Transactions are a low-level primitive that force users to manually navigate liquidity across fragmented chains and protocols like Uniswap and Aave. This exposes them to execution risk, MEV extraction, and constant gas optimization.
account-abstraction-fixing-crypto-ux
Blog
Why the 'Transaction' is an Antiquated Concept in Intent-Centric DeFi
The atomic, single-protocol transaction is a legacy construct from a simpler era. Account Abstraction (EIP-4337) enables multi-step, cross-protocol 'user operations' that fulfill a holistic user intent, rendering the traditional model obsolete.
introduction
THE PARADIGM SHIFT
Introduction
The atomic transaction is a legacy abstraction that fails to capture user goals, creating a fragmented and inefficient DeFi experience.
User intent is the high-level abstraction that separates the 'what' from the 'how'. A user wants to 'earn yield on USDC', not 'approve, swap, bridge, deposit'. Protocols like CowSwap and UniswapX pioneered this for swaps.
The intent-centric architecture inverts the model. Users submit signed declarations of desired outcomes. A solver network (e.g., SUAVE, Anoma) competes to fulfill the intent optimally, abstracting away execution complexity.
Evidence: UniswapX, which outsources swap routing to fillers, now processes over $15B in volume, demonstrating demand for intent-based execution that minimizes costs and maximizes output.
key-trends
FROM EXECUTION TO OUTCOME
The Three Pillars of the Post-Transaction Era
The transaction is a low-level, user-hostile primitive. The future is users declaring desired outcomes, not signing raw calldata.
01
The Problem: The User as the Execution Layer
Users today are their own integrators, manually stitching together protocols like Uniswap, AAVE, and Lido. This is a UX failure and a systemic risk.
- Manual Slippage Management: Users overpay for safety, losing ~1-5% per complex trade.
- Gas Auction Hell: Competing in public mempools against MEV bots.
- Fragmented Liquidity: Requires bridging, wrapping, and multiple approvals across chains.
~$1.3B
MEV Extracted (2023)
5+ Steps
Avg. DeFi Action
02
The Solution: Intent-Based Abstraction
Users submit signed declarations of a desired end-state (e.g., 'Get me 1 ETH on Arbitrum for <$3400'). A network of solvers competes to fulfill it optimally.
- UniswapX & CowSwap: Pioneered this model for swaps, aggregating liquidity and MEV protection.
- Across & Socket: Use intents for cross-chain bridging, abstracting away complexity.
- Anoma & SUAVE: Architecting generalized intent-centric networks and block spaces.
-90%
User Steps
~500ms
Quote Latency
03
The Enforcer: Cryptographic Commitments
Intents require a cryptographic guarantee that the promised outcome is delivered. This shifts trust from individual protocols to a verifiable settlement layer.
- ZK Proofs: Prove correct execution off-chain before on-chain settlement (e.g., zkSync, Starknet).
- Optimistic Verification: Use fraud proofs and bonding, as seen in Optimism and Arbitrum rollups.
- Atomic Settlement: Ensures the entire intent either succeeds or reverts, preventing partial failures.
100%
Execution Guarantee
$0
Revert Cost
WHY THE EOA MODEL IS OBSOLETE
Transaction vs. User Operation: A Technical Breakdown
A first-principles comparison of the legacy EOA transaction model versus the intent-centric User Operation, as defined by ERC-4337 and utilized by UniswapX, CowSwap, and Across.
| Core Feature / Metric | Legacy EOA Transaction | ERC-4337 User Operation | Implication for Intent-Centric Design |
|---|---|---|---|
Architectural Primitive | Atomic State Transition | Declarative Intent Object | User specifies what, not how |
Signer Abstraction | Enables smart contract wallets (Safe, Biconomy) and social recovery | ||
Gas Payment Token | Native Chain Token Only | Any ERC-20 via Paymasters | Enables sponsorship and gasless onboarding |
Execution Atomicity | Single Call, All-or-Nothing | Bundled Multi-Op via Bundlers | Enables complex cross-chain intents (LayerZero, Socket) |
Fee Market Design | Priority Gas Auction (PGA) | Auctions for Bundler & Paymaster | Decouples inclusion from execution, reduces MEV surface |
Typical Gas Overhead | 21,000 base + execution | ~42,000 base + execution | ~2x cost for account abstraction features |
Standardization | Ethereum Yellow Paper | ERC-4337 (No consensus change) | Faster iteration, deployable on any EVM chain |
MEV Resistance | Low (Frontrunning, Sandwiches) | High (via SUAVE, CowSwap solvers) | Searchers compete on outcome, not transaction order |
deep-dive
THE PARADIGM SHIFT
From Atomic Command to Holistic Intent
The transaction is a low-level, rigid instruction that fails to capture user goals, creating the complexity that intent-centric architectures solve.
The transaction is a liability. It forces users to specify low-level how (e.g., swap path, gas price, slippage) instead of the high-level what (e.g., 'get 1 ETH for the best rate'). This creates a usability barrier and exposes users to MEV.
Intent decouples declaration from execution. A user submits a signed declarative goal, and a solver network (e.g., UniswapX, CowSwap) competes to fulfill it optimally. This shifts complexity from the user to the network.
Atomicity is the bottleneck. A transaction's success depends on a single, sequential state change. An intent's fulfillment can be asynchronous and multi-chain, leveraging specialized protocols like Across and LayerZero for cross-domain settlement.
Evidence: UniswapX, which routes orders via off-chain solvers, has settled over $5B in volume, demonstrating that users delegate execution for better outcomes.
protocol-spotlight
THE INTENT REVOLUTION
Architects of the Post-Transaction Stack
Transactions are a low-level, user-hostile primitive. The next stack abstracts them away, letting users declare what they want, not how to do it.
01
The Problem: The User as an Executor
Users must manually navigate liquidity across 50+ chains, sign dozens of transactions, and pray for MEV protection. This is a full-time job.
- ~$1.3B in MEV extracted annually from naive users.
- >60% of DeFi users have lost funds to a bad route or slippage.
$1.3B+
MEV Extracted
>60%
User Error Rate
02
The Solution: Intents as Declarative Commands
Instead of signing a transaction, a user signs a statement of desired outcome (e.g., "Swap X for Y at best rate"). Specialized solvers (CowSwap, UniswapX, Across) compete to fulfill it.
- ~20-30% better prices via solver competition.
- Zero-gas experiences for users.
~30%
Price Improvement
0 Gas
User Cost
03
The Infrastructure: Solver Networks & SUAVE
Intent fulfillment requires a new execution layer. Solvers are the new miners. Flashbots' SUAVE aims to be a decentralized mempool and solver marketplace.
- Sub-second execution latency for cross-chain intents.
- Creates a $500M+ market for solver fees.
<1s
Latency
$500M+
Market Size
04
The Settlement: Intents Kill the Bridge
Traditional atomic bridges (LayerZero, Axelar) are transaction-based. Intent-based routing uses any available liquidity, treating all chains as one fragmented state. Across and Socket are early leaders.
- >70% cheaper than canonical bridging.
- Unlocks $10B+ in stranded liquidity.
-70%
Cost vs Bridge
$10B+
Liquidity Unlocked
05
The Risk: Centralization of Solvers
If a few solver entities (e.g., large market makers) dominate, they become the new MEV-extracting order flow auction. This recreates the Wall Street problem.
- Top 3 solvers can control >60% of intent flow.
- Requires verifiable execution proofs (ZK).
>60%
Flow Control Risk
ZK Proofs
Mitigation
06
The Endgame: Autonomous Agents & You
Your wallet becomes an agent that continuously fulfills complex intents ("Maintain 10% yield exposure"). The transaction disappears entirely into persistent state management.
- 0 manual interventions for portfolio management.
- Turns DeFi into a set-it-and-forget-it system.
0
Manual Txns
24/7
Autonomous
counter-argument
THE TRADEOFF
The Counter-Argument: Complexity and Centralization
Intent-centric architectures shift complexity from users to a new class of centralized infrastructure, creating systemic risk.
Intent solvers become centralized bottlenecks. The specialized execution logic required to fulfill complex intents creates high capital and technical barriers, leading to solver oligopolies like those seen in CowSwap and UniswapX.
User sovereignty is an illusion. Delegating transaction construction to a third-party solver reintroduces trust assumptions; the user's outcome depends entirely on the solver's honesty and capability, a regression from non-custodial wallet principles.
The MEV attack surface expands. Solvers compete in a zero-sum game for user surplus, incentivizing sophisticated front-running and back-running strategies within the intent fulfillment process itself, beyond simple transaction ordering.
Evidence: The leading intent-based DEX aggregator, CowSwap, relies on a small set of professional solvers; its centralized order matching is the core mechanism that enables its gas-free, MEV-resistant trades.
takeaways
WHY TRANSACTIONS ARE OBSOLETE
TL;DR for Busy Builders
The transaction is a low-level, user-hostile primitive. Intent-centric architectures abstract it away, letting users declare what they want, not how to do it.
01
The Problem: The User is the Executor
Today, a user must manually construct a precise sequence of calls across fragmented liquidity pools, bridges, and DEXs. This is slow, expensive, and fails often.\n- ~30% failure rate for complex cross-chain swaps\n- Wasted gas on failed frontrun transactions\n- Lost MEV extracted from predictable user behavior
~30%
Fail Rate
$1B+
MEV Extracted
02
The Solution: Declarative Intents & Solvers
Users submit a signed intent (e.g., 'Swap 1 ETH for best price of ARB on Arbitrum'). A competitive network of solvers (like in CowSwap or UniswapX) finds the optimal execution path.\n- Permissionless solver competition drives better prices\n- Atomic composability across chains via Across or LayerZero\n- Guaranteed execution or revert, no partial failures
10-50%
Price Improvement
~500ms
Solver Latency
03
The New Stack: SUAVE & Anoma
Specialized infrastructure is emerging to make this viable. SUAVE (by Flashbots) is a decentralized block builder and preference mempool for intents. Anoma provides a full-stack architecture for intent matching and privacy.\n- Decentralized sequencing removes centralized solver risks\n- Encrypted mempools prevent frontrunning (see shutter)\n- Native cross-chain intent settlement
0
Frontrunning
Chain-Agnostic
Scope
04
The Endgame: From Wallets to Agents
Wallets like Rabby and Safe are becoming intent-aware. The next step is autonomous agents that manage capital based on high-level policies (e.g., 'Maintain 20% stablecoin exposure'). The transaction disappears entirely.\n- ERC-4337 Account Abstraction enables sponsored intents\n- Agentic wallets auto-route for best yield/price\n- User gets outcome, not receipts
100x
UX Simplicity
Always-On
Capital Efficiency
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall