Author: Epistemic Meditations

Original title: A Framework for Network Effects in Web3

The terms "Web3" and "network effects" are now becoming more commonly misused than "machine learning" or "artificial intelligence." These terms are used indiscriminately by Twitter big Vs who follow hot topics, which makes people feel suffocated and headaches. Over time, these two terms have become buzzwords, but the essence of their operation has been ignored. Even within the so-called "crypto-native" community, there is a strong consensus that the Web3 network will become the mainstream Internet model, but no one seems to have a very clear idea of ​​how it will actually have network effects.

Therefore, the DAOrayaki community sorts out and compiles Web3 articles, and restores how Web3, as a network, can realize its potential. In this process, the DAOrayaki community filters the noise, breaks down terms like "Web3" and "network effects" into their simplest components, and then uses these components to model how the Web3 network can unleash/motivate a new network effect.

The content of this article is divided into three parts. In Section 1, we first establish the basic elements that make up a Web3 network, and the technical and economic design differences between Web2 and Web3 networks. Then, in Section 2, we explore how Web3 networks leverage these fundamentals to enhance network effects. Finally, in Section 3, we will examine the framework established in the previous sections to see if it can yield practical insights into the future of Web3 networks.

Section 1 - Modeling Network Effects, Web3 Fundamentals and Arbitrary Web3 Networks

Let's start from the beginning. What exactly are network effects? Simply put, a network effect is an increase in the value of a platform caused by an increase in usage of the platform. I find it useful to model network effects through the following process.

To expand new users, it is necessary to generate enough value to persuade users to stay, and a higher retention rate needs to fully meet the value needs of new users, and so on. In the early days of the web, both conversion and retention need to accelerate, or the flywheel will start spinning in opposite directions.

The founder who discovered the secret of network effects in the "Web2 era" has expanded social networks, applications, and markets to billions of users, but the Internet network era after 2010 is an era of zero-sum stagnation. The rapid decline in speed that managed to achieve this scale is evidence of that.

So why has the Web3 network shown insane growth over the past 12 months? Why a video game as simple as Axie Infinity can boost its monthly revenue from about $10 million to over $350 million in two months, or why a "creator token" network like Rally can Valuation at nearly $1 billion?

To the disappointment of Web3 skeptics, answers like "hype," "Ponzi scheme," and "virtual scam" don't quite tell the story. To move beyond these explanations, we must pinpoint key design differences between Web3 and Web2 networks.

Clarify the definition of Web3: "Move to Web3" = adopt the basic elements of Web3

For our purposes, Web3 refers specifically to a new set of discrete digital elements implemented by a decentralized virtual computer (aka blockchain). More specifically, we refer to smart contracts, which are the underlying primitive blockchains that support Turing-complete programming languages ​​such as Ethereum. We believe that "any smart contract" is the root node of the basic elements of Web3, and each basic element is itself a specific instance of a smart contract.

The composability of smart contracts enables them to act as conduits for these different building blocks to work together within and across Web3 networks. It is the interplay of all these essential elements, facilitated by smart contracts, that constitutes"Web3 network"Foundation.

Economic differences between Web2 and Web3 networks.

We can model Web2 and Web3 networks as a combination of raw material and capital flows. There are two kinds of raw materials in the network - users and technology platforms, and two kinds of capital - user capital (u) and platform capital (p). User capital includes any scarce, user-provided resource, including user-generated content, physical inventory, money, or even just attention. Platform capital is not only the income generated by the platform, but also includes the non-financial utility that the platform gives to users.

Users provide "user capital", and the technology platform converts it into "platform capital", some of which are allocated to users and some are withdrawn from the network-this is the basic action of any Internet network.

The essential difference between Web3 and Web2 networks is their treatment of platform capital. While the Web2 platform siphons off nearly all the capital they generate, smart contracts allow the Web3 network to return the capital generated by the platform to users at the value of the user capital provided by them via tokens.

From an economic design perspective, this means that Web2 networks are biased toward authoritarianism, while Web3 networks are biased toward liberalism/capitalism.

Web2 Network Design (Absolutism):

Web3 Network Design (Libertarian/Capitalist):

This is neither to say that all Web2 networks are completely authoritarian, nor that all Web3 networks can be labeled "liberal/capitalist". The web can certainly use the basic elements of Web3 to create a "tyrannical" economic model. However, enabling users to capture the value of their digital labor has been the guiding ethos of Web3 network design (and thus the kind of Web3 network we want to focus on).

The above assumptions about the Web3 network depend on "governance tokens" capturing the value of the platform capital generated by the Web3 network (giving them a non-zero financial value). Assigning value to tokens may sound counter-intuitive at first.

In fact, tokens can (and most do) have zero financial value. However, those tokens that capture value possess at least one of the following characteristics:

• Platform Utility - The Web3 network uses smart contracts to provide token holders with privileges within the network, which drives demand for tokens and thus financial value.

• Ownership of community treasury funds - revenue generated by the Web3 network is transferred to community-owned treasury smart contracts. Governance tokens derive their value from the network's treasury's future revenues, just like stocks derive their value from a company's future cash flow.

Governance tokens can also combine these two characteristics, tying application utility with network ownership. For example, holders of Uniswap’s governance token UNI can gain voting rights and ownership of the $7 billion currently in the coffers (https://openorgs.info/). Even if the network is not currently generating revenue, Uniswap's smart contracts contain a "fee switch" that can be turned on by a vote of UNI holders, which will transfer transaction fees to the treasury. Voting functionality and ownership of existing treasury assets, plus the potential for future income drive the financial value of UNI (currently priced at $17). The implication is that tokens that do not have these characteristics do not play a substantial role in the network effects we are about to discuss.

Section 2 - How Networks Use Web3 Fundamentals to Drive Network Effects

Now that we have built a robust model for any Web3 network and its associated fundamentals, we can begin to flesh out the role of these fundamentals in influencing the various components of the network effect (conversion, retention, value creation).

Not all users are created equal - "value producers" through token rewards

Contrary to this, when it comes to user conversions, what any network is trying to achieve is not the sheer number of new users. Going back to our arbitrary internet network model, the network is actually designed to maximize net user capital (invested by new users).

The nascent network is more concerned with acquiring "value producers" rather than "value consumers", because value producers invest much more user capital and help attract value consumers. The types of value producers also vary from network to network. For a network like YouTube, the value producers are the content creators. For ride-sharing apps like Uber, they are the drivers. For a dating app, although it might be a bit odd to use this framework, the users generating the highest value are beautiful women.

Although the flatness of the distribution of value producers and consumers in different networks is different, because value production itself is much harder than consumption, value producers usually come less than consumers.

The incentive for value producers to join the Web2 network is based entirely on the rewards they receive from value consumers on the network. If there are no guests on Airbnb, renters have no incentive to upload listings. If there are no riders on the Uber network, there is no incentive for drivers to sign up. And without a large audience on Twitter, there's no way to incentivize well-known users to tweet. This is also known as the often criticized cold start problem.

In contrast, the Web3 network can acquire value producers without a certain number of consumers, and we reward them with governance tokens in exchange for their investment of upfront user capital. For example, a DeFi protocol like Sushiswap rewards early liquidity providers with SUSHI tokens, while a Web3 social application like Rally Network rewards fast-growing early adopters of its network with a weekly airdrop of RLY tokens. Creator community.

This means that the Web3 network can acquire value producers purely by promising a share of future platform capital, as the financial value of this commitment is already priced into the governance token being distributed. Again, this dynamic does not arise unless the tokens being distributed possess the characteristics we outlined in Section 1.

While many types of token reward mechanisms have been implemented, effective mechanisms often disproportionately reward users who invest user capital early on, since all forms of capital are most scarce and valuable at the outset for any network. of value. Web3 networks typically reduce token rewards based on the number of new value producers they acquire to incentivize value producers to join as soon as possible.

By guiding the entry of early value producers, the token reward mechanism increases the utility of the network technology platform. In addition to this, the value (price) of the network’s increasingly scarce governance token increases along with the network’s platform utility. This means that token rewards have a dual means of increasing network effects, one by increasing the non-financial platform utility of the network and the other by increasing financial incentives to facilitate conversion.

Token rewards play an important role in the types of users who choose to join the Web3 network. A network with an effective token reward mechanism is designed to optimize its hierarchy of token holders, an important concept we will revisit later that largely determines the ability of a Web3 network to build a moat.

Second-order and third-order network effects of the token reward mechanism:

Since user capital is scarce (users have limited time, money, attention, etc.), how much they are willing to invest in any given network is determined by their subjective willingness to invest. The capital invested by any user in the network can be described by the following formula:

For Web2 and Web3 networks, W is a function of the relative rewards users expect to receive in a network. Both networks must compete for user capital by increasing the rewards users expect to receive, thereby increasing W.

The token reward mechanism is one way the Web3 network does this. As users earn governance tokens, their net incentive to continue investing capital into the network (to increase the value of their existing ownership) increases accordingly. Governance tokens are scarce by design, and networks that prioritize rewarding high-value users with these scarce rewards maximize future user capital committed to the network. More precisely, such a network maximizes the sum of the following formulas, where U is the set of all users in the network and I is the total user capital invested in the network.

In contrast, Web2 networks often acquire early value-producing users through cash-based incentives. Networks like Paypal have spent hundreds of millions of dollars on such tactics. However, governance tokens are fundamentally different from cash payments in that they contribute to the long-term financial well-being of the network. This means that the token reward mechanism gives the Web3 network a greater ability to intervene in W than the Web2 network. Another way of saying this is that the Web2 version of this strategy only incentivizes conversions, whereas the Web3 version incentivizes both conversions and retention.

The compounding impact of token rewards on network effects continues beyond the actions of individual users. In the Web2 network, value producers must engage in zero-sum competition for a limited share of platform-generated capital. YouTube content creators compete for views, Twitter users compete for followers, Uber drivers compete for riders, and so on.

This situation is also unavoidable in the Web3 network, but because governance token holders all benefit from the rise in network value, the negative impact of competition is offset, which is called "consistency of user incentives". This is why governance tokens are often referred to as "digital equity". However, unlike traditional equity, through token rewards, the number of "digital equity owners" can reach the scale of tens of thousands very early on. This means that in the Web3 network, the net amount of cooperation/positive sum interaction between users is greatly increased, which ultimately improves the efficiency of user capital deployment.

Network Effects Driven by Web3 Composability:

Unlike Web2 networks which are closed source and isolated, Web3 networks are open source and composable. Smart contracts, tokens, and other Web3 fundamentals in one network can interact naturally with fundamentals in other networks. Most importantly, the rules of these interactions are governed only by the code in the relevant smart contracts and do not require human coordination to enforce.

The network effect caused by composability can be interpreted as a mutually beneficial transaction of capital between Web3 networks. In other words, network composability enables users to invest capital in one network to generate benefits in another network.

DeFi protocols are a perfect example of composability-driven network effects. For example, cryptocurrency assets provided by users to Yearn Finance are used to provide liquidity to networks such as Compound in exchange for yield, while increasing the yield of Yearn Finance users and improving the application utility of Compound users. As the number of protocols reaching maturity in DeFi continues to grow, the opportunity for Yearn Finance to integrate more yield strategies (and the accompanying network effects) from other networks continues to increase.

As the broader Web3 space matures, we can expect composability to play a huge role in driving network effects, not only for DeFi protocols, but also for consumer applications of Web3. We've seen several examples of composability strategies at the intersection of DeFi and Web3 social, such as Rally Network using Yield Delegating Vaults on Yearn Finance to manage their community finances (https://amit-rally.medium.com /introducing-yield-delegating-vaults-f861a11afb0b). If we push this trend to its theoretical limit, it should come as no surprise that the growth multiple created by composability across networks is the main driver behind Web3 over Web2.

The relationship between open source, composability and moat/retention of the Web3 network:

While the composability and open source nature of Web3 networks can generate strong network effects between projects, it clearly increases the attack surface of new Web3 networks. For example, despite gaining massive traction after mainnet launch, Uniswap was still vulnerable to a “vampire attack” that temporarily lost liquidity to Sushiswap (a fork of Uniswap that has a similar reward mechanism for governance tokens). This attack allows capital to be taken out of one network and placed in another without the capital transactions that facilitate mutual benefit. Examples like this should be of legitimate concern to us.

Given that Web3 software is not proprietary and there are few exit barriers for token holders, is it possible for a Web3 network to build a moat for defense? Before answering this question, we need to understand what exactly "retention" and "moat" mean in the context of Web3.

For any Web2 or Web3 network, users can be considered "retention" as long as they continue to invest a minimum amount of user capital into the network. The Web2 network measures this by analyzing application usage data. The same usage data is available to the Web3 network, but retention can actually be understood as the willingness of users to hold (hodl) their governance tokens. When users withdraw their staked tokens, we can consider them quitting.

The durability of any Web3 network moat is tested every time there is a significant negative movement in token price, which could be caused by events such as vampire attacks, collusion among token holders, security breaches, changes in market conditions, etc.

When Web3 networks successfully build a moat, they will convince a certain number of users that holding their stake is a profitable strategy. As we mentioned earlier, some users are significantly more important to retention (value producers) than others (value consumers). One way to divide users in a Web3 network is to divide into "missionaries" and "mercenaries". Missionaries are generally value producers, believe at heart in the mission of the web, and have a disproportionate influence on the humanity of online communities. On the other hand, mercenaries are more about consuming value, lacking the internal motivation to develop the network, and their motivation is purely for economic interests. We can measure where users are on this spectrum by evaluating how they react to negative fluctuations in the token price.

"Preachers" have long-term confidence in the growth of the network they are a part of and will hold on to their token stake even during large negative price swings, while "mercenaries" take a short-term view when token prices start to falter will exit their shares.

We now introduce the concept of a network’s token holders, a term that refers to how governance tokens are distributed among the network’s missionaries and mercenaries. Different token holder structures have different "robustness", here robustness refers to a network's ability to retain its token holders after negative fluctuations in its token price. We can estimate the robustness of a network's token holder structure by computing the sum of the following equations.

The assumption behind this equation is that when the largest number of token holders on the network are the least price-sensitive users (missionaries) and the minority token holders are the most price-sensitive users (mercenaries), the network is the most robust.

This equation demonstrates the critical role token rewards play in establishing the durability of the network moat. If mercenaries can hijack the reward mechanism and gain substantial network ownership, then the network's token holder structure is unsound, most users will churn when token prices are disrupted, and the network will not be able to maintain its moat. On the other hand, if a network's token reward mechanism rewards missionaries more, then the network can gain a larger and more durable advantage over competitors.

This explains why networks like Axie Infinity have built durable moats without a materially differentiated product. Sky Mavis, the company that created Axie, quietly developed the network for years in the post-2017 crypto winter. Therefore, most of their early users are missionaries of value production, who can firmly support the game despite the bear market.

According to Jeff 'Jiho' Zirlin, co-founder of Axie Infinity, "the community first needs to be small enough so that it can be big enough". This is what makes Axie so hard to replicate - if you're a competitor right now, you're going to attract people who want to find the next Axie, not people who are really interested in pushing the game forward.

Part 3 - Applying the Web3 Network Effects Model to Generate Practical Insights

So far, we have outlined the main ways in which Web3 fundamentals can facilitate new and powerful network effects. However, the power of the analytical framework we have built will depend on how much practical insight it can provide us about the future of Web3 networks. Here are some insights I have drawn on building on the framework established in Sections 1 and 2.

Capitalist design will be the dominant web design pattern in Web3

Web3 fundamentals ultimately make it possible for users to receive revenue from their digital labor in the form of governance tokens, resulting in a fairer and unrivaled user experience. Those networks that choose not to design with this momentum-driven tokenomics will lose out on user experience and will be left out of the competition.

The corollary here is that Web3 networks will compete on the rewards they allow users to acquire, which means we can expect Web3 networks to become less and less adversarial to users over time - unlike Web2 networks and their The relationship between users of , evolved in the exact opposite way.

Web3 network moats will be easier to build than Web2, but harder to defend

When Web3 networks keep growing and optimizing their token holder structure to be strong enough, they can build moats. However, because the new Web3 network is so good at attracting high-value users, it will disrupt existing networks. This makes the moat of the Web3 network more volatile as it builds and collapses.

The value of each user of the Web3 network will be 10 times that of the Web2 network

As we discussed, the Web3 network unlocks capitalism on the Internet. Throughout history, every major transition to a capitalist economy has resulted in a huge increase in total output (and of course GDP). The Web3 network will increase the average production value per user by a similar magnitude. From a distance, it appears that Web3 networks are achieving more than 10 times higher value per user than Web2 alternatives.

The granularity of the token reward mechanism (Airdrops) is getting finer

Early experiments with token rewards were fairly straightforward. For example, Uniswap’s initial airdrop was to provide a uniform 400 UNI tokens to any wallet that used the protocol by a set date. As this space matures, Web3 networks will compete to maximize the amount of future user capital invested by their users. This means that token reward mechanisms will continue to upgrade in terms of measuring and rewarding value behaviors, inevitably making these mechanisms more granular and complex. This increase in granularity will be exacerbated by the addition of Web3 social applications, where investments in "user capital" are made through very small interactions.

Hybrid networks (networks with centralized and decentralized components) will scale more efficiently

As the Web3 ecosystem moves beyond the prosperity of DeFi and into the era of Web3 consumer social applications, it becomes difficult to accommodate a large number of user interactions on-chain. Since these new types of wide-reaching Web3 networks are "not fully decentralized," there is growing concern that these networks will not succeed because Web3 networks must be "community-owned and controlled."

As long as the network still allows users to gain ownership of the network through governance tokens, the lack of decentralization will not disrupt growth too much, despite the limitations placed on user control by centralized network components. That's because, generally speaking, users value ownership of network revenues far more than control over their technology platforms. In other words, a properly designed hybrid network can still achieve all the “Web3 network effects” we outlined in the previous chapters.

Niche Web3 networks will survive more easily than niche Web2 networks

Web2 consumer Internet networks will only survive if they scale to millions of users. Users in the Web3 network generate and capture enough value that they can be sustained with a much smaller number of users, benefit from composability-driven network effects, and effectively attract value-producing users.

at last

at last

We would like to hear your thoughts on the models illustrated in this article. Do you think there are other types of network effects in Web3 worth considering? Are there hidden assumptions in the proposed model that are worth dissecting? Agree/disagree with this prediction? Join the DAOrayaki Discord channel to discuss and analyze Web3 together.