Shared sequencers solve for liveness by decoupling block production from a single operator, preventing censorship and downtime. This creates a neutral, resilient base layer for transaction ordering, similar to how EigenLayer secures Actively Validated Services (AVS).
account-abstraction-fixing-crypto-ux
Blog
Why Shared Sequencers Without AA Are Incomplete
Shared sequencers solve ordering but miss the user. This analysis argues that without native Account Abstraction, they fail to capture the true value of a modular stack: seamless cross-rollup sessions and abstracted fees.
introduction
THE MISSING PIECE
Introduction
Shared sequencers are a necessary but insufficient upgrade for user-centric rollups.
Without Account Abstraction (AA), the user experience remains broken. A shared sequencer alone cannot sponsor gas, batch operations, or enable session keys. Users still face the friction of managing native gas and signing every action, which EIP-4337 bundles and smart accounts eliminate.
The integration is a protocol design imperative. A shared sequencer must be built with native AA support, like Starknet's sequencer and zkSync's native account abstraction, to enable intent-based flows and abstract chain-specific complexity at the sequencing layer itself.
key-trends
WHY SHARED SEQUENCERS NEED ACCOUNT ABSTRACTION
The Modular Stack's Missing Link
Shared sequencers solve for block space, but ignore the user. Without AA, they create a fragmented, high-friction experience that undermines the modular promise.
01
The Problem: Fragmented UX in a Modular World
A user bridging from Arbitrum to Base via Espresso's shared sequencer still needs separate gas tokens, seed phrases, and approvals for each chain. This kills cross-chain composability.
- User Burden: Managing 5+ native tokens for a multi-chain portfolio.
- Developer Burden: Building for the lowest common denominator of wallet support.
- Result: Adoption is gated by the most primitive user experience in the stack.
5+
Tokens to Manage
-90%
Composability
02
The Solution: AA as the Universal Settlement Layer
Account Abstraction (via ERC-4337 or native implementations) makes the user's smart account the portable identity. The shared sequencer (like Astria, Radius) orders transactions, but AA wallets handle gas and session keys.
- Portable Security: A single smart account works across all rollups using the sequencer.
- Gas Abstraction: Pay in any token; sponsorships via Paymasters become trivial.
- Atomic Composability: Bundle actions across sovereign chains into one user operation.
1
Account for All Chains
Any Token
Pay Gas With
03
The Killer App: Intents Meet Shared Sequencing
The real synergy is enabling intent-based architectures (like UniswapX, CowSwap) across a rollup ecosystem. A shared sequencer provides a neutral ordering venue, while AA enables expressive user commands.
- User Expresses 'What': "Get me the best price for 1 ETH across Arbitrum, Optimism, Base."
- Solvers Compete: Solvers on Across, LI.FI, 1inch submit bundles to the shared sequencer.
- AA Executes 'How': The user's smart account validates and pays for the winning bundle atomically.
~500ms
Cross-Chain Latency
10-30%
Better Price Execution
04
Entity Spotlight: How Fuel & Sui Nail This Combo
These L1s aren't modular but demonstrate the principle: native account abstraction is core infrastructure. Fuel's UTXO model with predicates and Sui's object model with sponsored transactions show that sequencing and AA must be co-designed.
- Fuel's Predicates: Enable complex, gasless transaction logic before execution.
- Sui's SponsoredTx: Allows any entity to pay for a user's gas, baked into the protocol.
- Lesson for Rollups: Baking AA into the settlement layer (like a shared sequencer's design) is non-negotiable.
Native
Protocol Feature
$0
User Gas Cost
05
The Economic Flaw: Sequencer Revenue Without AA
A shared sequencer capturing MEV and fees from dozens of rollups creates a powerful, extractive middleman. AA with decentralized paymaster networks can redistribute value.
- Without AA: Sequencer profits from user friction (failed tx, inefficiency).
- With AA: Paymasters (like Biconomy, Pimlico) can subsidize gas, capturing value by enabling new use cases.
- Realignment: Value flows to user acquisition and retention, not just block space ordering.
>50%
Failed Tx Rate
New Biz Model
Paymaster Stakes
06
The Verdict: Build the Stack Backwards from the User
The modular thesis fails if the end-user experience is an afterthought. The stack must be: User Intent -> Smart Account (AA) -> Shared Sequencing -> Execution Layer.
- Shared Sequencers (Espresso, Astria) are infrastructure for rollups.
- Account Abstraction is infrastructure for humans.
- Omitting AA means building a high-performance engine with a horse-drawn carriage interface.
User-First
Design Mandate
Incomplete
Without AA
thesis-statement
THE ARCHITECTURAL SHIFT
The Core Argument: Sequencing is a Commodity, Sessions are the Product
Shared sequencers that only aggregate transactions are selling a commodity, while the real product is the programmable user session.
Sequencing is a commodity because its core function—ordering transactions—is a solved problem. The market will converge on the cheapest, fastest provider, just as it did with block builders like Flashbots and MEV-Boost on Ethereum. A shared sequencer without a session layer is just a decentralized mempool.
Sessions are the product that captures value. A session is a programmable context for a user's chain of actions, like a multi-step DeFi trade or a game interaction. This is where protocols like Ethereum's ERC-4337 and Starknet's native account abstraction create sticky, monetizable relationships.
Shared sequencers without AA are incomplete. They offer atomic composability for a single block, but fail to guarantee execution across multiple blocks or chains. A user swapping on Uniswap and bridging via Across in one flow needs a session, not just sequencing.
Evidence: The EIP-7702 proposal and the growth of intent-based systems (UniswapX, CowSwap) prove the demand is for abstracted, multi-step user journeys. Sequencing is the pipe; the session is the service flowing through it.
INFRASTRUCTURE VALUE LEAKAGE
The Value Capture Gap: Shared Sequencer vs. AA-Enabled Stack
Compares the economic and technical completeness of a standalone shared sequencer versus one integrated with Account Abstraction (AA) infrastructure, highlighting where value is captured or lost.
| Critical Capability | Standalone Shared Sequencer (e.g., Espresso, Astria) | AA-Enabled Stack (e.g., Biconomy, Safe, Pimlico + Shared Sequencer) | Traditional L1/L2 Sequencer |
|---|---|---|---|
Native User Fee Capture | 0% |
| 100% (but user pays directly) |
MEV Capture & Redistribution | Sequencer-level only | User-level via intents & bundling | Validator/Proposer-level only |
Cross-Chain User Experience | |||
Transaction Sponsorship (Gasless) | |||
Session Keys / Batch Auth | |||
Modular Security Budget | Sequencer fees only | Sequencer + Paymaster + Bundler fees | Sequencer/Validator fees only |
Protocol Revenue Streams | 1 | 3+ (Sequencing, Bundling, Paymaster) | 1 |
Developer Abstraction | Execution only | Full stack: Account, Gas, UX | None |
deep-dive
THE ARCHITECTURAL GAP
Why AA is the Missing Primitives for Shared Sequencing
Shared sequencers without account abstraction are architecturally incomplete, failing to deliver on the core promise of a unified user experience across rollups.
Sequencer-level UX is insufficient. A shared sequencer like Espresso or Astria provides atomic cross-rollup transaction ordering. This solves for developers and liquidity, but the user still manages separate wallets, pays separate gas fees, and signs separate transactions for each chain. The user experience remains fragmented at the account layer.
AA enables intent-driven sequencing. With ERC-4337 smart accounts, a user submits a single signed intent (e.g., 'swap ETH on Arbitrum for USDC on Optimism'). The shared sequencer's atomic ordering is the execution layer, but the smart account's bundler and paymaster become the coordination layer, handling gas abstraction and cross-chain settlement automatically via bridges like Across or LayerZero.
Shared sequencing without AA is a half-bridge. It replicates the modular data availability model of Celestia but for transaction ordering. Without AA's smart accounts to interpret and execute complex intents, it's just a faster, more coordinated version of the existing multi-chain mess. The end-user abstraction is missing.
Evidence: Protocols like UniswapX and CowSwap demonstrate the demand for intent-based, gas-abstracted execution. A shared sequencer network that natively integrates with AA bundlers (e.g., via RIP-7560) will capture this demand by making cross-rollup interactions feel like a single transaction, not a bridge hop.
counter-argument
THE COUNTER-ARGUMENT
Steelman: "AA is an App-Layer Concern"
A valid critique argues that account abstraction is a user-facing feature best handled by individual applications, not a shared infrastructure layer.
The core argument is correct: Account abstraction is fundamentally about user experience. Features like sponsored transactions, session keys, and social recovery are product decisions. A dApp like dYdX or Uniswap should own this logic to differentiate.
Shared sequencers add overhead: Forcing all apps to use a single AA-enabled sequencer creates unnecessary complexity. It's akin to mandating every website use the same login widget. Apps lose fine-grained control over their gas sponsorship policies and fee logic.
The standard already exists: ERC-4337 is the dominant standard for smart contract wallets. It operates at the application layer, independent of the underlying sequencer. This separation of concerns is intentional and efficient. Stackup, Biconomy, and Safe build on this model.
Evidence: The Ethereum mainnet processes millions of ERC-4337 UserOperations without any L2 sequencer modifications. The bundler and paymaster infrastructure proves AA is a successful app-layer primitive.
protocol-spotlight
THE MISSING LINK
Who's Building the Complete Stack?
Shared sequencers solve MEV and liveness, but without account abstraction, they fail to deliver a seamless user experience.
01
The Problem: Gas Abstraction
Users still need native gas tokens for every new chain. This kills cross-chain UX.
- Breaks flow for non-crypto-native users.
- Adds friction requiring multiple token balances.
- Limits adoption to power users only.
~80%
Drop-off Rate
5+
Tokens Needed
02
The Solution: Paymasters & Bundlers
The complete stack integrates a paymaster to sponsor gas and a bundler to execute user operations.
- Sponsor transactions in any ERC-20 token (e.g., USDC).
- Batch operations for ~30-40% lower effective gas costs.
- Enable session keys for one-click interactions.
$0
Upfront Gas
1-Click
User Action
03
The Architecture: Intent-Based Routing
Abstracted accounts allow users to express what they want, not how to do it. The sequencer finds the best path.
- Integrates with solvers like UniswapX and CowSwap.
- Optimizes for cost/speed across L2s via bridges like Across and LayerZero.
- Captures MEV for user rebates, not extractors.
10x
Better Prices
~500ms
Solver Latency
04
The Winner: Stack + AA
The dominant shared sequencer will be the one that bakes AA into its core, not as an afterthought.
- Native 4337 support for zero-friction onboarding.
- Unified liquidity layer across all connected rollups.
- Protocol-owned flow becomes the default user experience.
$10B+
Potential TVL
100M+
User Target
takeaways
WHY SHARED SEQUENCERS WITHOUT AA ARE INCOMPLETE
TL;DR for CTOs and Architects
Shared sequencers solve for atomic composability and MEV capture, but without Account Abstraction, they fail to deliver the fundamental UX shift required for mainstream adoption.
01
The Atomic Composability Mirage
Shared sequencers like Astria or Espresso enable cross-rollup atomic bundles, but standard EOAs cannot sign transactions for multiple chains or dApps in one go. Without AA's session keys or sponsored transactions, users face a UX cliff where atomicity is a backend feature they cannot access.
- User Reality: Still signing 5+ TXs for a simple cross-DEX arbitrage.
- Architectural Gap: Atomic sequencing layer ≠atomic user experience.
0
User-Atomic TXs
5+
Avg. Signatures
02
MEV Redistribution is Stuck at the Wallet
A shared sequencer's MEV auction (e.g., Flashbots SUAVE vision) can optimize and redistribute value, but it hits a wall at the user's EOA. Without AA's payment abstraction, you cannot practically implement fee rebates, transaction bundling subsidies, or direct MEV share payouts to the end-user.
- Current Flow: MEV profit → Proposer/Protocol Treasury.
- AA-Enabled Flow: MEV profit → User's gas fee or wallet balance.
>99%
MEV to Users
ERC-4337
Required Std
03
The Interoperability Bottleneck
Shared sequencing creates a unified liquidity and state layer, but EOAs fragment user identity and assets across rollups. AA's smart accounts are portable, chain-agnostic identities that turn shared sequencing's technical interoperability into a usable product. Without it, you're building a highway with no on-ramps.
- Key Entity: Polygon AggLayer and zkSync Hyperchains need AA for seamless portability.
- Result: Shared L2 state with siloed L1-style user accounts.
1
Portable Identity
N
Chain Fragmentation
04
Security Model Regression
Decentralizing the sequencer set improves liveness but does nothing for account security. AA introduces social recovery, multi-sig policies, and transaction guards at the account level. A shared sequencer without AA offers a more robust chain but leaves users with the same vulnerable, irrecoverable seed phrases.
- Contradiction: Enterprise-grade sequencing with consumer-grade wallet risk.
- Solution: AA smart accounts as the mandatory security interface.
0
Native Recovery
ERC-4337
Security Layer
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