What Is An MEV Searcher? How They Profit On Ethereum
MEV searchers are bots that scan blockchain mempools for profitable transaction opportunities. Learn how they fit into Ethereum's block-building economy.
Key takeaways
- An MEV searcher is an entity that scans the blockchain mempool to detect and exploit profitable transaction ordering opportunities before a block is finalized.
- Searchers do not build blocks themselves. They submit transaction bundles to block builders, who compete to include them in the next block.
- The main strategies searchers use are DEX arbitrage, liquidation capture, sandwich attacks, and backrunning.
- Becoming a competitive MEV searcher requires significant technical depth, low-latency infrastructure, and the ability to outbid other bots in real time.
An MEV searcher is a bot or algorithm that monitors a blockchain's pending transaction pool (mempool) to identify and exploit profitable reordering opportunities before a block is confirmed. Searchers earn profit through the difference between what they pay and what they capture.
MEV searching is one of the least visible yet most consequential activities in decentralized finance. Understanding how it works reveals a lot about how value actually moves on-chain.
What Is an MEV Searcher?
| Quick answer: An MEV searcher is a participant, almost always automated, that scans blockchain data to find and act on Maximal Extractable Value (MEV) – profit that can be captured by controlling the order in which transactions appear in a block. |
MEV itself refers to the total value that can be extracted from a block by inserting, removing, or reordering transactions. It exists because blockchains process transactions in batches (blocks), and the party deciding the order of those transactions can influence outcomes, such as price changes, liquidation triggers, and arbitrage windows.
Searchers are the entities that do the hunting.
- They run algorithms that continuously watch the mempool and look for opportunities to profit.
- When they find one, they construct a bundle: a carefully ordered set of transactions designed to capture that value.
- That bundle is then submitted to a block builder for inclusion in the next block.
Searcher vs. validator vs. builder:
- A searcher finds the opportunity and creates the bundle
- A builder assembles the block from multiple bundles
- A validator (or proposer) selects and signs the final block
Searchers do not mine or validate. Their edge is purely informational and computational. Whoever spots and acts on an opportunity fastest wins.
Example of an MEV Searcher in Action
| Quick answer: The clearest way to understand a searcher is to walk through a live opportunity. The example below is a sandwich attack – one of the most well-documented MEV strategies. |
Step 1: Detecting a large swap
A searcher's bot continuously monitors the Ethereum mempool. It spots a pending transaction: a user is about to swap 500,000 USDC for ETH on Uniswap. A trade of that size will move the ETH/USDC price noticeably within that pool.
The bot flags this as a target.
Step 2: Calculating potential profit
The bot runs a simulation. Given the pool's current liquidity depth, it estimates:
- How much the price will shift after the user's trade executes
- How much ETH to buy before the user's trade to benefit from that shift
- What the optimal selling price will be after the user's trade moves the price up
- Whether the profit exceeds the cost of gas and builder fees
If the math is positive and fast enough, it proceeds.
Step 3: Creating a bundle
The searcher constructs a bundle of three transactions, in exact order:
- Searcher buys ETH (front-run) – before the victim's trade
- Victim's swap executes – raises the ETH price in the pool
- Searcher sells ETH (back-run) – at the now-higher price
This bundle is atomic: either all three execute in order, or none do.
Step 4: Submitting through flashbots
Rather than broadcasting to the public mempool, where the bundle could be copied or front-run by other bots, the searcher submits it privately through Flashbots (or another MEV relay). This keeps the strategy confidential until the block is proposed.
The searcher attaches a priority bid: a direct payment to the builder to incentivize including this bundle over competing ones.
Step 5: Winning block inclusion
The builder receives bundles from many searchers simultaneously. It assembles the most profitable combination and presents the block to a relay. The relay forwards it to a validator, who signs and proposes it to the network.
If the searcher's bid was competitive, their bundle lands in the block in exactly the order they specified. The sandwich executes, and the profit is captured.
Where MEV Searchers Fit in the Ethereum Ecosystem
| Quick answer: MEV searchers occupy a specific layer in Ethereum's transaction supply chain. They are not miners, validators, or users, but intermediaries who profit from the gap between transaction submission and block finalization. |
Traditional Ethereum transaction flow
Before MEV infrastructure existed, the process was straightforward:
- A user submits a transaction to the public mempool
- Miners (now validators) collect transactions, order them (often by gas price), and include them in a block
- The block is proposed to the network
In this model, validators controlled transaction ordering entirely. This created a race: bots competed by bidding higher gas fees to get their transactions processed first, leading to gas wars that congested the network and raised costs for everyone.
The modern MEV supply chain
After Ethereum transitioned to Proof of Stake and the adoption of Proposer-Builder Separation (PBS), the system became more structured:
Searcher → Builder → Relay → Validator (Proposer)
- Searchers detect opportunities and submit bundles via private channels
- Builders (e.g., Titan Builder, Beaver Build) aggregate bundles and construct the most profitable block
- Relays (e.g., Flashbots, Ultra Sound, BloXroute) act as trusted intermediaries between builders and validators
- Validators simply select the highest-paying block from the relay
As of late 2025, roughly 90% of Ethereum blocks are built via MEV-Boost, meaning the vast majority of blocks pass through this supply chain rather than being assembled directly by validators.
How searchers work with builders
Searchers do not submit individual transactions. They submit bundles, which are ordered transaction packages with a specific bid attached. Builders receive hundreds of bundles simultaneously and run algorithms to find the combination that maximizes block revenue.
The economic reality is stark: searchers often pay more than 90% of their gross MEV revenue to builders and validators in order to win block inclusion. The searcher's actual take-home is a small fraction of what they extract from users.
How MEV Searchers Make Money
| Quick answer: MEV searchers profit by exploiting predictable on-chain behavior, including price impacts, undercollateralized loans, and state changes created by other users' transactions. Their revenue model is built on four primary strategies. |
Author's perspective: What strikes me about MEV searching is how fully it resembles high-frequency trading on traditional financial markets – the same latency arms race, the same winner-take-most dynamics, the same thin margins masked by large gross volumes. The interesting difference is that in DeFi, this activity happens in full public view on-chain. Every sandwich attack, every arbitrage bundle, every liquidation race is permanently recorded. That transparency does make the economics unusually legible compared to what happens inside a traditional exchange's order book.
DEX arbitrage
When the same token trades at different prices across two DEXs – say, ETH is cheaper on Uniswap than on SushiSwap – a searcher can buy on the cheaper venue and sell on the more expensive one within a single atomic transaction, eliminating price risk.
This is the most common MEV strategy. Because it corrects price discrepancies across markets, DEX arbitrage is generally considered a net positive for DeFi. It keeps prices efficient across liquidity pools.
According to EigenPhi, arbitrage transactions generated $3.37 million in profit over 30 days in September 2025 alone.
Liquidation rewards
DeFi lending protocols (Aave, Compound, etc.) require borrowers to maintain a minimum collateral ratio. When a borrower's collateral falls below the threshold, typically due to a price drop, the protocol allows anyone to liquidate that position and collect a reward (usually a percentage of the seized collateral).
Searchers run bots that monitor collateral health ratios in real time. The moment a position becomes liquidatable, multiple bots race to submit the liquidation transaction first. Whoever wins captures the fee.
Like arbitrage, liquidations serve a structural function. They prevent lending protocols from accumulating bad debt.
Sandwich attacks
As shown in the example above, a sandwich attack places the searcher's transactions around a victim's large swap, buying before and selling after, to profit from the price impact the victim's trade creates.
The data from 2025 shows the scale of this activity:
- Over 95,000 sandwich attacks were recorded on Ethereum between November 2024 and October 2025
- Monthly attack volume consistently ranged between 60,000 and 90,000 attacks
- Sandwich attacks caused approximately $60 million in annual losses for traders
- Despite the volume, attackers' actual profit margin was only ~5% – the majority of the extracted value went to block builders via gas fees
Unlike arbitrage or liquidations, sandwich attacks extract value directly from users without providing any systemic benefit.
Backrunning profitable trades
Backrunning places a searcher's transaction immediately after a target transaction to capture the state change it creates without touching the victim's trade itself.
A typical example: a large swap on Uniswap moves the price of an asset significantly, creating a price gap between that pool and another. A backrunning bot submits an arbitrage transaction in the same block, directly after the swap, to close that gap and pocket the difference.
Unlike sandwich attacks, pure backrunning doesn't harm the original trader. It simply captures value that would otherwise disperse across the market. This is why backrunning is the basis for MEV-Share programs, where users can opt into having their trades backrun in exchange for receiving a portion of the profits.
Are MEV Searchers Good or Bad for Ethereum?
| In short: MEV searching has both stabilizing and destabilizing effects on Ethereum. The answer depends on which strategy is being run and who bears the cost. |
Where searchers benefit the network:
- DEX arbitrage: Keeps token prices consistent across liquidity pools, reducing the spread ordinary users experience when swapping. Without arbitrage bots, price gaps between pools would persist longer and widen.
- Liquidations: Keep lending protocols solvent. When a borrower's collateral falls below the threshold, searchers race to liquidate it, protecting depositors and the protocol itself from bad debt accumulation.
- Validator revenue: MEV is now a meaningful portion of Ethereum validator income. MEV-Boost lifts validator APY from roughly 4% to around 5.69%. This makes staking more economically attractive, which supports network security.
- Backrunning: Corrects pricing inefficiencies after large trades without directly harming the original trader.
Where searchers impose costs:
- Sandwich attacks: Extract value directly from users in the form of invisible slippage. Traders pay a worse execution price without knowing it – roughly $60 million in annual losses on Ethereum as of 2025.
- Block builder centralization: The infrastructure arms race in MEV has contributed to a small number of builders dominating block production, creating potential censorship and systemic risk at the block-building layer.
- Latency externalities: Even private relay systems generate spam on certain chains (particularly Solana), as bots fire off high volumes of competing transactions that mostly fail but still consume network resources.
The strategies most profitable for searchers tend to impose the greatest costs on users, while the strategies that benefit Ethereum's ecosystem (arbitrage, liquidations) operate on thin margins and are already highly competitive. The balance between MEV’s extractive and productive forms remains an active design problem for Ethereum protocol developers.
MEV Searcher Infrastructure & Competitive Dynamics
| In short: The barrier to entry in MEV searching is high, and the market is more concentrated than it appears from the outside. |
Core infrastructure requirements:
- Full node access: Searchers need low-latency access to mempool data; running a dedicated node is standard practice
- Simulation engine: Before submitting a bundle, every potential trade must be simulated to verify profitability; a failed bundle still costs gas
- Bundle submission pipeline: Direct integration with Flashbots, MEV-Share, or other builder APIs
- Speed: In competitive strategies like liquidations and sandwich attacks, milliseconds determine winners; co-locating servers near Ethereum validators is common
Searchers pay the majority of their gross revenue to builders and validators.
Data from EigenPhi and ESMA suggest that searchers retain less than 10% of their gross MEV revenue in competitive markets. The rest flows up the supply chain.
This creates a dynamic where the viable professional MEV searcher population is surprisingly small. Most of the well-known consistent profits come from a handful of entities with significant infrastructure advantages.
For example, one entity – jaredfromsubway.eth – was responsible for roughly 70% of all sandwich attacks on Ethereum over the study period tracked by Cointelegraph Research/EigenPhi.
Can anyone become an MEV searcher? Technically yes. The code and tooling are open source. Flashbots publishes documentation, and the Ethereum ecosystem is permissionless. But practically, new entrants are unlikely to be profitable in well-known strategies like DEX arbitrage or liquidations. The more realistic entry point today is through niche strategies or newer chains with less saturated MEV markets.
Sources and Further Reading
- Ethereum.org – "Maximal Extractable Value (MEV)" https://ethereum.org/en/developers/docs/mev/
- Flashbots – "Flashbots Documentation" https://docs.flashbots.net/
- a16z Crypto – "MEV Explained" https://a16zcrypto.com/posts/article/mev-explained/
- EigenPhi – "MEV Research and Data" https://eigenphi.io/
- Extropy Academy – "An Analysis of Arbitrage Markets Across Ethereum and Solana (2025)" https://academy.extropy.io/pages/articles/mev-crosschain-analysis-2025.html
- Cointelegraph Research / EigenPhi — "Sandwich Attacks on Ethereum Have Waned (Nov 2024–Oct 2025)" https://www.tradingview.com/news/cointelegraph:fa12ba092094b:0-exclusive-data-from-eigenphi-reveals-that-sandwich-attacks-on-ethereum-have-waned/
FAQs About MEV Searchers
Aggregators like 1inch or Paraswap reduce slippage by routing across multiple pools, but they don't fully protect against sandwich attacks unless they route through a private mempool or MEV-protected RPC. Some aggregators have added MEV protection explicitly. It's worth checking whether the one you use does.
