How does Base achieve a daily revenue of 180,000 USD?

This report will explore the fee structure of Base and highlight the activities that drive its revenue rise.

Written by: Marsbit

Base, created by the cryptocurrency exchange Coinbase, is the most profitable platform in the Ethereum Layer 2 (L2) network, with daily earnings often exceeding the total of all other top Rollup projects. Over the past 180 days, Base's average daily earnings reached $185,291, far surpassing the $55,025 of the second-ranked Arbitrum.

Clearly, Base has become a platform with consistently high revenue, but what drives its economic activities? What other advantages does Base have that leading L2s do not possess, allowing it to create such high value?

This report will explore the fee structure of Base and highlight the activities driving its revenue rise. We found that Base's sorting mechanism and decentralized exchange (DEX) activities are key drivers.

Base's Trading Order Mechanism

The trading sorting on Base is determined by two variables:

1. Transaction submission time (delay);
2. The fees paid by the sender (economic incentive) relative to the complexity of the transaction.

This mechanism is similar to the operating model of logistics companies like UPS: packages are sent in the order they are delivered by users, while allowing senders to pay an additional fee for "express" delivery to obtain faster service. The priority fee mechanism creates a dynamic auction market, consistent with the Ethereum EIP-1559 fee model, achieving a balance between delivery time and economic incentives.

Specifically, transactions on Base include a "base fee" (lowercase "b", please do not confuse it with the chain name) and a priority fee: all users need to pay the base fee when sending transactions, while the priority fee is optional and is only used to accelerate transaction execution.

But how does the sorter decide which "urgent" transaction to prioritize? It does not directly consider the total fee, but rather focuses on the quote for each unit of Gas (the computational resources required), that is, the cost-effectiveness ratio of the resources needed for the transaction versus the benefits it brings.

Let's illustrate this with the example of a logistics company: Suppose the delivery truck has limited space (similar to the Gas limit of a block), and the driver (the sorter) wants to maximize the tip income within that limited space. At this point, there are two packages:

• A large package, base shipping fee of 50 dollars, but priority tip is only 10 dollars, taking up half a truck space;
• A small package, with a base shipping fee of only 20 dollars, and a priority tip of the same amount, taking up very little space.

Although the total cost of the large package is 30 dollars higher than that of the small package, the driver will still prioritize loading the small package because it has a better cost-effectiveness in terms of space utilization.

The sorter of Base follows the same logic, prioritizing transactions with the highest priority fee per unit of Gas, thereby ensuring that the blocks with the "highest cost" of computational resources are also the blocks with the highest returns. Therefore, when two users submit transactions simultaneously, regardless of the complexity or total fees of the transactions, the user who pays a higher priority fee per unit of Gas is more likely to have their transaction included in the block first.

The figure below illustrates this process:

Why is this important? How does Base differ from competing chains?

The EIP-1559 style fee model of Base creates a continuous public market auction environment for block space, allowing users to directly bid for block space based on the urgency and profitability of their transactions. Therefore, although there are still latency competition factors, the economic benefits allow the income of sorters to rise directly with the demand for block space and the profitability of on-chain transactions.

This stands in stark contrast to Arbitrum's primary sorting mechanism. Arbitrum adopts a strict "First-Come, First-Served" (FCFS) model where users primarily compete based on latency rather than economic cost. In this model, the main competition is not about who can pay higher fees, as all users have the same Gas fee and do not use priority fees, but rather who can deliver their transactions to the sorter the fastest. This leads to a "latency race," where professional participants ensure their transactions are prioritized by investing in low-latency infrastructure. In this environment, Arbitrum's fees only rise with the scale of demand and do not effectively reflect the profitability or urgency of individual transactions.

In April 2025, Arbitrum launched the Timeboost feature aimed at creating a more flexible FCFS system, allowing sorters to gain profits similar to priority fees. In fact, Timeboost adds a "fast lane" for transaction execution in Arbitrum, where users can bid to use this lane for a limited time. Users entering the fast lane can achieve near-instant execution, while other users are still processed in FCFS order, with only a 200-millisecond delay added to compensate for the priority of the fast lane. Although Timeboost introduces a form of priority bidding, its mechanism is more predictive rather than reactive compared to Base's priority fee model. Bidders must predict the potential total profits for a future time period and bid based on estimates. This means Arbitrum receives fixed fees from the winning bidders, regardless of the actual profit situation during that time period. This proactive fixed-rate model is less effective than a reactive system where users bid separately for each transaction in capturing the value of sudden high-profit trades.

How much income does Base generate?

In the past 180 days, the daily average revenue of Base was $185,291. In comparison, the daily average revenue of Arbitrum was $55,025, and the total daily average revenue of the other 14 Ethereum L2 networks was $46,742.

Since the beginning of this year, Base's total revenue has reached 33.4 million USD, Arbitrum 9.9 million USD, and the other 14 L2 networks 8.4 million USD.

Relatively speaking, over the past 180 days, Base has accounted for 64% of the total revenue of the top Ethereum L2 networks ranked by total collateral value, consisting of 15 networks. Its share has significantly risen over the past year, increasing by 48 percentage points from an average daily share of 37% in July 2024, calculated using the 7-day moving average of daily revenue share. As of July 20, due to increased on-chain activities, Base's share of Ethereum Rollup revenue has decreased to 49.7%.

The Important Role of Priority Fees

The transaction fees on Base consist of two main components and an optional priority fee:

• Layer1 (L1) fees: Used to pay for the cost of submitting a batch of L2 transactions to the Ethereum mainnet. In March 2024, after the Ethereum Pectra upgrade introduces "blob (data block)" through EIP-4844, the L1 fee component for Base (and general L2 transactions) will decrease significantly. This is because submitting bulk data via blob is more cost-efficient than submitting it in the form of call data in L1 transactions.
• Base Fee: The mandatory fee for executing transactions on Base. Set by the protocol, it fluctuates based on the utilization of the previous block's space—higher when the network is busy, and lower when it's not.
• Priority fee: An optional fee, also known as a "tip," used for prioritizing the execution of transactions. Priority fees help transactions gain a better position in a block or ensure that the transaction is included in the current block rather than delayed to the next block. A block can contain thousands of transactions, which are executed based on their slot order. Typically, the first slot in a block has the highest value because transactions in that position are executed first and are not affected by subsequent transactions.

Priority fees are the main source of Base revenue, with users obtaining accelerated execution through bidding. In the past 180 days, the daily average priority fee revenue of Base has reached 156,138 dollars, accounting for 86.1% of its daily average revenue.

Specifically, the priority fees at the top slot of the Base block are a significant contributor to the sorter’s revenue, as users compete for positions near the top of the block. Since 2025, the priority fees paid for transactions in the first slot of each block have contributed 30% to 45% of the daily revenue. Additionally, during the same period, the priority fees paid for transactions in the top 10 slots of each block contributed 50% to 80% of the daily revenue. However, in the weeks following July 5, the share of priority fees from the top slot significantly decreased in relation to the total daily revenue. This is attributed to two factors: 1) An increase in traffic raised the base fees, which diluted the share of priority fee revenue; 2) The implementation of "Flashblocks" on July 16 (which will be detailed below) caused high-priority transactions to fall into lower slots within the block (but as we will see, this is not necessarily a bad thing).

Priority fees mainly come from a small number of addresses. Over the past year, 64.9% of priority fees came from only 250 addresses. The top address accounted for 3.6% of all priority fees during the same period, which, calculated at the ETH price at the time of payment, is equivalent to 1.99 million USD.

What are Flashblocks?

Developed by Flashbots, Flashblocks aims to improve transaction processing speed on Base. To achieve this goal, it introduces "sub-blocks", which are high-confidence pre-confirmations of block sub-parts created approximately every 200 milliseconds. For example, a block can contain three different sub-parts, allowing users to obtain transaction pre-confirmations for these sub-parts before the block is confirmed on-chain at the set 2-second interval. This means that even if Base's block interval remains unchanged, end users can experience transactions as if they were completed almost instantly, resulting in a smoother and more responsive experience.

Why is this crucial for Base network fee analysis by slot? Because from the perspective of transaction ordering, each "sub-block" is effectively treated as a new block. Therefore, high-priority fee transactions may fall into a lower slot of the overall "confirmed block," but are at the top of the "pre-confirmed sub-block."

The figure below illustrates the difference in priority fee distribution for the top 200 block slots before and after the implementation of Flashblocks. The black bars represent the proportion of priority fees for each slot; the blue line represents the cumulative proportion of all slots up to that slot (Pareto distribution).

In the week before the implementation of Flashblocks, the Pareto curve sharply rose in the top 10 slots and then grew linearly towards the 200th slot. In contrast, within a week after the implementation of Flashblocks, the Pareto curve was relatively flat at the lowest slots, only starting to steeply rise around the 50th slot of each block — indicating that high-priority fee transactions are falling into the later slots of confirmed blocks.

Impact of DEX Trading

The DEX activities on Base are very active. Among all Ethereum L2 networks, Base has the highest daily DEX trading volume, accounting for 50% to 65% of the L2 network DEX trading volume, and its DEX total value locked (TVL) is the highest among all L2 networks (excluding perpetual futures DEX).

The activity of DEX is an important reason why Base's priority fees remain high. Between 50% and 70% of the total fees earned by the Base sequencer daily come from priority fees for DEX trades. However, since July 7, the proportion of DEX trading fees in the total daily fees has decreased from 67% to only 34%. This is attributed to two factors: 1) the rise in base fees has diluted the share of priority fees; 2) increased competition for on-chain block space has forced users to pay priority fees for non-DEX trades.

Since 2025, the priority fees paid for the first slot of DEX transactions have contributed 30% to 35% of the total daily priority fees, while the top three slots have contributed 50% to 62% of the total daily priority fees. Recently, the proportion of priority fees from the top slots has decreased due to increased overall competition on-chain leading to higher priority fees for non-DEX transactions, as well as the implementation of Flashblocks causing high-priority DEX transactions to fall into lower slots within blocks.

Conclusion

By analyzing the DeFi and yield structure of Base, we found that:

• Priority fees constitute the vast majority of the earnings;
• In the past year, over 60% of the priority fees came from just 250 addresses;
• High DEX trading volume and TVL;
• The priority fees from DEX trading contributed nearly three-quarters of the total priority fees.

These points indicate that maximum extractable value (MEV) trading, particularly competitive activities such as DEX arbitrage, is an important source of revenue for Base sequencers. The EIP-1559 style fee model adopted by sequencers is the direct mechanism to achieve this: it transforms block space competition from inefficient latency-based races to efficient economic auctions.