How do Web3 projects design mature business models and token economies?

Original Title: 《Maturity for a successful next Web3 cycle: A case for Token Engineering》

Author: Achim Struve, Outlier Ventures

Compiled by: aididiaojp.eth, Foresight News

The recent collapse of the crypto market triggered by the FTX scandal has revealed the vulnerabilities in the business models and token designs of Web3 projects. This article will not delve into a specific failed project but will focus on the necessary conditions for Web3 token economics to mature sufficiently in the next adoption cycle. For today, better user experiences and lower technical barriers are important "frontend" issues. The "backend" requires better quantitative models through token engineering. Without good marketing, adoption rates, and user experiences, even a Web3 project with fully optimized token design cannot succeed; conversely, a project can have the best user interface and experience and conduct rich community marketing activities, but without thoughtful token design based on quantification, it will still fail.

How to Optimize Token Design?

Token engineering is a relatively young field that first gained widespread attention in 2018[1]. Token engineering refers to the design, modeling, validation, and optimization of token-based economic models through interdisciplinary approaches. According to Sayama, token economics is a complex system with two core concepts: emergence and self-organization[2]:

"Emergence is the important relationship between micro and macro properties of a system. When it is difficult to simply explain them from micro properties, macro properties are referred to as emergence."

"Self-organization is a dynamic process through which a system spontaneously forms rebalanced macro structures or behaviors over time."

Both concepts exist within crypto-economic models and can be seen as the convergence of a system towards certain unforeseen states, resulting from the interactive decisions of individual nodes or entire stakeholder groups. This self-organizing behavior can occur in real-time interactions with utility protocols or through participation in proposals and voting in governance. Therefore, to understand these complex interdependencies, token engineers must conduct data simulations.

Figure 1 illustrates some of the different disciplines that token engineers must utilize when analyzing crypto-economic models. In this article, "token economics" and "crypto-economics" are used interchangeably.

Figure 1: Disciplines related to crypto-economic systems as discussed by Voshmgir and Zargham[3]

Crypto-economic systems are complex adaptive networks composed of different entities with varying goals and incentive structures, interconnected through cryptocurrencies deployed on the blockchain. The term "crypto-economics" was first mentioned in the Ethereum developer community, referring to decentralized blockchain networks where different nodes can interact based on trustless smart contracts. These smart contracts are deployed in code form on the blockchain, ensuring that all transactions between different nodes are completely transparent and immutable. In most cases, certain nodes in the network will attempt to optimize their short-term or long-term financial situations. However, if there is a decentralized governance structure, some agents may primarily optimize to gain more voting power. Voting power enables them to lead the organization in a direction that aligns with their own interests. Note that Eyal's miner's dilemma example shows that optimizing for short-term maximum profits without considering the impact on the ecosystem may harm long-term profits[4].

To find the best decision models for maintaining system stability and sustainability tokens under different market conditions, engineers will leverage some or all of the disciplinary knowledge shown in Figure 1 to create data-driven models or digital twins. Zhang et al. proposed a modeling approach for node-based crypto ecosystems. They developed a general model of the basic rules and behavioral strategies that nodes must adhere to within technical systems[5]. Figure 2 shows a high-level mathematical description of this model.

Figure 2: A generalized model of node beliefs and policies concerning various factors beyond state and behavior, including private signals σ, private goals V(·), partially observable states X, with observable subspace Y⊆X, and some additional environmental stochastic processes δ extracted from potentially unknown distributions.

Figure 3 provides a more specific and simplified illustration of the different fundamental entities and their influencing factors within the token ecosystem. Consistent with Figure 2, the interdependencies between entities can be unidirectional, cyclical, or recursive, leading to a nonlinear complex interaction system. The unidirectional relationships in Figure 3 reflect the sentiments expressed by businesses in the market from a macroeconomic and geopolitical perspective. The transactions of adopters are influenced by token valuations, which in turn are affected by the distribution of token supply, and the token supply distribution depends on buying and selling, forming a closed loop. Through recursive relationships, adopters can also shift their market views based on their perceptions of business and token valuations. Of course, many more factors play a role in these illustrative interaction scenarios.

Figure 3: A summary and simplification of the different influencing factors among various entities in a token economic system without a decentralized governance structure.

Engineers can model various crypto ecosystems, including different layers of blockchains, DeFi protocols, GameFi, Metaverses, and NFT projects. For example, for a generic Web3 business based on fungible tokens, engineers need to make different design choices and assumptions in the model regarding ecosystem and market responses, such as:

• Token supply type: fixed or inflationary
• Monetary assumptions of the overall business model
• Liquidity design
• Planned token buybacks
• Investor token allocations, corresponding discounts, and lock-up schedules
• Token distribution and management
• Adoption assumptions and market sentiment
• Utility
• Granularity and behavior at the agent level

The results of the model will be closely related to its underlying assumptions and input data, and all parameters should be questioned and validated whenever possible. From the perspective of tokens, a well-thought-out business model needs to consider issues related to its intrinsic value and demand, and include strong goal statements, intelligently designed utilities, value flow diagrams, and stakeholder mappings. When token engineers complete this picture, they can elevate the entire crypto business to a more clearly defined and specific quantitative level. This step cannot be overlooked for any Web3 business or project that aims to have a stable, sustainable, and profitable future.

Without Token Engineering, There is No Sustainability

For smaller market cap tokens, macro market sentiment often manifests as higher volatility, thus the crypto market bear market accelerates the elimination of flawed small-cap crypto companies. While the decline in token prices is not the only concern for Web3 enterprises, the price behavior of tokens itself does not represent the overall ability to capture value, but it is an important market indicator. Figure 4 shows the results of a simplified quantitative token model example for a Web3 business. The red circles indicate the point in time when the company's reserves will run out of tokens in about 8 years. This is a typical case of fixed supply token design but was not properly numerically predicted through a token engineering approach.

Figure 4: An example of token supply forecasting through a quantitative token model. This case illustrates an unsustainable ecosystem where the reserve pool will run out of tokens in about 8 years.

To make crypto-economics more mature, improving stability and sustainability is a necessary attribute of crypto protocols. This is why Web3 enterprises cannot avoid conducting quantitative analyses of their ecosystems. Sustainability is a multifaceted term that encompasses different types.

The sustainability of a token ecosystem is directly related to token supply and the value capture of tokens. Supply sustainability means that there must always be enough tokens in the enterprise wallet to grant different stakeholder groups. To ensure long-term supply sustainability, L1 protocols like Ethereum, Cardano, Solana, Avalanche, etc., often have token minting capabilities. In most cases, L1 protocols implement burning mechanisms to counteract the dilution effects of token issuance and aim for long-term deflation. However, inflation rates and network-supported incentives must be predicted at a quantitative level to ensure that there are always sufficient incentives to run validating nodes, thereby enhancing decentralization and security. Projects built on these layer 1s tend to choose fixed token supplies, as projects where investors' shares are not diluted are more appealing to investors. To ensure that there are always enough tokens available for payments, Web3 businesses must carefully plan and complete quantitative forecasts, maintaining stable or appreciating token valuations that align with their interests. McConaghy proposed a sustainability loop model for Web3, drawing on successful Web2 businesses and national economies, which is worth referencing and learning from[6]. From a marketing perspective, a decline in token prices may be a negative thing, but it does not necessarily mean that if holders receive appropriate token rewards, their shares will be diluted in value. Similarly, all exact system states and parameter sets must be tested under different market conditions within the ecosystem, which is equally important to prevent one stakeholder group from potentially exploiting another within the ecosystem. Token supply and valuation are crucial for the success of Web3 businesses, and the data-driven nature of token engineering is necessary in this regard.

Profit Forecasting for All Scenarios

Another factor that token engineers need to consider is the incentives for protocol participation; Web3 enterprises must always consider their customers' motivations for purchasing, holding, and using their tokens. Token ecosystem models can not only help predict the potential profits of enterprises but also help forecast the potential profits of different stakeholders. The adoption of tokens and Web3 businesses heavily relies on monetary incentives for stakeholders. Token holders can qualify for governance participation, product usage discounts, or receive token payments in return through token staking mechanisms. Participation in token staking expresses token holders' trust in the Web3 initiative, allowing them to earn rewards. Token staking mechanisms can reduce circulating supply and create scarcity, with the key being how to design reward yields for adopters.

Since holders typically seek to monetize rewards from payments, reward yields may increase market sell-off pressure in the near future, potentially leading to a decline in token prices. Once participants see this situation continuing, more stakeholders will unstake and possibly sell their tokens, exiting the protocol, which could lead to a downward spiral in token prices. Depending on the lock-up periods, if the rewards for stakers are too low, they may not provide sufficient incentives for people to participate in the protocol and stake, thus reducing the expected effectiveness of that staking. Token engineers need to model stakeholder behaviors in these different scenarios to prevent unforeseen token outflows.

To examine L1 blockchains from different angles and assumptions, CADLabs has conducted many preset experiments with the open-source Ethereum radCAD model[7]. By adjusting the model's given merge[8] date to a past date, the Ethereum Foundation can show the expected profit margins for all proof-of-stake (PoS) validators in the network under different scenarios of total ETH staked and ETH prices in USD, as shown in the color mapping in Figure 5.

Figure 5 shows that if the amount of staked ETH increases, the profits for validators in the ecosystem decrease, indicating higher network validator participation, and vice versa. Buterin mentioned that this effect should lead to a healthy balance of the number of validators, where there are always enough validators to maintain the decentralization of the network, as profits will increase if less ETH is staked, while on the other hand, it can prevent overcrowding of participants[9]. Without the models provided by token engineers, these insights would be difficult to quantify.

Conclusion

This article emphasizes that in the next Web3 cycle, token engineers need to design more mature crypto ecosystems based on quantitative methods. The discipline of token engineering aims to simulate the processes of different important relationships within such systems. The design of token ecosystems is closely related to Web3 fields such as blockchain, chain games, NFTs, DeFi, and the Metaverse. Modeling crypto ecosystems quantitatively is the only way to predict the potential urgent impacts enforced by self-organizing agent behaviors. This article showcases examples of token supply and valuation sustainability, where token engineers need to maintain a data-driven knowledge base through data analysis.

Overall, quantitative modeling and forecasting are essential for the success of Web3 projects. We cannot determine whether the crypto industry will mature sufficiently in the next Web3 cycle, but fortunately, the token developer community, platforms, tools, and companies are growing, indicating an increasing awareness of the facts outlined in this article. This places some pressure on Web3 enterprises to transition from relying on subjective intuition to using deterministic, rational, and quantitative analyses to set their ecosystem parameters.