Want to join the conversation? Connect with us on Tribe.
Find “Founders in Arms” on Apple Podcasts, Spotify, and YouTube.
Transcript of our conversation with Roger Dickey:
Immad Akhund:
- Hi everyone, welcome to the Founders in Arms podcast with me, Immad Akhund, Founder and CEO of Mercury.
Raj Suri:
- And I'm Raj Suri, co-founder of several startups. And today we have Roger Dickey, who is know a founder's founder in my view. He's been creating interesting companies for I think 15 plus years. You had an exit with selling Mafia Wars to Zynga and then you started Gigster after that. I think you were running a studio and Gigster was one of the companies out of there. And and you've also started more companies since then since you've moved to Austin. Roger's one of the most creative guys I know in the tech business, in the tech industry. And that's saying a lot. I mean, a lot of creative people here, but Roger has like so many good ideas. He's the guy I go to whenever I wanna talk about ideas. I think you have some crazy long list of like startup ideas, Roger, right? I mean how long is your list these days?
Roger Dickey:
- Oh man, I mean, since running two startup studios in a row, actually three, if you count the studio, we, we built our game out of, I've got this list of ideas over the last 15 years, probably a hundred ideas in it. I actually, it's a shared Google Doc, so I can, we can share it with the audience afterwards if people wanna see it. 'cause I can't count how many times I sit down for coffee people and they're like, what should I be working on? Or what's cool in ai? At some point I was just like, I'm gonna write all this down and make it shareable.
Immad Akhund:
- How do you come up with new ideas? Like what's your process?
Roger Dickey:
- So we had, at the last startup studio, we had a very like, kind of structured process. So we call it the Matrix Approach. We take spaces, you know, like personal finance, healthcare, you know, so like social gaming. We cross those with new technologies. So things like, you know, mobile AI, blockchain, VR whatever. And within, within each one of those, you know, cells in the matrix, you have a row in a column, you know, and again, the column is like the space and the row is a technology or an approach. You know, for example, like Lambda School pioneered like ISAs. So we're like, well what can we use ISAs for? What is ISA meets home ownership or ISA meets personal finance? So we would brainstorm in each of those boxes. I would send out to the team, like every, I think Tuesday and Thursday I'd send out a list of 10 of those cells and I'd say, Hey, go home, think these over and bring your best like three ideas in each cell. And then we'd get together as a team and just do a long team lunch and kind of debate ideas. We'd write down our favorite ideas and everybody had to have like a, you know, like a two or three sentence pitch for each idea. And then after that, you know, we would have people write a one pager for like, kind of the best three ideas of the week. And then that one pager might turn into like a little prototype we'd build later. And it, it kind of went from there.
Immad Akhund:
- And you did this process like 15 years ago?
Roger Dickey:
- Yeah, I did this at three different startups actually. So the company that became Mafia, we did a version of this and then the company that became Gigster, we actually built 15 different prototypes for, we found the idea for Gigster and before the prototypes, I mean we generated I think 800, 900 ideas and that was for Gigster. We call that Liquid Labs.
Immad Akhund:
- Do you, have you gone back to those ideas and has any one of those ideas, like you did Gigster, but was there another, another idea that ended up being like a decacorn or like a really, really big company and you were like, oh, I should have done that one?
Roger Dickey:
- Yeah, for sure, for sure. Some of 'em we just missed or like, they didn't play to our strength, so we don't wanna build them like, but yeah, absolutely. There's some good ideas on that list.
Immad Akhund:
- Raj, what's your process for ideas? 'cause you, you're doing like three, three right now, like is, do you have such a structured process for coming up with things?
Raj Suri:
- I have a long list and actually Roger and I compare lists sometimes. So we, we, we, we talk about our list. I mean, my process is pretty straightforward. It's like problems that I've had personally that I think are are more global problems and, and also things that I've observed like in the world, you know, now I've had a bit more experience so I can, I've observed like some like discontinuities in the matrix, so I, so things that like are working in one industry but now working in another industry.
Immad Akhund:
- Discontinuities in the matrix.
Raj Suri:
- Exactly. Yeah. There's some, there's something strange go, I mean, for example, like, you know, with Presto, we, we actually went public, right? So we went on the NASDAQ we, you know, we, we rang the bell and stuff and it was a very unique experience, right? Like going through that, that NASDAQ thing. Then I was wondering like, why can't, why why are more like celebrations not like this, right? Like the why are they less visible? No, or not as visible as like the NASDAQ celebration. And it turns out that a lot of people now just try to use the NASDAQ and they go and they actually take pictures in front of Times Square for every, everything even for like series A and stuff like that, right? But I think there's actually like this market for celebrations and for visible celebrations in the world of social media. And, and so that idea was sparked by that experience, right? It, it doesn't necessarily mean it's gonna be a billion dollar company, but it shows that, you know, it, it just kind of gets your, your thinking when you have an experience. Roger, I think you're a parent now, right? So there's a lot of ideas around, I dunno if you've been generating ideas around parenting, you know, companies, 'cause there's a lot of things to do there.
Roger Dickey:
- Not yet. No.
Immad Akhund:
- I find that my process is like, I guess somewhere between the two of you. Not quite as structured as Rogers, but I, I like being, like getting really annoyed with things, and America is like full of super annoying things, right? Like, I mean Mercury really came from this. I was like, oh my God, I hate banking and it's so annoying. And I think like that with like going like top down and saying, okay, what is the biggest kind of GDP sucks in America, right? Like finance is a huge one and then it's like defense and healthcare and a few other things. So that's like a good kind of set of things to go after when they're like really huge tams.
Raj Suri:
- Yeah. Roger, a lot of your ideas also consumer. I I know you, you have a lot of consumer ideas,
Roger Dickey:
- Mostly consumer. Yeah, that's what I have the biggest passion for. I mean, I really caught the bug for consumer when I was at Zynga post acquisition of Mafia Wars. It's just incredible to like wake up every day and you know, see that you had 5 million active users the day before and you look in your database and like every record in your user table's a real human being who is having like a human experience using your product and telling their friends about it and you know, playing your game with their, you know, grandfather, grandmother to stay in touch. There's so many of those human stories and experiences we had at Zynga and at the games I made before I was at Zynga, I think I, I think my biggest game had a couple million users before Zynga. And every day I would go through my, my user table and I would just copy and paste email addresses out and email 10 people and I'd ask 'em a question like, Hey, I saw that you got to level seven but that you haven't played in like two months. Why is that? And they'd be like, oh, like I just got accepted to grad school and, but I loved your game, blah blah blah. You know, usually you learn so many things and that connection with that number of people I think is, is is really special. And of course in B2B you kind of have that, but people are using your product as more of a tool. It's a different relationship.
Immad Akhund:
- Do you feel like it's kind of, I guess, I don't know, sad or something that like there hasn't been that many startups in consumer in the last 10 years, right? There's, I'm having a hard time easily going like, oh, there's a huge kind of startup in consumer and I guess Uber, but that was more than 10 years ago.
Roger Dickey:
- Yeah, I, I think so a little bit. I mean even like ChatGPT is very much a consumer thing. I think there are a lot of these kind of like dual, I'll say dual use even though that has its own like defense definition. But you know, we'll say dual use consumer, you know, enterprise consumer B2B, there's a lot of those like, like Figma, like I was very envious of that. You know, this guy Dylan Field sat there and just built this like vector editor in a browser, which I could code in like, not probably six months, nine months. I could build something with 80% feature parity, shipped it, seeded it with the right users and it just blew up, you know? And now, and it starts off, it's like the, you know, classic grassroots B2B SaaS thing where people are using it on their own, like individual creatives are using it and then they kind of upgrade. 'cause 10, 20 bucks a month isn't a big deal and suddenly teams at companies are using it and suddenly it's this like massive SaaS tool.
Immad Akhund:
- I do think product led growth is, is almost consumery, right? Like I think there's, like Mercury has a lot of that. Like we, yeah, we have a few hundred thousand users and we can go talk to them and all of that.
Raj Suri:
- It's really fun talking to users. I really enjoy that piece at Presto where I can actually go to a restaurant and see people using the product and actually approach them and like, and actually have a conversation with them. And that should also observe them using the product. That's actually the benefit of using, of a hardware product is you can actually observe people using the product kind of stealthily, right? And you can see where they're stuck and where they're laughing and stuff like that. It's really, it's really a fun experience, you know, with a hardware product it's harder to do that, you know, when you have a mobile app like, you know, for example tribe, you know, a group chat, you can't really see people using the product. It's, it's all, you know, you have to do use analytics and stuff. Yeah. There is this category where you mentioned Figma, which is like, would you call that product led growth? And it's like basically like you build a great product and you kind of see where it go gets to
Immad Akhund:
- Yeah, a hundred percent just like self-service. It's, you know, distribution even to big businesses is happening by like individual employees kind of picking it. I think that's like the quintessential product led growth.
Roger Dickey:
- Yeah, I think of, I think of product led growth slightly differently. Like I could see putting that, putting it in that category, but to me it's, it's something that's inherently like the, the the core unit that you're building is inherently shareable, which makes it go inherently viral. Like something like Dropbox was the same way. You're, you're you're putting files in your folder but you're sharing them.
Immad Akhund:
- Yeah, I think that's different. That's like, that's from your Zynga days, like virality built in. I I think that's really powerful, but I feel like that is a different word for it. Like I guess like B2B virality or something is maybe the word is, there's very few examples of that though.
Raj Suri:
- Yeah, there, there any kind, anything that has a social element will have that, right? And so yeah, Figma would have that. Actually Tribe would have that too. There's a few others. And actually Mercury has that too, right? Because you have like these like payment requests and, and stuff like that, right?
Immad Akhund:
- Actually little growth hack that was never truly able to prove whether it works or not is like when a relatively early on when you send in payment with Mercury, it said in the memo via mercury.com. 'cause you know, normally the memo line is Yeah, exactly. But I never really quite proved whether it like drove usage or not. But I mean, must have a little bit.
Raj Suri:
- I can prove it for you. It works like, because I use it all the time and like people are like, Hey, what's Mercury like? I, I have these conversations all the time. I work with a lot of people who don't know Mercury outside the industry, so they're all learning about Mercury through these like payment requests and things like that.
Immad Akhund:
- I saw this really cool feature. I was talking to one of the execs at Wise yesterday and you know, wise is a, it has a bunch of features, but one of the things is like you can store balances and right, like very prominently in the app they have, they have a feature where you can hide your balances, like top left eye, top right icon. And the idea is that they, they want people to show wise to their friends, but no one wants to go like, oh look how much money I have. So they have this like really prominent feature, but it like prompts people to go, oh I should like, you know, why does this icon exist and you click on it and your balances, it reminds you to do word of mouth for them. It's super smart, right? I hadn't thought about like, I mean people think a lot about like generating virality, but people don't actually think that much about generating word of mouth. I thought that was like a pretty interesting kind of take of it.
Roger Dickey:
- That's classic product led growth in my opinion. Love that feature.
Raj Suri:
- What do you think of Wise, by the way, Immad, we never talked about this, but Wise has also really kind of grown to become quite dominant, right? In this money transfer space.
Immad Akhund:
- There's like a little bit of FinTech in London. It's like the, one of the few things that's working in Europe is like FinTech and London. And it's just because a lot of it's driven by like international money movements, because in Europe people travel way more between countries and whenever you travel between countries is so frustrating because all the previous kinda, you know, banking and financial payments were like basically price gouging on like those customers. So Revolut, Wise at least mostly exist because of that environment. And now they're kind of exporting that across the world. And yeah, I mean, wise is a great company. I use it all the time to send money back to the UK or whatever. If you look at their whole mantra, they're like trying to like, can you completely suppress the margins in this space? They're like, we're gonna take 10 bips and wherever we can we'll reduce it, reduce the pricing. I I think it's like super smart and really well executed.
Raj Suri:
- Roger, what are you working on these days?
Roger Dickey:
- I am back to back to the drawing board, looking at starting venture number four, of course looking at ai. But I think of AI as a horizontal, like AI is a tool I could use to solve a customer problem, right? So like what customer problem do I wanna solve? And of course it can consumer looking at personal finance, I think finance is just a lot more complicated now. And you ask people where they put their money or how they're spending it and you know, or what a dollar even means. Like it means one thing if you live in Thailand, it means, you know, another thing if you have a million dollars in your bank account, you live in like New York City in a expensive high-rise apartment, you know, so the, the, the actual value of a dollar is, is very subjective and how to invest. Like you ask people, do you feel like, do you feel like you're investing your money? Right? And almost nobody says yes. So I think, you know, there's a level of conviction and confidence that our personal planet tools aren't giving us. So kind of that's scenario I'm thinking about looking at gaming again. I think there's a lot of alpha still in gaming and I wouldn't mind getting back to that. And then also social. I think our social networks aren't really serving us. They're creating a lot of division, doing a lot of really bad things for, for us culturally. So I'm looking at that as well.
Immad Akhund:
- In gaming, there's a few things I've seen recently and I'm an investor in a couple of them. Generally I actually avoid gaming 'cause you know, I was deep in the gaming industry and I in my previous company and after you are deep in gaming or really in anything, it's hard to invest in it. But I feel like there's a few crypto games. I guess what's your take on crypto gaming? Like, do you think that's like, that's a horizontal that's that you're exploring or you think it's like not interesting?
Roger Dickey:
- I would, I'd say more not interesting. I'm not really sure what blockchain adds. I mean, you look at the promise of it and often it's positioned as you have portability, you have portability of game assets between ecosystems. Like I can get a sword in Minecraft and I can move that, you know, golden sword to some other game that's not really a burning need though. I think there's a lot of asset movement inside of an ecosystem, but asset portability across ecosystems hasn't really been figured out. I mean that if you just think like, you know, a little bit into the future on that, like that requires a lot of alignment between platforms about what does that game item even even mean and how powerful is it? What are its attack and defense points or however you have it game balanced. How does that then move into a new ecosystem or game until you solve that? There's really no use for blockchain. You can just use a database if you're porting assets around inside of an ecosystem. So I haven't really seen a need for it. I have invested in three crypto gaming startups, but that's because I love the founders. They have prior exits doing non crypto gaming.
Immad Akhund:
- I wonder if you did the same ones I did?
Roger Dickey:
- Yeah, we could compare notes after. I mean, probably a couple of them. One of them was in Andreesen Speedrun, which I think is smart. I'm glad they're doing that. Yeah, a couple of others.
Raj Suri:
- What are you excited about in gaming? Like, I'm curious. I mean I'm, I'm a gamer. I love playing games. What excites you in gaming in general?
Roger Dickey:
- Like, it's just incredible to see like how much the market has grown, how much more mainstream games are now. I think mobile really helped with that. Helped push games into the mainstream where everybody from like toddlers to grandparents are playing, are playing games. It used to be a lot more hardcore. So I, I think the market expansion is, is really exciting. It just means there's a lot more ideas you can go after. There's a lot more surface area, a lot more alpha. I think there are new business models too, like virtual goods and like, you know, in app payments that make it a lot more viable to, to do performance marketing effectively. Like you look at most games and like the way they're growing is kind of performance marketing. So I think the aperture's like wide open in terms of like what types of games you can build. And it's a bunch of crazy ideas. Like I know this guy who runs a studio in Thailand and they just made this game that's about like knitting or something and you're just like weaving thread. And they won like the Apple design award for games. 'cause it's just this really like, fun, relaxing game to play. And that game has like no point, right? Like nobody would've made a game like that 10 years ago, but now they're just so many. There's, the market's just so big. You can almost do whatever you want. If it makes you like, feel good to use it, turn it into a game.
Immad Akhund:
- Is there any like, new trend? Because like, I feel like everything you just mentioned also existed about like 80 years ago or something like that. Do you feel like there's AI or something else that makes like a new why now for a game? Oh, do you think it's like actually just not enough people looking at gaming anyway, so there's lots of opportunity regardless of like a why now?
Roger Dickey:
- I think there's still a lot of opportunity. I mean, people are, people are trying to use AI to make games more personal or more custom or like procedurally generate in-game assets or like, or make smarter NPC characters. I don't really think AI is quite there yet. And I don't think, I don't think like, I, I don't think there's an unlock around making worlds more deep or immersive. Like I don't think that's why people weren't playing games before. I think worlds were deep and immersive enough that there's certainly enough single player content and games that take a hundred hours to finish. And a game like Minecraft, you don't need AI because you're playing with humans and humans are more interesting than ai and there's more unique and generative content that's being made by humans that you see when you go in. So I don't really see where AI is gonna be a big unlock right now. I think it's just kind of, as you said, like more Cambrian explosion of like people exploring this like very large market.
Raj Suri:
- There's still a lot of innovation right, in gaming and, and social networks and all these types of different markets which are seen as kind of by VCs as mostly like, you know, uncool or passe or you know, like, you know, I think VCs get up too obsessed with like, oh, there hasn't been a big exit in this space for like the last 10 years, so this space is dead. And, and as founders I don't think we, we believe that, right?
Immad Akhund:
- I mean, VCs would've said AI was completely dead when open AI started. So like those are probably the places to look at. To be honest,
Roger Dickey:
- AI's died like six times over the last 20 years.
Raj Suri:
- Yeah. Yeah. There, there's a lot of cool things I think in gaming. And I, I agree with that thesis. I actually know also people like working on like AI gaming, like, like basically easier game editors or game creators kind of tools.
Immad Akhund:
- But do we think the, that the future will be like, you prompt engineer a new game, you're like, Hey, I like knitting and I'd love a mobile game where you could just like knit all day and then like, then AI kind of generates the game.
Roger Dickey:
- I think that future exists for everything. I think SaaS is in serious trouble conventionally. I mean, a lot of SaaS tools are just databases under the hood, right? So you could, you're able to just generate your own UI that works like for your exact business. I can see that for gaming as well. But at that point, like while I'll be out of a job anyway, we'll just be living in some like UBI post scarcity economy,
Raj Suri:
- I think like translating like a Figma set of mockups to like a game makes; will be very easy, you know, at some point, right? Like where you can just take mockups and like translate it. And of course you can generate the mockups with AI too, but I think that'd be a bit harder.
Immad Akhund:
- I mean, I, I'm a bit skeptical that you could make a deep game that with that I think you could definitely make like a relatively simple like hair leveling and it gets slightly harder and it's like a casual game. So maybe like most of the talent goes into like deeper kind of world creation type things.
Roger Dickey:
- Yeah, and I, and I think also, like right now a lot of the best AI products are like AI plus a bunch of like rules-based, more traditional software. You're seeing this with some of the coding tools, like, I think I just saw that this, this one regenerative coding tool that's outperforming oh one, it's not even like LLM based, right? Or it's like maybe partly LM based. There's another one from the former CTO of GitHub that's part LMs, but they also do a lot of kind of dynamic. Like they'll take what the LM outputs and then they'll like actually try to like run it in a virtual machine and see if it works and often it doesn't and they have to like correct it after the fact. So there are a lot of like hybrid approaches right now, and I could see if somebody's tried to make like a generative game builder, you couldn't just plug it into open AI as it is right now. You'd have to have like some rules engine. And to your point, like if you want leveling, you'd have to be able to like put in a leveling system and the L one might not do that. So you might need like human built modules, but if you glue those together with ai, you could think you could get something pretty interesting.
Immad Akhund:
- You also have some AI players that play the game and say, oh, it's a fun game. Like it just goes around in a big circle.
Roger Dickey:
- Testing. That'd be good. Yeah.
Immad Akhund:
- I thought it would be interesting just to change the subject a little bit to talk about kind of the entrepreneurial journey. I felt like, you know, all three of us have done lots of startups and like seen highs and lows throughout it. Roger, when you think about kind of your Gigster experience, I remember for a while it was a super hot company then like, you know, I guess it was under the radar and it sounds like you gotta exit out of it. Yeah. What makes you I guess A, handle all of this kinda turmoil, I think, and then B, come back for more?
Roger Dickey:
- I think that's not really maybe talked about enough. You know, it's, it's not, I think a lot of people fetishize doing startups and being a founder and think it's this like, awesome, cool job and you know, there are, there are a lot of things that are great about it, but it's really for people who are, who are like willing to, you know, chew, you know, e eat glass and stare at the abyss or like whatever Elon Musk said. Like you have to just be born like wired to want to do that and to wake up every day and have a different problem or challenge to solve. You know, people think like they're done when they like launch their, their company's product, but like, as we all know as founders, like every day you wake up and there's like a different puzzle to solve. It's like, oh, I'm trying to hire this executive who I wanna poach out of, you know, Dropbox. Like, how am I gonna get this person? I'm gonna have to like get on the phone with their wife and convince them to join, you know, whatever it is. Like, there's just everything is a little project, a little challenge. And as a founder you have like thousands of those before you reach, before you reach an exit, you may maybe 50 of those 1000 come like before you launch, but then after that there's another like 900 that you have to do. So you have to just like love to like start and successfully finish a lot of diverse little projects. And I think if you're that kind of person, you, you have to start companies because there's nothing else that's like it, you know, I used to tell people, like, I didn't, I didn't wake up wanting to like be a founder or start a company, but I like to work on hard problems with smart people and I want to make money. And if you put all those things together, like, you know, you have to start companies. You know, if you wanna get adults to work on projects with you, you have to pay them.
Immad Akhund:
- I made this analogy actually earlier this week that like doing startups is kind of life on hard mode just to go to the, the gaming conversation. It's like, yeah, I mean you could probably have a easier life, but it's kind of fun to play life on hard mode. Yeah. Raj, what do you think, what was, what was one of the, I don't know if you want to get too personal here, but what was like a really low point for you in kinda your startup journey and like yeah, what did you learn from that
Raj Suri:
- Many different low points, but you know, luckily, like I don't get too low, so I tend to bounce back relatively quickly. I think at Presto we had a lot of big enterprise deals and they're very binary outcomes. You know, it's a little bit like elections, you know, you, you have like, you know, long campaigns basically running pilots for these big customers and at the end of like, maybe even years of effort, they'll tell you whether you won or lost and you could get everything like tens of millions in revenue or nothing, right? And for a long time I didn't know how to win those deals, but, you know, I was trying to figure it out for first principles because there's actually not that much, there's actually not that much knowledge of how to win enterprise deals in Silicon Valley. And it's hard to find people who have really truly done it before and can teach you. So I had to like basically learn by failing a bunch. So I had a, you know, a couple of instances where, you know, campaigning for a big deal over multiple years, you know, whole team is super invested in winning. Everyone's working late night and then at the end of it they're like, Nope, we're not gonna go with it. You know, we'll delay it for five years or we'll go with a competitor or something like that. And those are heartbreaking, you know, now when, when that happened, and the good news is, I learned how to win those deals later on and we had a much, we had a, we had basically almost a hundred percent hit rate at some point, but, but learning the hard way is just like, it's tough.
Immad Akhund:
- Yeah. That's challenging. You know, weirdly, one of my hardest points, and I wonder if Roger relates to this, is like straight after I sold my previous company. And it's weird because it's supposed to be like the, the journey, like the end goal. Like it's supposed to be like, oh wow, you did it. Like, but it was just like, it was really difficult. Like it was a very long nine month process and then at the end of it you kind of lose the baby that you'd spent all this time kind of nurturing and creating. It's, yeah, it's weird in startup world that these like, these like end goals are often like disappointing.
Roger Dickey:
- Yep.
Raj Suri:
- You know, one interesting thing, I was, you know, thinking about is like, you know, the process of finding product market fit is actually a very important skill that founders have that they always need, you know, at every stage of the company. So like in my, like probably right now you're launching a bunch of new products and you're thinking about like, how do I get product market fit with this new product? Right? Like, so that skillset is probably useful at every stage.
Immad Akhund:
- It's actually almost harder when your company's bigger because you can, yeah. When you're like, when you have nothing, there's no way to like fake product market fit easily. Whereas like at Mercury scale, we can launch something and we can get like a thousand people using it very easily. And then you're like, oh, did did it actually work? Did they actually like it? Or it's just because we have a funnel of we can like send towards the product actually is tricky. When you tried all those different ideas, Roger, was there like a, like a thing where you were like, oh, like this is product market fit. Like was there like a metric you looked for? Like how did you decide to like double down on any one idea?
Roger Dickey:
- Yeah, I think your, your analogy at Mercury of like false product market fit, that's something we would think about. You can fake product market fit if you raise a bunch of money and like pump performance marketing into something like, it can make you think that it's taking off, but it's so important to be able to listen to those like kind of quiet signals and not overwhelm yourself with kind of, you know, some something inorganic. So we would try as much as possible not to spend money on launches. What we would look for is pull this feeling of pull or like overwhelming demand. Emmett Shear had this quote about product market fit saying it's like, it's, it's like turning a light on in a room. Like it's just a totally different experience. Or another analogy I've heard is like, when you don't have product market fit, you're like pushing this boulder up a hill, but then when you do, you're chasing the boulder down the hill. It's like you got into the top and now you're at the top and the boulder's like run going downhill and it's like crushing trees and houses and people and you're like, whoa, boulder, go a little left, go a little right? You're kind of like directing it, but it, it's taken on a life of its own. And I think when a product really works, we feel that pull and we feel it start to just be this kind of like runaway train of people telling their friends about it and kind of joining, even at a small scale, like a hundred people, they tell like 20 of their friends, you know, you've got something pretty special at that point.
Immad Akhund:
- Yeah, I think, I think distribution becoming easy is like, kind of how I describe it, but barrage in like enterprise sales, I've, you know, you spent a years probably getting to what felt like full product market effort, right? Like I feel like in some spaces it takes like way longer to figure it out.
Raj Suri:
- It does, yeah. And, and I, I think there's a model. I mean, I, I know in Silicon Valley, you know, folks, and especially, I mean, in venture studios, right? People will spend a certain amount of time on a comp, on an idea and then like if it doesn't take off, they kind of move on to the next idea. And I think both at Presto and Lyft, I mean it took over five years to get to product market fit, you know, and I, I don't see many founders like sticking it out. There are, there are stories of founders doing that, but it's pretty rare. I think,
Immad Akhund:
- Did it take five years at Lyft?
Raj Suri:
- Yeah. Well, you know, Lyft started at Zimride, and it was 2007 and this was way, this was even before the iPhone came out. And, and Uber wasn't even, didn't start till then, right? So both companies, Lyft and Uber really took off around 2012. And so like, yeah, it actually took a long time. And, and, and Zimride, Lyft had pivots along the way as well. So, but there were just like, you know, you know, the founders Logan and John and, and myself, I mean, we both, you know, those guys stuck it out for much longer than me. And there, there were like, you know, they just believed that carpooling would be a thing and like people would want to use, you know, a share of rides and, and there was a lot of pain along the way to get there. Same with Presto had a lot of pain. There was a lot of times where people were like, why would I have a, a custom tablet on my table to order and pay? And you know, and these, these restaurant chains are very conservative, you know, they're like all about the human experience. So like the idea of putting technology in a restaurant is, you know, anathema to them. So yeah, they're, it's, it's very difficult to stick it out that long, you know, and most people don't do that. Roger, I'm actually also wanna talk about like your, your third company, which is like very different from your other two, right? The, I think founders might be inspired. What, what was the idea behind the third one?
Roger Dickey:
- So third company was tech enabled home renovations. We started with bathrooms. We wanted to make it as easy as pushing, pushing a button and like a room in your house gets remodeled. And there's a lot that goes into that. Like principally three things like design ordering and procuring materials cost effectively, and then getting the construction done. So we just took that flow productized it, built some cool 3D visualizations, some cost calculators, project management flow pricing tools, kind of a marketplace for the contractors to come in and get involved with the customers. You know, communication software, all directions. We form relationships with around a hundred suppliers from Home Depot and Lowe's, all the way down to, you know, individual like tile manufacturers. And then we had about 200 general contractors on the platform. So, you know, the whole thing from design to construction was kind of covered and it was supposed to be a really streamlined experience. Worked well in some cases, didn't work in others, you know, depending on the contractor and the market and the homeowner and what their needs were. But the, the outcome of that was we, we ended up selling the technology to Home Depot and the operating business that did the home renovations went under that did not work. So that was a low point for me. Very, very challenging thing to do operationally. And you know, that was the outcome of that, but at least the technology worked well.
Immad Akhund:
- What was the reason? I guess the overall thing didn't work?
Roger Dickey:
- At a high level. Like you talk to a lot of founders who've raised a lot of VC money and I think they would echo this. We probably raised too much money and scaled too fast. You know, we raised like $3 million and we did like a couple of home renovations with our technology and they all went like really well. And we looked at the numbers and they were very contribution margin positive. So the average home renovation is like $35,000. You know, you subtract our costs, we're making like 5,000 to $10,000 on each home renovation. And we thought, wow, if we just scale this up, it's gonna be a money printing machine. You scale it up and like things start to break and, you know, the contribution margin starts to shift and maybe you don't notice that end of the day, like we scaled too fast. And operationally we were very weak in a lot of ways. And we were, I think too, too, too cost in efficient. And that ended up ultimately kind of killing the operational side. Just like CAC was very high. Our operational costs were very high. We ended up underwater on, on, on, on the margins.
Immad Akhund:
- I mean, one of the reasons we see so much more software only innovation is that it is much harder than anything that touches like atoms. So it is maybe it's like a little sad that like actually like our lives, like the physical lives don't improve as fast as like our software lives do. But it is just so much easier to do software startups.
Raj Suri:
- Yeah, it is. I mean, but you could see like, you know, when, when you can nail the physical one or it's, it's so much more inspiring. I mean we, you know, this week we had the Elon Musk rocket catch, right? It's always inspiring to see like when, when physical startups work, I think I'm excited that there's a lot more physical startups getting funded and you know, deep tech is a thing that, that, that taking off Immad you invest in a lot of these physical like technologies, right? Deep tech stuff.
Immad Akhund:
- Yeah. I mean it's still probably less than 20% of my investments, but yeah, I do, I'm a real sucker for like space investments and weirdly actually they, even the the more crazy ones, I actually get like, pretty good funding and like seem to, you know, there's a lot of stuff happening in space. I was an investor in Inversion, which is like a company that like, you know, there's no way right easy way right now to bring something from space back to earth. So they're doing like reentry vehicles. But yeah, that's like a very kind of difficult idea in that, you know, like actually reentry vehicles are like a very hard problem. There's only a few people that have actually done it, like NASA and SpaceX and a few people. So, but it's fun. I mean it like, it is, as you say, extremely inspiring. Like it's, you know, I get way more excited about them than like the next sales SaaS startup that's gonna like add some efficiency to the sales process.
Roger Dickey:
- Yeah, I, I remember seeing, I remember seeing Relativity Space, you know, go across at YC and it was like, 3D printed rockets for space. I was like, they just took a bunch of buzzwords and glued them together like this, this, this will never work at scale.
Immad Akhund:
- Yeah, they did your matrix.
Roger Dickey:
- They're just like, okay, like 3D printing cross space. Like boom, let's start a company.
Immad Akhund:
- I think they've raised like a billion dollars since then.
Roger Dickey:
- Yeah, no, they've done insanely well. Like I saw that one and I was like, I I I was, I was like, I'm not even gonna talk to 'em like this. This is, this is silly, but to your point, like so many cool ideas are getting funded in space.
Raj Suri:
- So I guess Roger, the other question I had for you is social networks. So what are you interested in in social networks? You know, you've seen the whole kind of rise and fall of social networks, you know, there's a big trend in Web 2.0 and now it's not trendy anymore. What, what do you think is the opportunity there?
Roger Dickey:
- I think there's a, I think there's a couple of things. I think one actually is in the business model, I think advertising, I, I think AI is gonna do a lot for advertising. And I, and I say this, you know, related to social networks. 'cause I don't know that this could be a standalone company, but I, if, if you look at someone like Facebook, they have a really big opportunity given how much they know about you and the, you know, relationships you have with brands, with friends to show you ads that are hyper targeted ads that will make you want to turn off your ad. Like for example, let's say you have a friend who's really into Doritos, you could see an ad with your friend, like holding a bag of Doritos standing on like a dinosaur, like rocketing off into space like, hey, you should eat Doritos. And like, I'm gonna wanna see that ad if it's like Immad like holding a bag of Doritos on a dinosaur, like that's really cool. Like, I'm gonna turn off my ad blocker to see that even if I don't click, it'll gimme a very positive association with Doritos. And you might ask like, why would people opt into that? Well, I would love to opt into being in like badass commercials, doing cool things like being the guy in the Old Spice commercial on a horse, you know, super jacked, like holding Old Spice, like I would opt into that. They can use my face all they want, you know, and celebrities, you know, can just be in any, any number of ads they want. So I think there's an opportunity around content and it could be advertising, it could be other types of content that I think AI brings currently. Like the agents so to speak on social networks have been humans. It's all like the agent to agent communication has all been humans, but I think we're gonna see AI enter the chat, so to speak, in like very interesting ways. I think that's one opportunity. I think every couple of years we see a new social network launch and I think that will continue. I don't think we're we're done on social networks. I think humans change and evolve over time and as we evolve, like we'll want that whatever the next TikTok is. So I think we do need founders continuing to work on that problem. And either we'll see something like TikTok grow into its own as the next big social network, or there'll be another Instagram style acquisition where Facebook will buy you for a billion dollars. But I think there's a lot of alpha there.
Raj Suri:
- I mean TikTok is probably the most successful, maybe the most successful company in the last decade altogether. I mean, apart from maybe Tesla or something. I might have you thought much about this space. Have you made any investments in consumer?
Immad Akhund:
- Very rarely in social, I'm trying to think if I have any, I guess Substack, does Substack count as social? I mean it's media. The interesting, I guess counter argument is it used to be that Facebook was like stealing time from TV or just actually you sitting around, whereas now people are consumed by their phones already. So the only way a new social network arises is if it takes time from another social network, which obviously TikTok managed to do, but that does create like the barrier to entry of like open spaces much smaller. So it is much harder, I think, to do a new social network. But I think there's, there's ideas, I mean like people, weirdly we have so many social networks, but people feel more lonely and there's like less community. So I think there's probably ones that can like bridge that somehow and like, you know, do things in the real world that are still like kinda around social networks. There's a really cool idea, you can probably Google the idea, but it's like they drop like little treasures in like lots of different locations and basically the app is like, you open it and you can do it with your kids or whatever, and you can see like, oh, near, you know, within a mile of you there's like five like drops and then you have to like find them in this. Yeah, but it's like real life things and like different, it's all volunteer based networks. So people are like running, running all the like, and then there's like little puzzles to like try to unlock it and so it's like just for fun. There's like, no thing behind it. But it's...
Raj Suri:
- So the treasures are physical treasures?
Immad Akhund:
- Yeah, the treasures are physical. Like, you'd like take like a silly crown and you'd like hide it somewhere and then there'd be a puzzle to like try to find it. It's like a, it's like a global treasure hunt. I thought that was really cool. But you know, it's, it's actually been around for a little while and it's, it's fairly popular but very like volunteer kind of run. But those kind of things are like, I think probably in the next kind of social network is like that. At least I would hope that.
Roger Dickey:
- Or could be inspired by group chats. I mean you look at like where a lot of people are going and you see, you know, these like subreddits that are hyper vertical social networks or you see people in a bunch of group chats with their friends. Like it might look more verticalized, it might look more compartmentalized. That's maybe what the future of social looks like. I mean a lot of that behavior is happening, but if you can find a way to turn that into something new, I mean that that could be part of the answer too.
Immad Akhund:
- I think Raj has some opinions on that one.
Raj Suri:
- I haven't caught you up Roger on what I'm working on. So that's exactly what Tribe is, is next generation group chat app. So it's one of the companies I'm working on. So yeah, we should chat exactly that thesis. I think group chats are ready for, for, you know, a next generation version. And I think group chats are like the best form of the internet. It's a way to, for people to kind of like be their, be their authentic self versus a lot of the internet is about performing and Immad and I are so many group chats together and I think we, we enjoy that, those conversations that happen and there's a lot of really good conversations happening on group chats. It's really fun to see what's kind of emerged almost as a like antithesis to Twitter and some of these like more public internet, you know, and Reddit, you know, public anonymous, you know, type of conversations, you know, on group chats.
Immad Akhund:
- There's a weird kind of FOMO around group chats as well. You're like, oh my God, what am I, like, you know, all the people listening to this are probably like, oh, what are Raj and Immad's group chats about and how do I get in kind of thing? And that's, I think that's part of the appeal to kind of these group chats that are like, not just family and friends.
Roger Dickey:
- Yeah, a lot of social things started with a little hook like that, you know, like Snapchat was like, oh, like the messages disappear. Like what kind of messages am I gonna get from people if they're disappearing? Like that could be cool. So it's like, yeah, I think to your point, like having that kind of allure could be an interesting product hook or distribution hook.
Immad Akhund:
- Yeah. We should probably wrap actually, if someone does want to speak to Raj and I, we are in a group chat together on Tribe that we'll link to at the bottom that's about this podcast and you can give us some feedback on that. And otherwise subscribe, leave us feedback on all of the normal podcast and video channels. Really awesome having you here, Roger.
Roger Dickey:
- Yeah, thanks guys.
Immad Akhund:
- We should do this again soon.
Share this post