The Social App Thesis

Why every winning onchain app will be social

It’s been a minute, or hundreds of thousands of minutes, since my last piece here, but I’ve published a couple pieces in that time on consumer apps I’d love to work with potential founders to build: namely, Hackathons-as-a-Sport, and Reality TV Shows. The reason I haven’t been writing is because I’ve been working to build a tool to enable these, called Joke; Bankless very graciously gave me 25 minutes to describe it here, and my dms remain open for anyone who would like to work together.

I am shilling, I know, I know, but I want to stress that the piece below isn’t just some supercilious manifesto about how apps need to behave, but an attempt to frame for myself *what I need to be doing* right now as a builder. Last fall, I published The Proto-App Thesis, where I argued that every app would increasingly operate like a protocol. This piece, The Social App Thesis, is its sequel—an attempt to argue that every app will *win* by operating as a social app.

Obviously this is a tricky double burden for your generic DeFi app to bear, this task of being both a protocol and a social app. But it’s also, I think, what gives apps so much opportunity over what’s come before.

Enough prefacing. Let me try to explain.

***

I.

Once you see it, you can’t unsee it. The influencer living off Prada gift bags in a rat-infested studio in the Lower East Side. The musician from the streets whose beat stops hitting the moment he becomes an over-orchestrated superstar. The rich husband bursting out of his shrunken, wrinkled button-down next to his zealously coiffured, haute couture wife. It’s everywhere.

What I mean is the inverse correlation between financial capital and social capital—between our contemporary merchant class (the financiers) and religious class (the taste-making kulturati).

This is a bit of a taboo subject, I think, in a world where capitalism has trained its adherents and detractors alike to believe that money can buy anything. Instead we find that to be rich not only means gaining one type of cultural power in political influence, for sure, but losing another type of cultural power in the blindness of privilege. The cost of controlling society is to become, well, something of a social loser within its norms.

If you are one of those poor people afflicted by billions of dollars of life savings, I know you may be fretting when you hear this. Please don’t. In theory, you still have the three classic ways of alchemizing financial capital into social capital. You can build a relationship with someone cool (by marrying), you can invest in something cool (by buying art) or, you know, you can do both at once (by becoming a Consumer Venture Capitalist). In theory, this hoary playbook remains as useful for you today as it did in the late 19th century. All you need to do—you bursting-button-downed financier, you—is to pull some baddie with taste in linens and jewels who can help you to sprinkle a George Condo or Vik Muniz on their walls. All you need to do is invest in the hottest new disposable-audio app that every kid in America will be using for the next 7-12 days. And then, surely, you’ll be cool. Right?

Right?

The only thing is that in practice…

When the investor renowned for their money gets in cahoots with a tastemaker renowned for their status, it is the reputation of the tastemaker that remains intact. The tastemaker might get the investor’s money. But the investor can never get the tastemaker’s status. 

I am trying to get at something uncomfortable here, something that the past two years of building a social-financial product has taught me over and over. It is easy to trade social capital for financial capital. But while you can cloak yourself in blue-chip designers all you like to impress your fellow financiers, it is extremely hard to trade financial capital for social capital.

You’ve seen this with every washed-up celebrity you know: when the coolest people become rich, even they can’t remain cool.

II.

What I’m trying to say is something that Web2 taught us long ago: for the masses of humanity, social incentives will always trump financial incentives. Most humans will happily let corporations harvest their data to the highest bidder if it gives them even the remotest chance of appearing aspirational online. Privacy and civil rights advocates can complain, but most people will happily incur massive financial opportunity costs for the sake of social connection that can signal status. Those of us who work in crypto often forget this fact. Most people are normal, and they’d rather score someone who listens to them than score a million bucks.

And besides—please forgive me this dark thought—they know that accruing social capital is one of the few mobile paths to accruing financial capital in the attention economy.

Web2 knew this. And if you want to know why nearly every Web3 social app has failed, your answer is here: it’s because Web3 disastrously decided that Web2 was incorrect, that financial incentives are strong enough to build retention, that people can buy their way into status.

Of course, Web3 had decent reason to think that financial incentives were all that was needed to bootstrap a zealous user base. After all, the original blockchain communities of miners and validators were driven entirely by financial incentives. So were the communities of DeFi protocols. I mean, financial incentives were the whole original unlock of blockchains’ permissionless financial rails! And they seemed to work so, so well in speculative bull cycles as buyers aped into soaring prices to help them soar some more.

But with the advent of crypto apps, DAOs, and NFTs, it started to become clear that financial incentives were often deadly to building meaningful social communities. To believe that blockchains were simply financial tools and that financial incentives were sufficient to bootstrap social communities—well, this was wrong.

It was wrong, first of all, that financial incentives could build retention. In fact, the reason financial incentives are so good at user acquisition is exactly the same reason they’re so bad at user retention—because a mercenary who will use an app to profit will leave it just as soon as the opportunity is better elsewhere. The same people who come for a price that goes up will leave for a price that goes down. Their loyalty means nothing unless you can continue to get them paid.

And it was wrong, above all, that people would be able to convert financial capital to social capital, that as so many 2010s elite coworking spaces promised, people could buy their way to cool. Of course, it’s not wrong that some small mass of delusional buyers will always try to buy their way to being cool. But they’ll quickly kill their own investment since there’s no club that genuinely cool people would like to be part of less than one whose membership can simply be bought. These clubs don’t just exclude the genuine builders and the marginalized voices that have built culture for millennia; they include (I’m sorry) anyone who’s ever decided to sell out. 

If you want to know what crypto social apps keep failing, it’s this: you can’t buy status. In fact, the attempt to do so will only accomplish the opposite. It will mark you as kinda lame.

III.

None of this means, however, that financial incentives don’t play a crucial role in unlocking onchain social apps. Just as it’s popular to believe that financializing social activity is enough to produce a killer app, it’s equally popular to argue against the supposed degeneracy of a casino culture of mercenaries and gambling addicts. The latter view is a reasonable response to the former, but it reeks of snobbishness towards a global underclass that might actually want to make money to feed its family. And more importantly, it’s wrong.

Blockchains are financial rails, and their most radical value propositions for social apps are also their most boring: they let you perform microtransactions with every tap, they let you disintermediate credit card and app store fees, and they give you an open API in the form of onchain metadata for anyone to build on top of. Ideologically, all of this is far less exciting than the revolutionary vision of collective ownership, artist royalties, and decentralized work that inspired and exhausted us in 2021. Financially, all of this probably sounds a good deal less exciting than pure and simple speculation as well. Probably, it just sounds like technicalities.

But consider what this means. Blockchains transform both the ways that social apps can be built, as well as the kind of social apps that can be built, for a very simple reason: they let users monetize directly from other users. Think about the entire history of web2 social apps outside of gaming, and you won’t find a single major app for which this is true.

Financial sustainability for users alone is huge. In fact, it’s never really been done.

IV.

Because here’s the real issue with Web2: it successfully monetized off of social behaviors. But its users didn’t.

So strong were the networks of friends, frenemies, bosses, colleagues, lovers—and maybe most of all, the network of potential friends, frenemies, bosses, colleagues, lovers—that it wasn’t only users who surrendered their data for the harvesting. Corporations themselves gave up the moats they would have had by hosting comms, forums, and job opportunities on their site.

This was the power of social networks: social incentives won, and they did at financial and reputational incentives’ expense. No, you would not earn money from your valuable content; the social network would. No, you couldn’t programmatically own or access or share the reputation you were building as a star creator on a given platform; the social network alone could leverage it for new users and ads. The goal was to become famous on one platform in order to monetize literally anywhere else.

Another way to frame this, I think, is that web2 was an app era, which is to say that it was a closed-data era. An individual’s data lived in the siloes of a given app, and this model is what enabled the apps to monetize by selling this data to advertisers. In short: in closed-data eras, ads and apps will win. Everyone needs to congregate on their platforms to be able to share their data with each other.

But then came crypto, and we entered the onchain era.

Crypto marked the start of a protocol era, or rather, an open-data era. Now individual’s data could be ported freely between apps, and there was no proprietary data to sell in open-source onchain networks. And in place of ads, a new model arose: tokenization.

At their core, tokens offer a somewhat gawky solution to the very real problems of permissionless technologies that anyone can input any kind of data into a system. Tokens are, essentially, legitimacy technology for a mass of users to put up economic collateral attesting that one transaction is legitimate and another is not. You no longer make money selling the data to ads. You make money by putting down economic stakes to attest that the data is true.

The reason to participate in crypto from the start, in other words, was the financial incentives.

This blessing, never possible in web2, was also a curse. By this point in this piece, you know the problem: in every bull market (including this one), the quick gains would draw masses of mercenaries to spam chains, farm protocols, buy tokens, shill bags, and launch new tokens, chains, and platforms. But the same financial fervor that overtook individuals during bull markets would turn to financial frigidity in the bear. Just as quickly as the prospect of getting the bag could pull people in, the prospect of losing it would push them away.

There’s another problem here too, though, that’s much less-discussed. Financial incentives on their own tend to be zero-sum at best. One person’s gain is another’s loss, and in the realm of pure speculation, you stand as much to gain in a bull as you stand to lose in a bear. This is why prediction markets—possibly the most-touted use cases for crypto apps for the past 7 years—command a total market of only around 10,000 users during their most popular periods (election cycles). And many of these are probably bots. The expected return is 0, so users have to be quite confident that they know the future better than other users who are also confident they know the future better than them. Having deep insight doesn’t necessarily help you when you’re also competing against others with the same deep insight.

So how do prediction markets get users? Well, by appealing not to rational bets but irrational ones that are tribal in nature: namely, elections, and sports games. People will bet on their own team winning because it matters to them.

You see where I’m going with this: for financial products to truly make money, they have to tap into social incentives.

We knew this, of course.

Web2 had extraordinary social incentives, but awful financial and reputational incentives.

Web3 had extraordinary financial and reputational incentives, but awful social incentives.

Financial incentives were good for making quick money. But social incentives were necessary for building a long-lasting business.

Crypto wins when—and only when—it enables both.

V.

You might not believe me—I know far too many people in this space who think I’m wrong.

So let’s talk about a specific case study: Uniswap.

Uniswap’s protocol has clearly won: it’s used not just by Uniswap, but Cowswap, 1inch, etc. And that’s the issue. Because it’s a fully open protocol, it can be cannibalized by its competitors. Uniswap presents a uniquely crypto-native problem, the likes of which we’ve never really seen in tech: you can lose to your own product. 

The issue here is that onchain apps don’t make fees on their protocol. Partly, that’s for legal reasons. But a protocol with fees would also incentivize competitors to fork it and fragment liquidity for all parties. That might be worth it if there were no other way to make fees—but of course, there’s an obvious one.

Uniswap, like every other onchain app, makes its money on the frontend. The frontend is where it needs to win. Only the frontend, not the protocol, is exclusive to a company in crypto. If projects can’t ultimately drive users to their site, they can’t monetize effectively.

And what drives users to a frontend? Brand, features, UI/UX all matter of course. But one of the great lessons of web2 was that the most important frontend driver is user networks. You go to a site because other users are there to find—and to find you. Just as financial liquidity matters for bootstrapping a protocol, user liquidity matters for bootstrapping a frontend.

Today, you can see that in every decision Uniswap is making. The wallets? The domain names? The acquisition of Crypto: The Game? These are all ways of making users loyal to its frontend. These are all ways of turning Uniswap ever-so-slightly social.

I have no idea what Uniswap has in store, but I imagine we’ll see many similar features in the upcoming year or two. Want to launch your own token? Uniswap could be the place for any LPs to congregate, join a chat, launch campaigns for others to join in.

What I’m trying to get at is: to win on the frontend, you need to win on social.

To build a financially sustainable model at all in crypto, you need to win on social.

VI.

I said before that this is a lesson I’ve been learning personally over the past year.

At Joke, we let anyone create an onchain contest for people to submit and vote on entries. Broadly speaking, contest players might win any of three ways: they might win money, they might win status, and they might win friends. The money is the financial incentive; the status is the reputational incentive; the friends are the social incentive. These are really all the incentives there are.

For example, let’s say someone ran some kind of Shark Tank onchain. The top winner could earn prize money (financial incentives). All of the contestants could earn status from every vote they get (reputational incentives). And voters could form teams around contestants to create an organic community backing them from the start—creating tribes and making friends (social incentives). 

And when I frame it that way, it should already be clear that financial incentives are the least compelling incentives at play. Only the winners earn money, and it’s far from guaranteed. But everyone can earn status by winning even a single vote. And everyone can make friends by creating teams.

Besides which: the act of building reputational and social profiles can lead to all sorts of financial benefits in terms of jobs, communities, and airdrops. But the financial rewards can only offer money.

You can see why it’s popular to believe that to be money-motivated is to be superficial: because it is. Your reputation and your friends represent your underlying values as a missionary for your cause. But your money represents, so often, your ability to sell these out as a mercenary for the highest-bidder.

If this sounds scandalous, crypto has proven it over and over again. One of Web2’s greatest lessons was that social incentives operate something like a marriage: slow-burning, long-lasting, deepening over years of activating relationships for an hour or two a day.

Web3’s lesson, meanwhile, has been that financial incentives operate more like an affair: all-consuming, short-lived, incinerating itself in the ashes of its own passion until it finds a new hot opportunity to pursue. The airdrop-farmers will float on the winds of the highest yield.

Of course, in a world where we all have to pay for food and shelter, we’re all somewhere on the mercenary spectrum, our attention open to the highest bidders. So I don’t mean to shade financial incentives. I simply mean that passion is a powerful tool for acquisition—but only if it can lead to the retention of a marriage. To recognize this means recognizing that blockchains are not simply tools for globally interoperable finance, but for globally interoperable coordination and globally interoperable reputation as well. They are, in fact, the solution to their own problem, the top problem plaguing moats and monetization in this space that we need true social tools to solve. Loyalty.

***

With special thanks to Daisy Alioto, Sophia Drew, Parker Jay-Pachirat, Sean McCaffery, Peter Pan, Kinjal Shah, and Seyi Taylor.