Dan Romero is an entrepreneur and early Coinbase employee who recently left to start Forecaster, a decentralized protocol for Web3 social networks. Recorded on February 1, 2023 for Crypto Packaged Goods Genius Call series.
Dan Romero moved to Silicon Valley in 2013 where he was introduced to Bitcoin and was inspired to read the white paper. This catalyzed his interest in the blockchain and its potential for a new computing paradigm. With Forecaster, Dan is using the Ethereum blockchain to create a decentralized social network where creators and audiences have direct ownership of their relationships. He has designed Forecaster for ease of use for the average user and for developers to have direct access to the underlying data and APIs.
In this episode, you will learn the following:
1. How decentralized, permissionless blockchains can be used as a new computing platform
2. How Forecaster is creating a new decentralized social network that allows creators and developers to have direct relationships with their audience and users
3. How the idea of RSS Plus lead to the creation of Farcaster as Twitter competitor.
Follow Dan at https://twitter.com/dwr
Follow Club CPG at https://twitter.com/CPGCLUB
To learn more about Crypto Packages Goods, visit https://www.cryptopackagedgoods.com/
Welcome back. We are here again on another genius Call for Club CPG. I am a big fan of Forecaster. I'm pleased to introduce Dan Romero. Dan was employee 20 at Coinbase and we have a lot of great info coming from that. He recently left and started Forecaster, which is a decentralized protocol, a sufficiently decentralized protocol for Web Three to do social networks as well as build interesting clients. Welcome to Crypto packaged goods genius call.
Speaker B 00:00:33
Thanks for having me.
Speaker A 00:00:34
Well, I'm sure I left a lot out of your introduction and I think we're going to start with how did you get into Web Three and crypto?
Speaker B 00:00:44
Yeah, so I moved to Silicon Valley in 2013 and I had started working at a SaaS company. I really wanted to work in startups. And one thing I found was when I talked to interesting people, specifically engineers, they were all really into Bitcoin. And I had known Frederom, one of the co founders of Coinbase, in college. And when I was actually originally moving out to Silicon Valley, he kind of pitched me on, hey, come and work at Coinbase. We just raised a series A from Union Square Ventures. And Fred Wilson, who I had read his blog for almost a decade and I think I made the mistake of not actually doing my homework. I hadn't even read the white paper. I just kind of dismissed bitcoin as magic internet money and a Ponzi scheme. And so in kind of immersing myself in Silicon Valley and meeting other smart people and realizing I didn't have a strong opinion on Bitcoin, I finally actually went and read the white paper. And that was a big turning point for me because it kind of unlocked the idea that not only is this potentially useful for finance and payments, but the idea that you could actually use the ledger as a computing platform. And I think this really started to crystallize a few years later when Ethereum was released and you now had something that was actually Turing complete. But even in 2014, when I had just joined coinbase, I remember going to bitcoin 2014 in Amsterdam, and a developer who was into bitcoin at the time, mike hearn, had built something which I still think is one of the most impressive things, considering how difficult it is to build on top of bitcoin from a kind of like smart contract or programmatic standpoint outside of just the core payments. But he had built something called Lighthouse, which was a decentralized version of Kickstarter. And I think that that had captured my imagination and frankly still captures my imagination and that cryptocurrency and these public permissionless. Blockchains is a new computing paradigm and it's a computer that is more expensive and slow for sure, but it enables some other interesting properties. One of which is you can actually write code that if architected correctly, will run forever and can't be changed. Which you can't say the same for some open source code that you release on GitHub and then you have to run on AWS at some point. You do have to pay for that computing bill, whereas with a blockchain like Ethereum, that smart contract that you deploy, you pay upfront and then for as long as the theorem continues to run, that code will execute any time it's called. And so I think that was the thing that got me excited. I think the time at Coinbase and we can talk about it, was a little bit of an unexpected detour in that most of what Coinbase does is financial services, right? It's a technology company at heart, but it is extremely regulated and it is really kind of this on ramp from Fiat to crypto. So in many ways, you're kind of treated like a bank. And so I think that the kind of exciting development on top of Blockchains as a computing platform was never really a focus at Coinbase, at least the five years that I was there. I think that's changed a little bit. Now the company is, I think, becoming a little bit more Blockchain, or kind of like crypto native. But yeah, I think that was what captured my interest. And one of the things that I'm still interested in today, and one of the things that Forecaster is using is using the Ethereum blockchain as a computing platform to enable a new type of social network, which is what Forecaster is. Great.
Speaker A 00:04:15
So let's go into what Forecaster is for our listeners and how I guess we'll get to how you really form that part for our entrepreneur base. But what is forecaster?
Speaker B 00:04:27
Forecaster is a sufficiently decentralized social network, and the simple way to describe it to someone in an elevator is Forecaster is like Twitter, but it's decentralized for the average user. That actually really doesn't matter. Most people are actually just kind of fine with how Twitter works. But I think that the two underlying promises to Forecaster as kind of both a social network and a protocol matter to two different audiences. So the first is, if you have a large audience on a network like Twitter, you actually don't own that audience, you don't own the relationship, or your audience actually doesn't even own the relationship with you. So an example would be, I have 100,000 followers on Twitter. I violate whatever terms of service they decide I violate. They can just remove my account at will. I have no ability to export my followers, let my followers know, hey, I'm on this new network. So the relationship between the publisher or the creator and their audience is owned by a third party in the case of Twitter or Twitter, but you could apply the same to YouTube or TikTok or Instagram. So on Forecaster, the relationship between a creator or publisher and their audience is actually direct. And specifically, if you actually think about it, the audience controls their relationship with the creator or director, I choose to follow you on this network. And I know that if I follow Mikey or DWR, which is my username, I can get access to whatever content I am posting to Forecaster and not have to worry about if there's some third party company that comes in and says, sorry, you're not allowed to communicate between the two of you. The second audience that forecaster is solving for, and this is actually more for the kind of protocol side of things is developers. And for developers, they have direct access to the underlying data and APIs of the forecaster network. The goal is to make it so that any new developer can come in, spin up our version of what we'd call an ethereum node. We call them forecaster hubs. So you can think of it as a kind of dedicated server that will sync all of the content on the Farcaster network at any given point in time. And that should be a pretty easy process. And in doing so, you have the same access to all of the data in the network as the initial founding team or any other big company that would come in. That's the same deal as a theorem, right? If I run into theorem node, and whether I choose to run it myself or through Alchemy, I have access to the ethereum network the same way that a company like Coinbase or Binance or uniswapped us. And I think that is a critical component to building a protocol for social networks. Because if you actually look at the history of the open APIs for social networks, there was an era in the early 2010s when both Facebook and Twitter had very open API platforms. And in the case of Facebook, they invented this term open graph, which we kind of still use today, but it's kind of a shell of what it originally was, where you as a developer could actually publish kind of activity into the Facebook news feed. And as a developer, you could also build kind of applications on top of Facebook. This is kind of Facebook pre going to mobile. And a prominent example is Zenga built a very large business on top of Facebook, only to have Facebook change its strategy in a very short period of time and totally handicapped Zenga's business. And so I think that that was a very prominent example of don't have platform risk as another company. And so I don't think you have nearly as many companies wanting to build on top of APIs. And then similarly, Twitter had a thriving third party ecosystem. In fact, they didn't even have a dedicated mobile app for a few years. They actually ended up acquiring one of the more popular third party apps, Tweedy. And in that period where they didn't have the dedicated first party mobile app, you had this thriving ecosystem tweet bot Twitter effect tweetie, you had a lot of innovation. So a lot of what we take for granted. In terms of Twitter's, core mechanics today were invented by both users and third party developers. Some examples at replies weren't originally native hashtags, weren't originally native retweets, quote, tweets, all things that were invented by users and third party developers. And then eventually Twitter took all these great ideas and incorporated them into the first party client. But in around like 2014, the company had to focus on monetization and really severely limited the size of these third party apps in terms of how many users they could have, and actually pretty timely. In the last two weeks, twitter has essentially cut off all third party apps, and it didn't even give it any communication. So some of these apps have been kind of existing in this weird handicap state for ten years, and then overnight we're just completely cut off. And so with forecaster, the goal is to make that impossible so that any developer you choose to spend the time and effort to build an application, whether that's open source or a business, you have a direct relationship with the users who choose to use your app, and no one else can come in and muck with that. And I think a couple of other examples of ecosystems where this does exist one, browsers. You get to choose what browser you use, at least on desktop. Mobile a little bit more tricky because Apple locks things down browsers and email clients, right? So if you own your own domain, you can move between Outlook or Gmail, and no one can actually kind of hold you hostage. And so that's the goal. You have a relationship with your audience as a creator or publisher, and as a developer, you have a relationship directly with your users and access to the data and APIs.
Speaker A 00:09:53
So much to unpack there. For me, personally, everything that you've said resonates with early days of Facebook and Twitter. Farmville for sure was something that you probably had to go into your feed and suppress if you had one of those friends or relatives who was constantly on there asking you for it. But as an entrepreneur, a big risk that you've highlighted is if you're behind somebody else's API, your business is at their whim. And I think it's both an amazing innovation and a bold statement to say these are the APIs that are going to be in perpetuity for you to build a business on. And I think for creators, as more and more folks get the idea that you've presented, more and more will be asking for it to own that relationship with their audience, specifically the folks that are now creating content that they now have to go multiplatform just to monetize and then be at the whim of it. I think Web Three and the ownership model that you put forward is you're owning your audience. And that's a huge piece that has not really been available to us and I think is a beautiful innovation so let's talk about how did you get to the ideation phase, or through the ideation phase of this. You were at Coinbase for five years. How long has it taken you to get this level of clarity and where did you start with it? Or was it pretty well formed?
Speaker B 00:11:25
I think it was an evolution over a decade plus. I think my first love on the internet, kind of when I was in high school, was RSS and blogs and I'm an info vor. And so the idea that you could subscribe to all of these different independent sources of information, whether they were professionalized blogs or just one individual sharing their thoughts, has always been something that I've been interested in. And I was a power user of Google Reader until they shut it down. And I still use RSS today they use a service called Feedbin. But I think one interesting thing that happened during the kind of like two thousand and ten s was RSS, which for a while kind of felt competitive with Twitter, very much lost to Twitter, right? The kind of mobile oriented nature of the user experience. And feed of Twitter just outcompeted RSS. We could talk about kind of the shortcomings, but I think fundamentally it comes down to consumer UX, both in that a native app with a feed is a lot better than a feed reader, which feels a little clunky and slow. And then the second thing is discovery, at least traditionally in RSS apps, you actually found the website and then added it into your feed rater versus on Twitter, if someone retweets something or you're clicking around, you can stumble across new profiles pretty easy. And then if you like the content, it's one tap and it's there. So some of that kind of ease of use I think people tend to underrate, especially technical people, because if you're used to using an RSS reader, you actually like that you have full control over it. But the average user, they want convenience. And that's actually something that I learned at Coinbase, is I have this concept I talk a lot about on Forecaster as stated versus revealed preference. And so in crypto you have a lot of stated preference of the ideology of decentralization own, your own kind of keys, all this stuff. I think that's critical to making the system work, right? Crypto wouldn't be nearly as interesting if you couldn't actually self custody. But the reality is that the reveal preference of most users is they want password reset and they don't want to have to think about a mnemonic. And so I think with that in mind, when I kind of started thinking about what I wanted to work on next after leaving Coinbase, the idea that had been rattling around in my head for a while was how could you make RSS a permissionless decentralized protocol, more competitive with Twitter? And so some of the things that I mentioned could you actually improve RSS or kind of build a layer on top of RSS to actually make it competitive with Twitter. The original name for the idea was RSS Plus. And then over the course of the last two years, it's kind of morphed into what Forecaster is today. And so Varun and I have spent a lot of time thinking about kind of what are the minimum amount of things that you need to add to something like RSS to actually make it so that it as a protocol can be competitive with Twitter. And it goes back to the kind of two tenets of the relationship between the consumer and the the creator and making that user experience really, really slick. And then the developer experience of how can you actually make the protocol approachable enough for a developer who's busy and has a bunch of other things competing for their attention and time, want to hack on building something new on Forecaster? And so thinking about user experience and developer experience is kind of the two core tenets of what we've tried to do over the last couple of years and are going to continue to do for the next few.
Speaker A 00:14:46
If you go to the Forecaster site now, Forecaster XYZ, you're presented with a very wide array. It's actually three very simple buttons, but when you dig into them, you really can see the thought with which the thoughtfulness which you and Varun have put together, forecaster, it's apparent that you've done a lot of research and thoughtful product. I think for me, and I think for folks that are trying to get into Forecaster, what is the first step for them? And I'll break it down into like, I want to understand Forecaster better, or I'm a developer and I want to go play on it. Where would they hit it?
Speaker B 00:15:30
Yeah. So I think the full caveat on the website is it's actually intentionally sparse and bare bones because I think we're still very much at an early stage. And I have a strong belief that consumers, you get like kind of one initial impression, and in some ways maybe that's hypocritical and that the initial impression right now is, what do I do? How do I sign up? But I do think there's a version of whenever you get a chance to install an app, that first five minute experience sets the stage for whether or not you're going to want to check it out again or you kind of say, hey, this isn't for me. And I think we just have a lot of room to improve both at the kind of protocol level, which the user doesn't experience, that they experience it indirectly and then more importantly, at the initial client level. And so we're kind of in this open beta in the sense that it's easy enough you DM me on Twitter to get an invite, but the friction there keeps the community relatively small, and it's allowing us to kind of like iron out all these new user experience details and just kind of stuff that I think are cable stakes in order to actually make it so that when we are ready to kind of open the floodgates in terms of anyone can go sign up. The initial experience is on par with something that is kind of like a mature web two company. But in terms of figuring out like understanding the protocol, I think about it in three ways. The first is how does this whole thing work? And at a simple level Farcaster is a hybrid of on chain and off chain. And so what we actually put on chain and in this case it's a theorem. Today we're on testnet, eventually it will be on main net. It's just a very simple identity primitive that it solves for a couple of things. So the first is if you sign up for forecaster, your identity is associated with a public key which is an ethereum address and you actually then become the owner of that kind of identity that no individual company organization could come in and change that. So you have sovereign ownership of that on chain kind of pointer which just basically says this user maps to this public key. And what's important about that is it allows you to be able to switch which clients you're using in the forecaster ecosystem and still retain the exact same identity. And then anyone who's trying to follow you this idea of the direct relationship between you and your audience can always find where you are. And so by doing that on chain that has some costs, it's going to have some gas. We only do that upfront during onboarding afterwards. Every other thing that you're doing on the network for the most part is actually off chain. And we have this concept called Farcaster hubs, which I think I mentioned before. It's our version of an ethereum node for the forecaster network. So it's kind of like a node for the forecaster network but we call them hubs and by running a hub you have full access to the underlying data and APIs. So every we call them CAS. But you can think of that as just like a tweet. So cast the likes the profiles, all of that information is permissionlessly available to any developer. And as a user you don't have to think about any of this because we've kind of constructed in a way that the experience that you use as a user is you just use an app, right? So in our case we have a kind of default app. We're probably going to rename that just to actually make it clear that forecaster is a protocol. Right now it's forecaster and forecaster but I think you could imagine something like email as a protocol and Gmail everyone realizes is a separate app than the protocol. But to imagine us renaming this kind of default app as something that is distinct and then there are going to be a bunch of other apps, right? Just like you can use Hotmail or Gmail or Yahoo or in the case of browsers, you can use Chrome, Safari, Internet Explorer, I guess they call it Edge now. But the idea of that is you get to choose which client you use and then from there you just use it like you use a normal app and all of the magic of distribution and kind of like how the content gets routed to everyone. That's what's happening actually at the hub level. And then the kind of authenticity of any post is actually mapped to what's on chain. And so you kind of have this three part system of on chain, minimal amount, but a critical important part of the ecosystem. The kind of off chain where all the data lives and kind of everything gets routed. And then as a user, you just use a client. And I think it's actually in some ways similar to email. Whereas if you think about email, everyone has an email address that's actually tied to a domain DNS. So that kind of lives at one layer. Then you have this layer in the middle that most people don't know about SMTP. That actually is how the mail servers exchange messages. When I send an email to you at Club CPG, you're getting the email sent to your server. And that just magically happens with the mail servers and SMTP. But the reality is we just use clients. So you might use G Suite, I might use Outlook, and we don't have to think about like what version of email you're using. It just kind of magically.
Speaker A 00:20:36
One of the things that strikes me about how you're building and constructing this is that this is not in the typical Silicon Valley sense, a hyper growth type company. I think I've been watching and following along with your updates and graphs and you are definitely achieving the scale, but I think you're really pacing yourself intentionally. What is the inspiration for that and how are you thinking about forecaster in 6710 years from now?
Speaker B 00:21:11
Yes, so the inspiration I think is two part. So the first is the most successful social network in terms of daily active users today still remains Facebook, despite it having actually been on a decline in the US. And such. Globally, I think it's something close to 3 billion. I forget, it's either Dow or Mao, but monthly active or daily active users. But Facebook I think a lot of people forget was invite gated. You had to have a.edu email address, but you just can't go sign up for it. You have to be enrolled at basically an American higher education institution. But it was.edu gated for over two years and we're coming up on two years of forecaster. So we're in the same realm of we want to kind of take this deliberate approach early on to figure out a all the kinks. But two, seeding the initial community with as high quality of a group of people as possible. And that's not meant to be exclusive. I actually want to be as inclusive as possible because I think there are a lot of great people out there. But I do think that there are a couple of risks. And this is the second point that we are trying to control for. So if you just have open sign ups, one, you're just getting a lot of spam and squatting and a lot of these other problems that we're going to have to solve for long term. But by keeping it gated, you can actually delay when you have to solve that problem because you have a limited amount of resources and you can only solve so many problems at any given point. And I think the second, which I think is probably more important is, and this is actually something I learned at Coinbase and in any hyper growth company, you can actually see this with the employee base. So you have a company, you're adding x percentage of employees, it's only a few people every month, that's fine the percentage of overall people relative to the size of the company, you're giving them a chance to kind of absorb the culture and norms of the company. And then you hit something like Hypergrowth, which happened to Coinbase in 2017, 2018, and you significantly increase the hiring rate because you just have all these things that you're trying to solve for. And now all of a sudden, at any given point for any 60 day period, 90 day period, 30% of the company is brand new. And when you have that large of a percentage of the company that hasn't really been at the company, what ends up happening is the culture starts to change. Because it's not malicious. It's just people are bringing the culture from whatever company they were at before, and they haven't had time to kind of throw osmosis. And just being around what the culture is, you now have this kind of morphing culture that's a little bit of a Frankenstein of every other culture that the companies are coming from in addition to what the core foundational culture that actually got the company to the hyper growth stage. And I think the same thing applies to a social network, especially a public social network, where the interactions between any given person are very easy. It's a little different on LinkedIn or Facebook where you're actually friending people or messaging apps, so it tends to be a lot more small communities all tied together. Whereas on a public social network like Forecast or Twitter, there's a lot of interactions all across the spectrum. And so giving the network chance to absorb new people coming in is a belief that we think is going to help scale quality on the network. And by scaling quality, we have the best shot that at some point when we do have permissionless, open sign ups, you are going to have more people retain because the network at its core is actually much higher quality. And so that's what we're trying. And then the last point there also just something I mentioned before is consumers kind of give you one shot. And so while we're in this early stage, you just want to be deliberate about people who are actually very interested in using something that is kind of in a beta and rough around the edges. We're trying as much as we can to polish things, but if you just kind of have it open up and you get a normal person who has kind of no patience and is comparing you to a web two app, the likelihood that they're going to churn and not come back is much harder or much more likely. And it's much harder to convince those people to come back versus the first time someone is giving your app a shot for those first five minutes.
Speaker A 00:25:10
Great insight. I think I picked up Twitter and then I put it down for a while and I eventually came back for probably a variety of reasons. One of the things that you strive for with Forecaster is to be a public, intellectual social network. And it seems to be tied to that deliberate growth that you're going for and making sure that people actually want to be there, expound upon kind of what does it mean to have Twitter kind of doing its thing and Forecaster being this public, intellectual social network.
Speaker B 00:25:44
Yeah, look, I think the hard thing is that Twitter is at scale and people who are kind of the what I would say generally well regarded public intellectuals tend to have a lot of followers on Twitter and even people who are interested in decentralization. The reality is they probably have several hundred thousand followers, like the more famous ones. And so it's very difficult for them to give up that audience despite Twitter being centralized. And day in theory want to use something that's decentralized. But bridging that gap of audience size I think is challenging. And so it's actually been a mistake in that when I first started onboarding people to the Forecaster beta almost two years ago, I went through kind of like my call it Internet friends who a lot of have our into crypto decentralization but also have large audiences. And they went through the onboarding as a kind of favor to me and said, oh, this is kind of neat. And then they didn't come back. And so I think the lesson there was actually Eugene Way has a great blog post called Status as a Service. Highly recommend anyone who's interested in social networks to read it. It's pretty long, but I think I've become convinced that this post is quite accurate in that what you actually want to focus on in the early days of a social network is you want to find people who. Don't already have a ton of social capital on the preexisting social network. And immigration is kind of like a rough analogy for this where if you're doing really well in the kind of like mother country, you're very not likely to go immigrate to the kind of frontier country. Whereas if things aren't that great for you there and there's potentially a new land of opportunity where land is cheaper or there are other opportunities, you're going to be more likely to want to go emigrate to that new country. And so for us, what ended up happening was that the group of people on Forecaster who have the largest audiences and have built kind of like new followings basically are starting from scratch. They didn't really have much of an audience on Twitter to begin with and so as a result they're building a new kind of social graph and new social capital on this network which Eugene Way talks about in this post where he calls it actually proof of work. And he says that you have this opportunity to put the social liquidity right so you can kind of think of a social network like DFI, like TBL for DFI, total value locked in social. It's actually just the number of posts that are happening there of some amount of quality. And so these people come in, they start using the network, people find the content interesting, they start following them, which creates this reinforcing cycle where I don't have any specific magic beam that you can go do to grow a social network outside of. We were just persistent for the first year, year and a half in trying to find a group of people that it resonated with despite it being rough around the edges. We had a bunch of crypto native features which I actually think appealed to a certain type of person. But basically last summer we hit a point where usage significantly increased on the network because we actually had enough social liquidity just by people using it every day, where now when you opened up the forecaster app, you actually probably saw new content almost every time you open it up, which I think that threshold takes a while, but we've reached it. Now. I think that the challenge is if you think in orders of magnitude. We're coming in on roughly 10,000 people on Forecaster in the beta, so it's still pretty small, highly engaged user base. But going the next ten X I think is going to be challenging for us. And so it's probably going to require us to shift our strategy a little bit and figure out how to broaden. Maybe some of the people who are interested in this are maybe going to be a little less crypto. Native and figuring out how do you kind of make it more general purpose that's, like, on the docket for 2023 is how do we go from the kind of 10,000 people in the network to 100,000 people in the network? And then the next year is probably 100,000 to a million? And then from there and maybe at some point you hit some escape velocity where it grows. But I don't think it's worth banking on that. I think in some ways that's lazy thinking of just like, oh, I'm going to hope the app goes viral because I actually think we've seen some apps go viral in the last few years in social. And what ends up happening is, one, you get a lot of churn because there's this initial excitement and then you hit hit a point where, okay, you know, there are still a lot of rough edges, so people turn and or you have big competitors. Just copy that one of the core features that's 80% as good and it actually really dampens the growth.
Speaker A 00:30:10
Does anybody have a Clubhouse invite?
Speaker B 00:30:12
Yeah, Clubhouse. I'm an investor or a huge fan. I actually wouldn't count them out, especially that it seems like I'm not counting.
Speaker A 00:30:18
It out, but it's exactly what you had described happening is like everybody Twitter has spaces, right? Just came right along.
Speaker B 00:30:27
Right. And it seems like Twitter is under investing in spaces now. So that's an opportunity for Clubhouse. But it's just going to be more of a grind to grow. And so our point of view is don't attract as much kind of hype and attention for now, really try to nail the core product experience and the developer experience of the protocol while holding ourselves to some amount of consistent weekly growth. Because it's not a science experiment. We are actually trying to make something that achieves internet scale. And so I think the best way to do that is you kind of focus on quality, but you have to kind of balance that with forcing yourself to grow and specifically grow with retention, right? So it's really easy to potentially just keep ramping up the invites and then by definition, you can kind of always be moving your daily active user number up. But what you actually want is, can you actually consistently retain 2030, 40% of users who sign up after some amount of time, a month or two months? And then you're kind of starting to build like a cake where each layer of new people coming in is adding more toward the social liquidity and vibrancy of the network. And so I think we're making progress on that, but we have not solved that by any means. And so it's actually one of the reasons we continue to kind of be in this invite only mode is work out those kinks really get to a place where we know that if for every 1000 people that sign up, we're going to retain 400 and hopefully that number gets higher and higher. And at a certain point, if it does happen like that, then, you know, you can actually just have open sign ups and you have a really sticky social network. And I think Facebook is by far the best at this in terms of their traditional get you to sign up, really try to push getting your email contact book uploaded, which, by the way, also helps for people in the future sign up, because they could start to map who you're friends with based on who's existing on the network. And then Facebook, I think it was, get ten friends, and if you got ten friends, you became kind of a retain user on average. And so I think we're still trying to figure that out for Forecaster in our client. Now they're going to be other clients, but I think it will benefit the network if we can have a point of view on, okay, this is the ideal profile for a person to sign up for forecaster. And if you can actually get them to do X and Y, then they're most likely to continue to retain. And it's not to get addicted, but it's to actually hit a level of utility for the individual that signs up that they actually want to come back. Because if you think about social, it's a zero sum game in that it's all about time. If you look at Facebook's metric, it's time spent. And so to get a user to, for example, install the forecaster app is one thing, but then to have them put the app on their home screen, that's another threshold very hard to measure. You don't get like a report from Apple of the percentage of users that have it on their home screen. You can approximate that by how much time they spend. But that's actually what you're trying to drive towards is having an app on the home screen, which is a validation that the end user views the app as some amount of utility. That I want to actually check this on a regular basis with the caveat that I actually think time spent is a very web two ad centric metric because it's essentially correlated with if you're a company like Facebook or Twitter. Twitter pre subscriptions is very focused on increasing time spent, because time spent means that you're going to see more ads. And if you see more ads, there's more revenue. And in a case like Forecaster, what we're actually trying to do with our client is we're going to actually try to do subscriptions. We just personally prefer that. And our view is if you can build an app that 80% of users are using for free, but maybe 20% of users are paying for, spotify is like a good example of this. I actually think their subscriber rate is even higher. And if you can get to that place, what you can do is you can shift the incentives from a product standpoint. You no longer have users, they become customers because they're paying you. And then the other thing I think that's important is you can actually start to prioritize features that might decrease time spent but increase customer satisfaction. So we don't want you to endlessly doom scrolling the app because we're not showing you ads. What we actually want to do is use machine learning, for example, to show you the ten most relevant tasks and then maybe even say, hey, you're caught up. We don't think that based on your preference there's going to be anything else that's particularly interesting to you. That's a hypothetical at this point. But the point is you can do that because the user is actually paying you to say, give me signal rather than give me kind of like a way to waste time. And I think that's like an aspirational thing that we'd like to achieve with our client, but I fully expect other Forecaster clients. The beauty of a permissionless protocol is some people may build apps that are completely free and use ads and that's been the revealed preference for a lot of consumers in web too. But I think we'd like to start the experiment that there maybe is an HBO version to the network television or a Netflix version to the network television of a social network. And so that's what we're going to try.
Speaker A 00:35:23
Wow. The builder and the entrepreneur in me is just racing through with how much benefit getting into Forecaster and really leaning into the network there can give. I actually have a funny sort of aside and it's more of not necessarily complimenting you on it, but thinking about I have a Blob and I only got a Blob and downloaded that app because on Forecaster you are like, all right, I'll give Blobs a chance. And it's the beginning of the year and it's pretty much a steps app where you get a little 3D avatar to track your steps and feed it. And it raises in me this idea that where Forecaster is right now is a lot of very committed people to the cause and they're also very giving of like what they're trying to do by integrating Forecaster. And so my Blob is named okanfunk, it's an interesting concept, but for our folks that are listening that in our accelerator program and that are building Web three products, everything that you've said to this point about growth should resonate very strongly with them. And I think on top of that, again, the cake that you're describing, forecaster becomes a very enticing protocol to understand and build against and get involved with. I've been impressed with so many different experiments that have come through and it very much feels like early Twitter in that sense of people are genuine. You've done a fantastic job of not having the bots come through, which I think is kind of a downside and the attention economy or the engagement farming, because I'm not pro or against engagement farming. I think it's something that's useful at times and you can take it or leave it. I think all of it feels, in a sense on this bleeding edge of this frontier like you've described. Dan has been gracious enough to give us, like, a secret word for you to join Forecaster. So if you want to describe a little bit about how that process would work, that'd be great. And then we're approaching we have a bit of time. I have some other questions teed up for you, but how can folks that are in our builder community get involved in Forecaster?
Speaker B 00:37:44
Yeah, so the current way to get on Farcaster is to DM me on Twitter. And the reason for that is that actually, it becomes a really great filter. Right. The network right now feels a lot like Twitter, so if you don't really have much of an active profile on Twitter, maybe less of a great candidate for it. That said, I think in targeted instances like a builder community, maybe this is actually a chance, if you haven't been into Twitter, to try something like it, because the network is a lot smaller, more intimate, people who are more like minded. That said, I think that the other thing is it gives me a way of contacting people, especially if they start using the app. And all this data is public because it's a protocol. Say you show up, you start casting, and then you stop. I can just reach out to you a little later and ask, hey, what worked for you for that first few days that you were using it? And then it didn't. And that's been a source of amazing product improvement feedback. So I wanted to keep this as long as possible. They were coming to the end of this onboarding process, but to use like a Paul Graham do things that don't scale. It's taken a lot of time for me personally, but I do think it's actually given me a really good feel for who's joining the network, just by virtue of I can look at people's Twitter profiles and have a way of contacting users who are no longer using the app. Right. I think a lot of apps would just kind of use some email drip campaign that says, hey, it looks like you didn't use the app anymore. Why didn't you use it? Most people ignore that. Whereas if I've taken the time to onboard you via DM, if I send you that one question, I tend to get a high hit rate if someone will just respond back with a short message. And so that's been really useful. The word for you all is Ohio. I use US states, so we haven't gone through all of those. I'm not from Ohio. I just picked it on the map today, and it's easy to spell. So if you put Twitter DM to me, I'm DWR on Twitter and you mentioned the podcast and just put Ohio, it'll kind of pop to the top of my inbox, and I'll make sure to send you an invite.
Speaker A 00:39:45
Fantastic. What you just mentioned there is so thoughtful in terms of how to grow and get product feedback simultaneously, and you mentioned it. I'm sure you have a lot happening through your inbox. I want to shift our attention more to this experience as an entrepreneur and the desires that you have. And one of the AMAs this morning on Forecaster, I think it was really poignant in. Somebody asked you, like, do you ever have any doubt? And I thought your answer was really poignant. And for folks that are here listening, that are in our accelerator program, if you could share, I'm going to guess that you remember what you wrote. Do you have any fears or doubts and what are you thinking about Forecaster? Constantly.
Speaker B 00:40:43
Yeah. I mean, I have 100 of them a day. Is this the right strategy? Are we growing fast enough? Should we be hiring more people? Should we be hiring fewer people? Should we be focused on the protocol? Should we be focused on the product? I'm not even working on this. Anyone who claims that they have this kind of perfect devotion to their idea and are unwavering, I think is lying because I actually think in some ways, the paranoia there's the famous book Only the Paranoid Survived by Andy Grove, who's very famous in Silicon Valley from his time at intel. And if you kind of have that entrepreneur mindset and let's be clear, not a successful entrepreneur yet. I worked at Coinbase. It was part of an early team that helped to grow Coinbase. So that was a big win. But for me as an entrepreneur, I haven't built a significant revenue generating business. So working on figuring that out, but I think is in kind of talking to other successful entrepreneurs and actually also seeing the experience at Coinbase through, you know, how Brian approached it and and Fred, I think the paranoia of always kind of being willing to look for truth seeking because effectively success in a start up. I'd say the most part is you found truth in that the market is rewarding you for building something valuable or something that people want. PG paul Graham, from my combinator, has all these trite phrases, but the more I've spent time building as an entrepreneur, the more I realize the reason he keeps saying these things that sound really trite is that they actually work. So I do things that don't scale. This idea of a founder schlep, it is not glamorous. There's just a bunch of stuff that you have to go do and that I think contributes to the doubt of, like, why am I doing this? Right? It's like, why am I setting up something in Rippling? And it's like, okay, well, that's just a part of what I need to do. And it's not all kind of like strategy in perfect product thinking. And then you have product failures, right? We've launched things that haven't worked as well as features that haven't hit our quality bar. And that's really humbling. And so going all the way back to that, I think doubt is inevitable. It's just how you have your relationship with it. And I think one thing I was fortunate in my time at Coinbase is I had plenty of doubt when I was at Coinbase when I wasn't even a founder. But I think seeing it work out at Coinbase, the biggest lesson I got from that. And yeah, you could say, well, maybe you got lucky, but I think it helps really crystallize in your head. Persistence is the thing that is super underrated in that it's so easy to get kind of off course from whatever mission you had and or discouraged, right? Like the period of 2014 to 2017, basically there was basically no growth. And there are a lot of people who left at Coinbase, I almost left myself. And there's something about persistence where you kind of stick it out because you fundamentally believe in what you're working on. And it might not always work out like that. It's not a guarantee just because you're persistent, it will work. But I think the thing that I strongly believe is that just sticking out longer than most people, assuming you're making some good decisions and constantly improving, much easier said than done. That is the kind of like whenever I feel doubt, it's just kind of like, okay, well, what can I do right now to make the doubt go down a little bit? And that's usually just trying to make forward progress, right? Whether that's recruiting someone or thinking through something on product. And in the case of forecaster, it's really great. I can just use the product. So sometimes being able to just use this public social network in a kind of irreverent or casual way helps me get a better feel for a as a user, what do I want out of this product? And then also just kind of seeing what other people respond to and or the fact that people are constantly giving me product feedback. So there is a little bit of a unique aspect in that the product itself with forecaster is inherently very social, right? So you remove some of that loneliness. But the other thing I would say is having a kind of co founder that you respect and a co founder that when you're down, they can be up and give that pep talk or a partner, right? Like, maybe you're a solo founder and you have another person in life who, when you're kind of feeling down on a day, they can remind you of the reasons that you originally pursued what you're working on. And that's not to be Pollyannah or just kind of like saccharine in the sense of just giving you, you're going to do this, you're going to make it. But it's actually more about just you're going to have an emotional roller coaster working on anything, whether you're an entrepreneur, an artist, or anyone creating something. And putting themselves out there publicly. There's a vulnerability that you're inevitably going to have to have and having mechanisms by which you can actually to kind of modulate the really extreme ups and the really extreme downs. I actually think that's one of Brian Armstrong's strongest things as an entrepreneur quality as an entrepreneur in the kind of like worse situations when everyone else is kind of panicking, he comes across as extremely calm and level headed. And it's something that I actually am probably more of an emotional person and I've kind of worked to try to make that a little bit more of how I approach working persona. And it takes work. But I do think over time, the highs that don't hit the highs and you don't hit the lows as bad. And I actually think that is the best way to continue to make forward progress. Sorry, a bit of a rant, but I don't know that's how I'm approaching it. I'm not successful yet with that, so we'll see.
Speaker A 00:46:17
I think it's interesting how you framed your success as an entrepreneur to this point. I am a maxi, a stand for Dan Romero, and I think the idea that forecaster well, you and your co founder, that it will continue. I'm one of the faithful and also I think it's really important to hear from another founder. Yeah, there's a lot of tactics that you have to implement on a daily as a founder to survive and that grit and determination is a key important attribute that you might not have at the beginning, but you have to build it. I most definitely can empathize with those feelings of doubt and just trying to make your way through the world, especially the past couple of years.
Speaker B 00:47:10
I was just going to say, I think that's a great point. It's a muscle and I think where you can develop that is one obviously the best way to do that is you become a founder and then actually kind of are forced to do it. But if you're not quite ready for that yet I think the other thing is sometimes if you're working at a company that is growing, so you try to identify a company that you respect the founder based on their public statements or how they're building the product, and you can actually learn a lot from there. It's not quite the same, but I think you can develop that muscle, especially being in that early is kind of seeing what it's like. And so then when you are actually starting your own thing, it's still a new experience because the buck stops with you now as the founder. But you have a little bit of more of a framework of, okay, the persistence and grit if actually applied and you're making some right decisions, you can actually make progress despite hardship or setbacks and things like that.
Speaker A 00:48:05
It's an interesting time in the valley where there's a ton of layoffs. I actually used to work at Meta. My old team got hit very hard in the last set of layoffs. I think one of the things that we're seeing in the CPG community is folks are trying to move into Web Three more full time. And you've had this really great approach as to your mindset at Coinbase in the beginning as an early employee. Can you share that a little bit in terms of just that mindset? So you're not a founder there, but you are very early and you're wearing many, many hats. What does that mindset look like for the folks that are trying to make their way in the world and join a new Web Three company or just a new startup?
Speaker B 00:48:59
Yeah, so I don't know if my opinion is controversial, but I think there's this meme sometimes that it's kind of like founders versus employees and maybe that's the case at a much larger company. Right. And it's more like founders and Execs plus or management versus employees. But at an early stage startup, I think that the most successful strategy is, one, trying to do as good of a job as you can on picking and that's fundamentally picking the right kind of market, team and product. So it's like Marcus and has like three things to consider. And he would argue that the only thing that matters in most cases, which if you actually kind of break it down, it's probably right, is if you pick the right market. And there's obviously Web Three as a giant market, but like to think about the specific the product is trying to sell to X person, that's essentially the market. And so thinking through who's going to use the product, whether it's a business or an individual, I think that's where I'd spend most of my time. If I'm evaluating a company or as an entrepreneur of what I'm trying to offer is really understand who you're targeting and do as much research and validation. So whether you're starting a company or you're joining, really having a good handle on that. And then I think the next thing is team and then product. You can actually have a presub par product in the scheme of things, but if you hit the right market and you have an excellent team, from an execution standpoint, you're going to have a lot of success. And I would actually say that Coinbase, there are certain aspects of the product that were clearly sufficient in the sense that we needed a way to move Fiat to crypto. There are a whole bunch of issues with Coinbase's product and very public issues of the site going down and all this stuff. And people still continue to use it. Right. It could continue to grow because it was in the right market with the kind of right team approaching on an execution standpoint. And so I'd say that picking is really, really important. Again, as an entrepreneur or as someone joining an early stage startup. But for the employee joining an early stage startup, you really, really have to kind of like put the company first and realize that what the most valuable thing you can be doing is solving the hardest problems for the company, assuming you can kind of work on those things. And as a result, every time you solve one of those hard problems, the company makes more progress and gets closer to hitting some kind of escape velocity or significant amount of growth. That when that happens, most likely you are not going to be able to scale. It's just, it's just the reality of human development as fast as an exponential growth curve, right? So if you don't have that and it's more linear, maybe you actually over five years get to kind of some level. But in the case of Coinbase, I was, I was VP of Ops in 2017. So it's kind of like a COO type role. And within six months it was so clear that the role required was so far outside of where I was going to be able to get to in a reasonable amount of time that both myself and Brian we got together and we said we need to actually go to COO search. And I think it's really difficult for early employees to be layered or kind of like realize that the company is growing faster than their capabilities. But if you can internalize that is that that's actually a very normal situation. In the case of where you're actually starting to have a lot of growth and you treat it as a team player, I think the benefits in terms of, one, the trust that you're going to have with the founder, because they're like, wow, this is a person that can actually put the team and the company ahead of their own personal that's going to be tremendous. You're probably going to have a very good reputation with people on your board which from a network standpoint starts to actually really improve your future prospects, whether that is continuing at the company, leaving the company to start another company, or getting a job at some other company based on your experience there. And so I think people way under rate their reputation as a team player because it's a natural, human thing of like, wait, I worked really hard, and now we're bringing in someone who didn't have to take as much risk and tends to be a little bit more senior in their career. And they're just coming in and getting to do this, and it feels really unfair. But I think it's reframing that to realize that, no, this is the launch pad to your career, you got lucky and that you hit a company with a lot of growth. This basically gives you a ton of optionality. And why would you want to go solely your reputation assuming you want to continue to play the Silicon Valley game? Right? The Silicon Valley game. Is very much about network and whether that is, again, starting a company, the next company you're working at, you get into angel investing, that's all. So kind of optimizing for your reputation as a team player and someone who is extremely helpful, I think is dramatically underrated in this world where people are like, what am I getting out of it, right? And so I think people way over optimize on titles at early stage companies. They optimize on things that I say are a little bit more superficial versus the number one thing I look for when I interview someone is walk me through the hard problems you've had to solve. And it's so clear that people who actually have solved the problems and gotten their hands dirty versus the people who were really good on the political side of getting the right title or kind of having the right scope of role, but basically didn't have to solve the hard problems themselves. Startups is just an endless stream of hard problems. So obviously, whether you're a founder or an early employee, that is the skill set that you really want to optimize for.
Speaker A 00:54:22
We are just about at time. I wish that we had more time because there is a lot to go dive into. At Metafacebook, the popular blog post internally was how to Give Away Your Legos, which is, I think, really for my simple brain that has a joyful toy type reference. It's impressive, everything that you've done and the way that you're articulating it. I'm excited for the future of forecaster. I'll leave you the last minute to cover anything that you want to cover that we have not. And yeah, thank you again for coming on. And shout out to both Janet and Chad for their work behind the scenes and for your team. And also TZ martin, who's the guy that has dragged me to a bunch of Forecaster meetups in SF.
Speaker B 00:55:09
Well, thanks for having me. I really enjoyed this conversation. The thing I'd leave everyone with is something I mentioned before is, I think, one of the best cures for a down day. As an entrepreneur, early stage employee, if you're trying to figure out what to work on is just try to talk to customers or users and you'd be surprised how willing a lot of people are. Like to give you 15 minutes to hop on a zoom, call and talk through something that you're thinking about. And I think that there's this kind of myth of the Thomas Edison light bulb in a lab. That's how entrepreneurship works. And I think that the more time I've spent is every time I spend time talking to users or in the case of forecaster developers on the protocol. The amount of insight that I get just from that conversation is a it's a super big mood booster and it reaffirms why you're working on what you're working on. And it is incredibly useful in terms of especially if you stack up a few of those calls in one day, it can really get your creative juices flowing in terms of solving next hard problem for whatever you're trying to do. So that would be my one piece of advice.