In Crypto, Apps Need Infrastructure Like Infrastructure Needs Apps

Or, maybe the real infrastructure was the applications we made along the way

In 2018, Dani Grant and Nick Grossman of USV published The Myth of The Infrastructure Phase, arguing that apps inspire infrastructure, not the other way around. It seems that five years later (almost to the day!), this debate is more prevalent than ever. 

Tweets from the last six months: 1, 2, 3, 4, 5

If you’re looking to get caught up, the TL;DR is that there’s merit on both sides. Team Infrastructure argues that better tooling improves app UX and allows developers to build more quickly, while Team Applications maintains that across major technology platform shifts, breakout applications (like the lightbulb, the airplane, or AOL) came first before inspiring the infrastructure (the electric grid, the airport, or web browsers) to support broad adoption of those apps.

As much as I’d love to take sides in this debate, I believe it’s too binary of a conversation to begin with. Infrastructure and app needs constantly evolve off each other. Just like the move from dial-up to broadband to Wi-Fi, or the transition from calling for delivery to Seamless to DoorDash, the direction towards success isn’t linear — it’s adaptive on both sides. In crypto, we often like to anchor on existing technologies or use cases, but building in the space fundamentally requires being comfortable with the fact that it all may change. 

One heuristic that largely feels like a distraction in this conversation is the notion of “mass adoption” in crypto — building products for consumer scale. What does “scale” look like in crypto, anyway? On one side, you have businesses that look and feel like traditional consumer products but use blockchain rails under the hood, oftentimes for payments (like Beam or Sphere) or loyalty (like Medallion or Blackbird). Consumers of these products usually have a lack of interest and/or understanding in crypto, and so abstracting as much of the technology away as possible makes for a more seamless user experience.¹ At the same time, on the entirely opposite side, you have crypto-native products for crypto-native consumers — a growing user base of individuals who love trying new products (often jumping through hoops to do so) and spending real sums of money within them.

Credit: @WilsonCusack

Although both of these user personas are attractive from a business perspective, they often feel fundamentally at odds with one another. Existing tooling may “work” for the users and applications who have been here thus far — but if we want to evolve the users, then we’ll have to evolve the infrastructure, too. This is less of a binary (“we need a killer app!”) and more of a push and pull based on consumer need. Let’s use wallets as an example, as the consumer gateway to all activity in the ecosystem:

The first iteration of the crypto wallet (let’s call this Wallet 1.0) was largely limited in terms of use cases. Many, if not most, early crypto adopters cared deeply about the fundamental pillars of the technology, like decentralization, transparency, and immutability. Notably, if users wanted to engage with the technology, they needed to value, or at least understand, the concept of self-custody. A good example of a product from Wallet 1.0 is MetaMask, created by ConsenSys in 2016.² MetaMask had 545,000 monthly active users in July 2020. This measure shot up to over 10 million by August 2021, driven by a surge of interest in “DeFi Summer,” where yield farming acted as a hook for consumers. At that time, it was an obvious choice for consumers to use MetaMask to take advantage of DeFi yields, as MetaMask was far more accessible (widely supported) than competitors.

The implication of this is that successful applications during this time were built with existing MetaMask users in mind — crypto-natives who had a wallet and wanted to use it. It made sense for consumer applications to be inherently crypto-native, based on the market opportunity provided by the infrastructure available. 

Many popular wallets today are still primarily self-custody, serving a similar user base. However, over the past few years, we also saw the pendulum swing entirely to the other side, as consumer needs continued to evolve. Specifically, we saw the proliferation of tools that enabled an easy onboarding to a crypto wallet, so net new users could engage in blockchain-based applications (like NBA Top Shot or Starbucks Odyssey) without having to worry about seed phrases. Let’s call this Wallet 2.0, brought forth by infrastructure like Magic and Web3Auth. In many ways, this solved the problems of Wallet 1.0, making wallet creation seamless and enabling traditional consumer companies to add crypto-native elements to their businesses. But in other ways, it created entirely new problems: instead of having a few self-custody wallets to be used across all dApps, users now had a new wallet for every single application, almost like getting a co-branded credit card for every store. These app-specific wallets siloed consumers into vertical brand ecosystems and removed the potential for portability, as keys were no longer custodied with the user.³ Without the ability to truly own your assets and take them with you… why use crypto at all?

“Though wallets and applications are users’ first touchpoints into crypto, we’ve positioned ease-of-use and interoperability as diametrically opposed.” — Nitya Subramanian

Consumer interest largely forced this change. While Wallet 1.0 catered to crypto-native users, largely focused on trading, Wallet 2.0 swung the pendulum far, introducing brands and mass-market consumer experiences. Now, the pendulum seems to be settling somewhere in the middle — what we can call Wallet 3.0, which combines the interoperability of the first iteration (true self-custody) with the powerful consumer UX of the second. This is being accomplished by service providers like Capsule, which leverages distributed multi-party computation to empower developers to build applications that can interact in-context with user assets on users’ terms, without sacrificing interoperability. As applications and infrastructure continue to pair, wallets become expressive building blocks for consumer identity, and more of the ecosystem begins to compound. 

The evolution of wallets over the last decade shows a constant push and pull between infrastructure and applications. It felt impossible a few years ago to imagine crypto wallets without seed phrases. Now, the opposite feels equally true. In an ever-changing ecosystem, it often feels comfortable to anchor on existing use cases, but success requires being fundamentally open to evolving them.⁴ All of this can change, and it will. And that is a reason to be excited.  

---

Thank you to Jarrod Dicker, Nitya Subramanian, and Ben Roy for their perspectives and feedback on this piece.

TCG Crypto is an investor in Sphere, Medallion, and Capsule. None of the information discussed herein is intended to be, or should be construed as financial advice, or an offer to sell or a solicitation of an offer to buy an interest in any security. The information set forth herein has been obtained or derived from sources believed by the author to be reliable and has been provided solely for informational purposes. Nevertheless, the author does not make any representation or warranty, express or implied, as to the information’s accuracy or completeness. Certain companies referenced herein are included by way of example and not companies in which TCG has invested to date nor companies in which TCG intends to invest.

1. This is a clear evolution of technology moving from front to back. As I referenced in an earlier piece, when tech is fully adopted, it disappears.

2. MetaMask was far from the first crypto wallet, but is a helpful example due to the scale it reached within a short period of time. For a more robust history of crypto wallets, check out Vitalik Buterin’s 2012 review of various Bitcoin wallets, or read about Ethereum’s original wallet MyEtherWallet (and its 2018 split into two competing services).

3. From America Onchain: “You’ve Got Scale?”: “Dumbing down has always necessitated the transfer of sovereignty from the user to the platform.” 

4. See this tweet on crypto semantics from Dan Romero.