exciting for a lot of the younger users because they would find the way to use Snap that nobody else would know. So how did you guys balance that? Right? Because that was always, I think, like, was, you know, as you talk to more older people, they would give this feedback. But then the younger people love those, this hidden gem inside Snap. So as a product design, how do you guys handle that and how do you dealt with it as Snap, you know, keep adding more and more users and more and more older people came to the ecosystem?

Yeah, that's a good question. You know, I think in the early days, Snap was the way that it really grew was via word of mouth. And we really leaned into the. The fact that we're very lucky that we had this product that worked in this way. But we really leaned into the fact that in order to learn how to use it, you maybe needed to get a friend to show you, you know, tell you to download the app and show you how it worked. In my mind, in the early days, that really helped to fuel the growth of it because it almost like promoted word of mouth spreading. But it only worked in my mind because the actual core product itself, you know, sending disappearing photos and videos back and forth was so sticky and compelling to these younger users. So I would say in the early days we really relished that and leaned into the idea of designing the app in a way that was very novel and new. It really helped. You know, these were like in the earliest or very early App Store days where mobile design paradigms were still being explored. And Snapchat was one of the first apps that really you could navigate by swiping around the app. And these days it's very common. But, you know, we got to really sort of pioneer some of these interaction designs and I think over time, so. So that strategy in my mind worked very well for the first chapter of Snap, where it sort of caught on like wildfire within Gen Z, within younger users. And younger users have more time in their day to kind of like explore how an app works. They're like digital natives. They grew up essentially using different types of applications and software, so they're just better at understanding how to navigate these new design paradigms. But I do think that that strategy ended up limiting our growth after we had started to basically saturate Gen Z and there came a time when we wanted to basically also cater to a slightly older demographic. We wanted to age up in the interest of finding further growth in the Snapchat community. So there's sort of like a philosophical change, at least that I felt where it used. At the beginning of Snap, we were very focused on designing these novel interactions, these novel designs that people had to understand just by using. And then in a later sort of mid chapter of Snap, at least for me personally, we were more focused on usability. We were more focused on, you know, creating slightly more traditional navigation constructs like a tab bar at the bottom of the screen that would be more accessible to a broader and older audience. So I guess maybe that's a long winded way of me saying that. I think there's different strategies from a design perspective based on like what your growth goals are at different points in the life cycle of growing a consumer app like Snapchat.

Got it. So let's move on. After Snap, you went to Match Group, you know, again, another big consumer product, you know, business. They have a bunch of brands from Tinder to Hinge. They also unhinged. Right. So what are the big learnings that you had, you know, at Match Group about the dating landscape?

You know, I mean, my time at Match Group was One of the most profound learning experiences of my life and my career thus far. And one of the reasons for that was that I got to experience a vastly different culture, company culture, than what I experienced at Snap. You know, I'm, I guess relatively early in my career, maybe mid career, but Snap was really my first actual job out of college. It was kind of all that I knew. It was my entire career essentially minus my startup which was very, very small. So, you know, sort of doesn't count. And while Snap had a very design led ethos as a company culture, Match Group was very financially minded and business focused. And it was quite shocking to me to join that after having spent my whole life at Snap. Essentially I grew up at Snap and then I went and I joined Match Group which is a much older with incredible financial operators. And when I joined Match, I remember sitting in on some of my first meetings and almost having like a little bit of a panic attack because I felt like everyone was speaking a different language. There was so much like financial vernacular that, you know, for you as an investor it would be very easy to understand. But for me, I'm sitting in these meetings and people are talking about like the AOI margin expansion and like RPP revenue per payer. And this was basically like, you know, the, the everyday language that people spoke down to, even ICs at the company, you know, IC product managers at the company would have a very strong financial mindset. And I felt like I had to learn a whole new language in order to figure out how to like make that transition to such a different company culture. So to me what was such an interesting learning was the fact that these two apps or these two companies that have similar apps serving similar user bases, you know, Snapchat and Tinder were founded I think in the same year in 2012 and used by the same types of, you know, users, younger people, Gen Z, et cetera. It was so fascinating to me that the like sort of operation of these apps behind the scenes could be so vastly different. And I think there's pros and cons to, I'm not saying either is, you know, better than the other in terms of company culture. There's just I think pros and cons of both which I found to be really interesting.

What are some pros and cons in your mind?

Match Group was so successful at driving the business, at driving revenue growth, at margin expansion and I think one of the reasons that they wanted to bring me in was that they wanted like a little bit more innovation and sort of design led Design thinking, creativity in the company. And you know, I was lucky to get that opportunity. And I brought in a bunch of super talented folks who could sort of manifest that ethos. One of the things that shocked me joining Match was that they're a very profitable company. And for some reason after being at Snap, where getting to profitability was sort of like always one of our key goals, I had thought that joining Match, I was joining this very profitable company and everything was just going to be so golden and so easy, which was such a naive thing for me to think. But in actuality what I found to my surprise was that although Match is highly profitable, they are still in this turnaround situation. What is happening in the world is that people are saying that Gen Z is dating less, Gen Z is having less sex, they're getting married less. And for a business like Match, which, you know, is in the business of basically getting people into relationships so that they essentially stop using the apps, it's critical for them to always attract the next generation of daters, the next generation of people who are, you know, just starting to think about sparking romantic connections. And I think in order to attract the next generation, you really have to stay innovative and creative and design led to really understand how to solve problems for, you know, the next set of users that you need to be attracting to your business. So it was an interesting learning and just understanding that even a highly profitable company could still be in a turnaround situation and a learning that it's really necessary to have both a strong financial revenue mindset as well as balance that with a consistent approach to innovation and creativity and design. If that makes sense.

That makes great sense. So I do have a question though that sparked me. If, you know, as you pointed out that in online dating you continue have to bring new daters in the ecosystem and each generation speaks differently, has different, different way of connecting people. Do you think that there could be an universal dating app or do you think that companies like Match Group continue to have to find newer app that's attracting to the newer, you know, dating cohort and acquire them and that's the only way they can be long term sustainable because, you know, you think that like, you know, Bumble comes out, you know, obviously that's not part of Match Group do really well and then suddenly, you know, like one day it's a cell blow up, right? Like I've been looking at this dating businesses going back, oh my gosh, 25 years and it always feel like it's like these businesses are like running on a Trade belt. Right. At some point it just doesn't work, and you just don't know when it doesn't work. So is that a fair assumption or what's your view on that?

I think that there are generational preferences when it comes to these, like, social consumer apps. I think that it is hard to maintain an app and a brand that appeals continually to the next generation. I mean, we see this with even, you know, beyond dating apps, with social media, the fact that there was a certain demographic that started to use Facebook, and Facebook became so popular, but then Facebook sort of became less cool. So then there was Instagram and, and Snapchat and TikTok, et cetera. I think it's just sort of hard to have these social consumer apps remain, you know, the hottest club in town, essentially in perpetuity. And you see that analogously in the online dating world as well. You know, match.com, which was the, you know, how match group essentially got started in the 90s, the first, you know, one of the first websites in the web 1.0 days where you go to do online dating. It still exists today, but it's less popular with the younger demographic because the brand ethos, the user base has sort of like aged up over time. I mean, I guess it underscores the importance of innovation and design and also maybe M and A just to make sure that for, for the operators of a business like that, you're able to continually attract the next generation, which is just so critical in the longevity of, you know, a consumer business like that.

So any interesting tidbit from looking at all this different dating app, like, what makes them go viral or what's, you know, make them break out.

I don't know about, like, what makes them go viral in, in, you know, in actuality, it's. It's quite hard to get a new dating app to go viral and break out just because of the fact that the whole product premise is that there's other people using the app that you can connect with. So I think dating apps are sort of notoriously challenging to bootstrap with a critical mass of users. But once you do have that critical mass, obviously it becomes a big moat because it's hard for competitors to really copy that. I think word of mouth certainly goes a long way. The brand is very, very important. You know, Hinge, for example, is doing phenomenally well right now as a dating app. And they have an incredible brand, which is not an accident. They put so much thought into, you know, creating a brand that's really in service of users designed to be deleted really. So I think it's like a brand ethos that people have come to really trust and respect. Another interesting, I guess tidbit or learning for me from Match was the benefits of operating a portfolio of brands and businesses is very different than what I experienced at Snap where, you know, the primary business was Snapchat. Match Group has something like 30 + different dating brands within its umbrella. And they've got different apps serving a wide variety of different audiences. They've got apps specifically geared for daters in Japan for example, and Japan has very different dating cultural norms than, you know, somewhere like America. They've got apps specifically geared for different affinity groups. But then they've also got this sort of like general population apps like Tinder and Hinge. But it was interesting to me to see the benefits of operating a portfolio like that. And one example, one tan or maybe two tangible examples of that one. I can't claim any credit for this, but during my time at Match, you know, there's always sort of pressure from Wall street to find ways to increase revenues and revenue per payer. And there was basically, I basically got to see how you could have a learning in one small app within the portfolio, but those learnings were universal and you could actually propagate those learnings across the whole portfolio. So there was an instance where in one very kind of relatively small dating app that had just a few million users, very small from a revenue perspective, we experimented with different subscription offerings, tweaking the subscription offerings essentially. And we found that it actually worked very well. And when we propagated that across the whole portfolio of Match Group, it ended up being something like 150 million annual run rate revenue, run rate win. And to me it was just so interesting that you, like, you have one, you have a little test bed in the form of these small apps where you can experiment really quickly without worrying about the downside impact too much on the overall holistic business. And then once you find things that work in that test bed, you can propagate them across the portfolio. And even like these tiny little tweaks can result, you know, due to the fan, due to the fan out effects can result in huge, you know, 100 million plus dollar revenue wins. So that was just a really interesting learning for me, understanding how to really operate a portfolio business. Another example of that, I was working with the Hinged, the Hinge team and we basically applied some insights and learnings from Tinder where we saw that on Tinder users could have AI a Machine learning model automatically select the best ordering of their Tinder profile photos in order to garner the most likes on Tinder. And we basically took that learning, applied it to Hinge and created an ML model that automatically changes, chooses the best Hinge profile photo to be at the top of your profile, like the first profile photo that everyone sees. And that had huge, huge positive impacts on outcomes for users on Hinge, where when they actually used this top photo recommended by the AI, they would actually go on significantly more dates. They would actually meet up with people much more frequently. So that was actually one of the biggest, one of the first product initiatives in, in a while at Hinge that really significantly lifted the number of people going on dates every month. So to me, I guess just understanding this portfolio mindset of operating these different businesses and these different products was a huge takeaway from my time at Match Group.

All right, so we should talk a little bit about AI. Know the AI is also, you know, we didn't have a lot of consumer, new consumer product for a long time, but then AI brought in, you know, all ChatGPT and other consumer product perplexity and others. What do you think? I know. What's your view on AI? You know, how do you think it's going to play out? You know, what's your view about that new kind of consumer product can be built on top of AI? How are you thinking about it?

Since the launch of ChatGPT, I have been almost like the ultimate AI optimist. I, I was lucky that when ChatGPT came out, I was the CTO at Match Group and I was therefore made sure that we were getting chatgpt into the hands of all our employees as early as possible. I mean, I think that AI is really going to change every facet of life as we know it over the next decade. I'm actually really excited to see it pan out. I know there's also some concerns about, you know, what that means for humanity if we, if we do actually build this artificial general intelligence that's basically smarter than every human on earth and can simultaneously, you know, reason across every single knowledge domain. But I think just having programmatic access to intelligence and reasoning is going to improve our lives in so many ways over the coming years. So, you know, right now actually I'm for fun using AI coding tools to, to prototype different ideas, to explore different possibilities when it comes to consumer AI products. To me it just feels like a greenfield opportunity. It feels almost like every legacy product out there stands to be disrupted or significantly improved with AI. What is your view on AI these days?

Pretty similar, obviously. Huge. You know, I think we are so early innings of AI. You know, it reminds me of 1995. People are building optical network and ISP services. You know, ChatGPT to me is like IRS. You know, I see her right, that the communication, we're talking to people. I think what we have seen is that the greatest value gets created when you take that technology and disrupt an existing businesses. Right? Online travel disrupted this massive travel industry, or Netflix disrupted this DVD rental business, or Amazon started as a book business. And same thing happened with the apps. If you look at apps, Snap was just a better texting apps than imessage. So new product create a better way of doing things and that's really, really create massive value. But I do have to ask you, how do you think this AI impact dating? Online dating apps, I think it's going.

To empower people to find higher quality matches, essentially. I think that AI can get to know each individual dating app user much more deeply and thoughtfully than ever before through something as natural as just talking, you know, like to a language model, essentially. And through that natural conversation, I think the dating app or the AI can really get to understand each person's preferences, their personalities, their sense of humor, their hobbies, et cetera. And over time, given that deeper understanding of each person, we'll be able to give them recommendations for people to connect with that are basically more compatible or have higher likelihood of compatibility. So I think it's still early innings for AI in online dating, as it is for so many different categories of products. But I'm very bullish on just the ability for AI to basically be a better matchmaker for people around the world. I kind of think of it like human matchmakers. People have been working with human matchmakers since the beginning of time. It's always friends of friends or something like that. There's an informal matchmaker, there's professional matchmakers in the world right now where you can pay them a bunch of money to have them look into their Rolodex of eligible singles and try to pair you up with someone. A human matchmaker is limited, constrained by just how many hours in a day that they have, how many clients they can talk to at any given time. But an AI model is completely unbounded, unconstrained. So an AI model could theoretically be talking to millions of people simultaneously to deeply understand who they are and what they're looking for and then be able to match those people with someone who's highly compatible. So I'M very bullish on AI and the ability for it to have a positive impact on online dating.

Great. This is a good way to end. Will, thank you so much for making the time and it was a wonderful conversation and great to hear your insight.

Thanks so much. Imran.

Thank you again for tuning in and I really hope you enjoyed that conversation as much as I did. Subscribe to our show on Apple or Spotify or wherever you listen to your podcast so that you don't miss an episode. See you next time. All opinions expressed on this podcast are solely those of the host and guest, are for general information purposes only, and do not constitute professional investment advice.