How much service fee should a Web3 platform charge?

Fees are far more than a simple extraction tool; they can also serve as a coordination mechanism.

Written by: Gérard Cachon, Tolga Dizdarer, Gerry Tsoukalas

Translated by: Luffy, Foresight News

Web3 aims to reduce reliance on intermediaries, thereby lowering service fees and giving users greater control over their own data and assets. For example, Gensyn (a decentralized AI computing platform) offers AI computing services at a fraction of the price of Amazon Web Services (AWS); Drife (a decentralized ride-hailing platform) promises to help drivers escape Uber’s commission exploitation of up to 30%.

However, while the idea of reducing costs for users is appealing, establishing reasonable fee and pricing standards requires platforms to balance the interests of multiple parties. The most successful decentralized marketplaces do not completely abandon fees; instead, they combine “decentralized pricing” with thoughtfully designed, value-creating fee structures to achieve supply-demand equilibrium.

Based on our research, this article will elaborate on the following: the role of pricing control and fee structures in platform economies and governance; why “zero-fee” models are doomed to fail regardless of the designer’s good intentions; and how blockchain platforms should formulate pricing strategies. We propose a new “affine pricing” model based on transaction volume, a mechanism that resolves the conflict between private information and market coordination.

Why Pricing and Fees Matter

The rise and fall of digital platforms depend on their ability to manage two core levers: pricing control and fee structure (i.e., how much the platform charges buyers and sellers for using its services). These are not only revenue tools but also market design mechanisms that shape user behavior and determine market outcomes.

Pricing control determines “who sets the transaction price.” For example, Uber uses a centralized algorithm to set fares, optimizing for supply-demand balance and price stability; in contrast, Airbnb gives hosts the autonomy to set their own prices, providing only algorithmic suggestions as moderate guidance. Each model addresses different priorities: centralized pricing ensures coordination efficiency in large-scale markets; decentralized pricing allows service providers to incorporate private information (such as costs, service quality, and differentiation advantages) into their pricing strategies. Neither model is absolutely superior; their effectiveness depends on the specific application scenario.

The impact of fee structures goes beyond platform revenue—it also determines which participants enter the market and how the market operates. The Apple App Store charges up to a 30% commission, a fee used to screen for quality app supply and fund platform infrastructure, which may dissatisfy developers but usually does not directly affect users; in contrast, if ticketing platform Ticketmaster’s high fees face alternatives, artists and fans may shift to other channels. On the low-fee end, Facebook Marketplace’s free product listing service has bred scams; several near-zero-fee NFT platforms have seen an influx of low-quality NFTs, resulting in a chaotic user experience.

The pattern is clear: excessively high fees drive away suppliers; excessively low fees harm the quality of services/products.

Many blockchain projects adopt a zero-commission model, based on the logic that if the platform gives up value extraction, it will yield better outcomes for suppliers and users. However, this view ignores the critical role of “well-designed fees” in effective market operation: fees are far more than a simple extraction tool; they can also serve as a coordination mechanism.

The Trade-off Between Information and Coordination

The core dilemma in platform design is: how to balance “leveraging private information from service providers” with “coordinating the market for greater efficiency.” Our research shows that the way pricing control and fee structures interact determines whether this dilemma is resolved or exacerbated.

When the platform directly sets prices, it is easier to achieve supply-side coordination and competitive alignment among service providers, but since it cannot know each provider’s private costs (such as operational or marginal costs), pricing often leads to mismatches for both supply and demand: prices may be too high for some users and too low for some providers. Since platforms usually charge commissions based on transaction amounts, this inefficient pricing ultimately leads to profit loss.

If service providers set prices themselves, in theory, their prices can reflect true costs and service capabilities: low-cost providers can gain a competitive edge by lowering prices, thus achieving better supply-demand matching and market efficiency. However, a lack of coordinated pricing can backfire in two ways.

When products or services are highly homogeneous, a race to the bottom on price is likely. High-cost providers are forced out, reducing supply; meanwhile, demand is often rising, ultimately weakening the platform’s ability to meet market needs. While the average price drop may benefit consumers, it directly undermines the platform’s commission-based revenue model.

When products or services need to be paired to maximize value, providers tend to set prices too high. Although many providers may flood the platform, their individually high prices push up the market average, eventually driving users away.

This is not just theoretical: in 2020, Uber tested the “Luigi Plan” in California, allowing drivers to set their own fares. The result was that drivers generally set fares too high, causing users to switch to other ride-hailing platforms, and the plan was terminated after about a year.

Key conclusion: these outcomes are not accidental, but rather equilibrium results under standard commission contracts. Even optimizing commission contracts may still lead to such persistent market failures. Therefore, the core issue is not “how much commission should the platform charge,” but “how to design a fee structure that ensures the market works for all participants.”

How to Solve the Problem

Our research finds that a targeted fee structure can cleverly solve market coordination problems while retaining the benefits of “personalized pricing.” This affine fee model uses a “two-part tariff” mechanism, requiring service providers to pay the platform:

• A fixed base fee per transaction;
• A variable fee: which increases with transaction volume (a surcharge), or decreases with transaction volume (a discount).

This model has differentiated effects on providers based on their costs and market positioning.

In such markets, providers’ costs vary significantly: some have lower costs due to advanced technology, access to renewable energy, or efficient cooling systems; others have higher costs but can offer premium services such as high reliability.

Under the traditional commission model, if market competition is excessive, low-cost GPU providers will set aggressively low prices and capture too much market share, leading to the aforementioned market distortions: some suppliers exit, limiting transaction volume, while the market average price is driven down.

For this scenario, the optimal strategy is a “transaction volume surcharge”: the more customers a provider serves, the higher the fee per transaction.

This mechanism imposes a “natural constraint” on aggressive low-cost providers, preventing them from capturing an unsustainable share of the market at excessively low prices, thus maintaining market balance.

When market competition is moderate or insufficient, the optimal strategy becomes a “transaction volume discount”: the more customers a provider serves, the lower the fee per transaction. This mechanism incentivizes providers to expand transaction volume by lowering prices, effectively enhancing market competitiveness without pushing prices below sustainable levels.

For example, on decentralized social platforms, “creators with higher user engagement” could be charged lower fees, encouraging them to set more competitive prices for paid content and attracting more user participation.

The brilliance of the affine fee mechanism is that it does not require the platform to know each provider’s specific costs; the fee structure creates positive incentives, guiding providers to self-adjust based on their private cost information. Low-cost providers can still gain an advantage by pricing below higher-cost competitors, but the fee structure prevents them from monopolizing the market in ways that harm the overall ecosystem’s health.

Our mathematical simulations confirm that a well-calibrated “transaction volume-based fee structure” enables the platform to achieve over 99% of theoretical optimal market efficiency. In theoretical frameworks, it far outperforms “centralized pricing” and “zero-commission” models. The resulting market will have the following characteristics:

• Low-cost providers retain a competitive edge but do not capture excessive market share;
• High-cost providers can continue to participate by focusing on “differentiated service niches”;
• The overall market reaches a more balanced equilibrium with reasonable price differences;
• The platform achieves sustainable revenue while enhancing market functionality.

Additionally, analysis shows that the optimal fee structure depends on “observable market characteristics,” not each provider’s “private cost information.” When designing contracts, platforms can use observable signals such as “price” and “transaction volume” as proxies for “hidden costs,” allowing providers to retain pricing power based on private information while solving the inherent coordination failure in fully decentralized systems.

The Future Path for Blockchain Projects

Many blockchain projects, by adopting traditional commission models or zero-fee models, have harmed both their own financial sustainability and market efficiency.

Our research confirms that designing reasonable fee structures is not contrary to decentralization, but rather a core element in building a functional decentralized marketplace.