Smart accounts break existing indexers. Indexers like The Graph and Covalent are built for Externally Owned Accounts (EOAs), which treat a wallet as a single key. Smart accounts, powered by ERC-4337 and ERC-6900, separate logic from ownership, creating a multi-signature, upgradeable, and modular identity that current indexing architectures cannot track.
Why Smart Account Indexing Is the Next Infrastructure Battle
Smart accounts break traditional indexing. Querying UserOp history and account state requires new paradigms, sparking a data war between The Graph, Subsquid, and new entrants. This is the critical infrastructure layer for the next billion users.
Introduction
Smart account adoption is creating a new, critical data layer that existing indexers cannot parse.
The battle is for user intent. Legacy indexing captures what happened (e.g., 'Wallet 0x... sent 1 ETH'). The new frontier is indexing why it happened—the user's bundled intent across actions like a token swap on Uniswap, a cross-chain bridge via LayerZero, and a subscription payment, all executed atomically by a paymaster.
Infrastructure follows application demand. The rise of account abstraction toolkits from Stackup, Biconomy, and ZeroDev forces the stack to evolve. Just as The Graph emerged for DeFi, a new indexing standard is required for the permissionless, intent-driven transactions of smart accounts, making this the next critical infrastructure layer.
The Core Argument
Smart account adoption will shift the infrastructure war from raw RPC performance to the quality of the data pipeline.
Smart accounts invert the data model. Externally Owned Accounts (EOAs) are simple state objects; smart accounts are complex, event-emitting programs. Indexing EOAs means tracking native transfers and standard contract calls. Indexing smart accounts requires parsing custom events from ERC-4337 EntryPoint transactions and interpreting internal state changes, a fundamentally harder problem.
The query layer becomes the bottleneck. Today's infrastructure race is about RPC speed (Alchemy, QuickNode). Tomorrow's is about query expressiveness. Developers need to ask, "Show me all UserOperations for this account that failed due to a specific signature scheme" or "Find all accounts that used a specific Session Key module this week." Generic indexers like The Graph struggle with this custom logic.
Evidence: The Ethereum Attestation Service (EAS) and ERC-4337 Bundlers already generate non-standard data streams that existing indexers cannot natively parse. Teams building account abstraction wallets like Safe{Wallet} and Biconomy are forced to build custom indexing infra, which is the new moat.
The Data Tsunami: Three Unavoidable Trends
The shift to smart accounts (ERC-4337) and intent-based architectures doesn't just change wallets—it explodes the data layer, creating a new bottleneck and a trillion-dollar opportunity for the indexer that can tame it.
The Problem: The Graph Can't See Your Wallet
ERC-4337 UserOperations are off-chain mempool objects, invisible to traditional RPC/indexing. The Graph's subgraph model fails for real-time, stateful account data.
- UserIntent Discovery: Indexers must parse and track ephemeral intents before they're on-chain.
- Bundler & Paymaster Orchestration: Real-time visibility into which bundler is winning which auctions is critical for MEV capture and reliability.
- Cross-Chain State: A user's smart account state is fragmented across L2s; no indexer provides a unified view.
The Solution: Intent-Aware Indexing Layer
Next-gen indexers like Goldsky and Covalent are pivoting to stream UserOperations and account abstraction events. This isn't about historical queries; it's about real-time state for dApps and wallets.
- Predictive Pre-Fetching: Indexers will cache likely next states based on intent patterns, slashing latency for dApps.
- Bundler Performance Analytics: Providing data feeds on bundler success rates and costs becomes a core service for wallet providers.
- The New 'RPC': The indexer becomes the primary data source for smart account dashboards, abstracting the chaos of the mempool.
The Stakes: Who Owns the User Graph?
The entity that indexes the smart account layer owns the definitive map of user behavior, intent, and capital flow. This is the most valuable dataset in crypto.
- Monetization Levers: Sell bundled data to DeFi protocols (e.g., Uniswap, Aave) for intent-based routing and risk modeling.
- Protocol Criticality: Wallets (Safe, Coinbase Smart Wallet) and intent solvers (UniswapX, Across) will be primary enterprise clients, creating $100M+ annual revenue streams.
- Vertical Integration: Expect wallet providers (e.g., Rainbow) to build their own indexers to capture this value and ensure data sovereignty.
Indexing Paradigm Shift: EOA vs. Smart Account
Compares the technical and economic implications of indexing for traditional Externally Owned Accounts (EOAs) versus emerging Smart Account (ERC-4337) architectures.
| Indexing Dimension | EOA (Traditional) | Smart Account (ERC-4337) | Infrastructure Impact |
|---|---|---|---|
Primary Data Source | Raw transaction mempool, on-chain events | UserOperation mempool, on-chain events, off-chain signatures | Requires monitoring new mempool (P2P/alt mempool) and bundler endpoints |
State Complexity | Single nonce, direct balance | Multiple nonces (key, batch), deposit in EntryPoint, deferred execution | Indexers must track abstracted state; requires simulating UserOperations |
Query Pattern | Who owns this NFT? What are my token balances? | What sessions/allowances does this account have? What actions are batched? | Shifts from simple state queries to intent and permission introspection |
Latency for Fresh Data | < 1 sec (mempool) | 1-12 sec (bundler queue simulation + inclusion) | Real-time indexing now depends on bundler strategy and gas auctions |
Fee Capture Surface | ~100% goes to L1/L2 sequencer | ~70% to sequencer, ~30% to bundlers/paymasters | Indexers must become bundlers or partner to capture new fee layer |
Security Model for Indexers | Validate digital signature | Validate ERC-4337 rules, signature aggregation, paymaster sponsorship | Increased computational load; risk of indexing invalid UserOperation bundles |
Market Examples | The Graph, Covalent, Etherscan | Stackup, Alchemy Account Kit, Biconomy, Etherspot | Incumbents must adapt; new entrants building native SA stacks |
The New Stack: Mapping the Contenders
Smart account proliferation creates a new data access bottleneck, making indexing a critical infrastructure battleground.
Smart accounts fracture user state. Unlike Externally Owned Accounts (EOAs), a user's assets and interactions are distributed across a dynamic set of smart contract addresses, breaking traditional indexers like The Graph.
The new indexer must be stateful. It must track user intent, session keys, and modular account components, requiring a paradigm shift from event-based to state-based querying.
This creates a winner-take-most market. The first indexer to reliably serve complex queries for protocols like Safe, Biconomy, and ZeroDev will capture developer mindshare and become a core primitive.
Evidence: The Graph's subgraphs fail for cross-chain smart accounts, creating demand for solutions like Airstack's intent-aware APIs or custom indexers from Pimlico and Stackup.
Contender Analysis: Strengths & Fault Lines
As ERC-4337 adoption grows, the ability to query and analyze smart account activity becomes a critical infrastructure layer, creating a new battleground for data providers.
The Graph's Subgraph Bottleneck
The Graph's subgraph model, built for contract events, struggles with the dynamic, user-centric nature of smart accounts. It forces indexers to model complex user operation flows as monolithic subgraphs, creating friction for developers.
- High Latency: Indexing user operation mempools and bundler actions introduces ~5-10 second delays for real-time state.
- Developer Overhead: Requires custom subgraph deployment per account logic, unlike simple contract queries.
- Cost Inefficiency: Paying for full-chain indexing when most queries target specific account abstractions.
Pimlico's Bundler-First Advantage
As a leading bundler infrastructure provider, Pimlico has a natural edge. It can index user operations and account states directly from its mempool and execution flow, offering real-time data as a service.
- Native Data Access: Sees 100% of user ops flowing through its network before they hit a blockchain.
- Real-Time Analytics: Can provide ~500ms latency on account state changes and transaction success rates.
- Monetization Path: Bundling service creates a built-in customer base for premium indexing APIs, mirroring the Alchemy playbook.
The Specialized Indexer Thesis (e.g., Goldsky, Nxyz)
Next-gen indexing platforms like Goldsky and Nxyz are built for real-time, application-specific data streams. They can treat smart account activity as a first-class event stream, bypassing traditional blockchain indexing constraints.
- Streaming Architecture: Delivers sub-second updates on account state via WebSockets, crucial for wallet UIs.
- Cross-Chain Aggregation: Can index ERC-4337 activity across Ethereum, Polygon, Arbitrum into a unified API.
- Fault Line: Must establish direct relationships with bundler networks or RPC providers to access the full mempool, creating a dependency.
RPC Giants as Sleeping Giants (Alchemy, QuickNode)
Enterprise RPC providers already have the deepest node infrastructure and customer relationships. Adding smart account indexing is a logical feature extension, leveraging their existing data pipelines.
- Instant Scale: Can deploy indexing to their entire $100M+ enterprise client base overnight.
- Full-Stack Data: Combine mempool, execution traces, and state diffs for unparalleled debug and analytics.
- Strategic Risk: May treat this as a checkbox feature rather than a core innovation, leaving room for agile specialists. Their move will commoditize basic indexing.
The Wallet Data Monopoly Play
Wallet providers like Safe, Rainbow, or Coinbase Wallet have the ultimate advantage: they control the client interface and can index the user's private transaction history and intent data directly.
- Proprietary Intent Data: Access to failed user ops, gas preferences, and cross-app session flows that public indexers cannot see.
- Vertical Integration: Can bundle indexing with wallet-as-a-service offerings for dApps.
- Fragmentation Risk: Creates walled gardens of user data, antithetical to Web3's open data ethos but potentially the most valuable.
The Interoperability Protocol Endgame (LayerZero, CCIP)
Omnichain messaging protocols are evolving into general-purpose state synchronization layers. They could index and attest to smart account state across any chain, becoming the canonical source for cross-chain account status.
- Universal State Proofs: Use light clients and attestations to verify account state on Chain A from Chain B.
- Native Cross-Chain Queries: A dApp could query a user's combined Safe balance across 10 chains in one call.
- Existential Threat: If successful, they bypass the need for traditional indexers for cross-chain account data, absorbing that market.
The Bear Case: Is This Just a Feature?
Smart account indexing risks being commoditized as a protocol-level feature, not a standalone business.
Indexing is a commodity. The core function—querying on-chain state—is a solved problem. The Graph and Subsquid already provide generalized indexing; smart accounts add a new data schema, not a new technical paradigm.
Protocols will internalize it. Major L2s like Arbitrum and Optimism will bake native account abstraction explorers into their block explorers to improve developer UX, mirroring how Etherscan dominates EVM chain data.
The moat is data standardization. Without universal standards like ERC-4337 for user operation indexing, each smart account provider (Safe, Biconomy, ZeroDev) creates a fragmented data landscape that resists aggregation.
Evidence: The RPC market demonstrates this path. Despite specialized providers like Alchemy and QuickNode, the largest L1s and L2s now operate their own public RPC endpoints, eroding the standalone market.
Critical Risks & Failure Modes
As ERC-4337 and smart accounts move from testnet to mainnet, the infrastructure for indexing and querying their complex state becomes a critical, non-obvious bottleneck.
The Graph's Looming Obsolescence
Subgraphs are built for simple EOAs and token transfers, not the multi-op, session key, and policy-driven state of smart accounts. Querying a user's "effective permissions" or "spendable balance" across chains becomes a graph traversal nightmare.
- Sub-second latency for complex intent resolution is impossible with current architectures.
- ~80% of smart account logic is off-chain validation, which subgraphs cannot index.
Centralized RPC Catastrophe
Teams default to centralized RPC providers (Alchemy, Infura) for eth_call simulations, creating a single point of failure and censorship. The entire user experience of a smart account wallet depends on this opaque, rent-seeking layer.
- A single RPC outage can brick onboarding for millions of abstracted users.
- Providers see all user intent pre-execution, a massive privacy leak and MEV opportunity.
The Verifier's Dilemma
Indexers must cryptographically verify Paymaster signatures, policy compliance, and gas sponsorship logic. This shifts compute burden from L1 to the indexer, requiring a new class of verifiable compute proofs to prevent fraudulent state reporting.
- Without proofs, you're trusting $10B+ in sponsored gas to an indexer's honesty.
- This creates a direct competitive moat for teams like EigenLayer and AltLayer providing AVS services.
Cross-Chain State Fragmentation
A user's smart account has different permissions and balances on Ethereum, Arbitrum, and Base. A unified view requires a cross-chain indexer that understands account abstraction semantics, not just bridged assets. Without this, wallets show wrong balances and failed transactions.
- LayerZero, Axelar, Wormhole become critical but must index beyond token balances.
- This fragmentation doubles infrastructure costs for application developers.
MEV Extends to the Index Layer
The indexer that first sees a user's batched transaction intent can extract MEV by frontrunning, sandwiching, or censoring specific ops. This turns infrastructure into a predatory financial service.
- Flashbots SUAVE aims to solve this for blocks, but not for pre-chain user intent.
- Indexers become trusted intermediaries, negating crypto's trustless promise.
Solution: Intent-Centric Indexing Protocols
The winning stack will index based on user intent and smart account semantics, not raw logs. Think The Graph for ERC-4337, built with ZK proofs for verification and P2P networks for censorship resistance.
- Espresso Systems for decentralized sequencing data.
- Hyperbolic for proving RPC responses.
- This is the next $1B+ infrastructure vertical after RPCs and oracles.
The 24-Month Outlook
Smart account proliferation will shift the infrastructure war from raw execution to sophisticated data indexing and intent resolution.
Smart accounts fragment user state. Externally Owned Accounts (EOAs) concentrate activity on a single address; ERC-4337 accounts and ERC-6900 modular accounts scatter logic, assets, and session keys across a dynamic graph. This breaks existing indexers like The Graph, which are built for monolithic contract queries.
The new battle is for the 'user graph'. Indexers must now track cross-chain intents, modular plugin permissions, and gas sponsorship meta-transactions. This creates a winner-take-most market for the first service that provides real-time, composable user state, similar to how Google indexes the web.
Intent-centric architectures demand new primitives. Projects like UniswapX and Across already separate declaration from execution. The winning indexer will not just log transactions but resolve and rank competing intent pathways, becoming the oracle for user preference.
Evidence: The Graph's subgraphs fail for cross-chain smart account activity. Emerging players like Goldsky and Covalent are building custom pipelines for this, but no standard exists. The protocol that defines the indexing standard for ERC-4337 will capture the market.
TL;DR for Builders and Investors
The shift from EOAs to smart accounts (ERC-4337, 6551) breaks existing indexers, creating a new data layer race for user-centric intelligence.
The Problem: EOA Indexers Are Obsolete
Current infrastructure like The Graph indexes contract events, not user intents across multiple smart accounts and token-bound wallets. This creates a fragmented, incomplete view of user activity and capital.
- Misses Cross-Chain & Multi-Account Activity: A user's assets and actions are split across a Vault, a 6551 NFT, and a DeFi smart wallet.
- Blinds Protocol Analytics: Cannot track cohort retention or capital flow for AA-powered dapps.
- Inhibits New Primitives: Social recovery, subscription billing, and automated strategies are invisible to legacy indexers.
The Solution: Intent-Centric Data Graphs
Next-gen indexers like Goldsky and Covalent are building graphs that map user intents to on-chain outcomes, aggregating activity across a user's entire smart account ecosystem.
- Unified User Profiles: Track a single entity across all their owned accounts (ERC-4337, 6551, Safes).
- Real-Time State Analysis: Monitor pending UserOperations in mempools for ~500ms latency alerts.
- Enable New Business Models: Power hyper-personalized dapps, risk engines for session keys, and cross-protocol loyalty programs.
The Battle: Who Owns the User Graph?
Control over this new data layer is a winner-take-most market. It's a race between specialized indexers, wallet providers (Safe, Coinbase Smart Wallet), and bundler services (Stackup, Alchemy).
- Monetization Leverage: The entity with the best user graph sells premium analytics and arbitrage signals.
- Vertical Integration Risk: Bundlers with exclusive mempool access could create data moats.
- Standardization Wars: Competing schemas (e.g., Covalent's Unified API vs. The Graph's subgraphs) will fragment developer adoption.
The Opportunity: Build the "Plaid for Web3"
The killer app isn't another block explorer. It's infrastructure that lets any dapp answer: "What is this user's total cross-chain net worth and transaction habit?"
- Permissioned Data Markets: Users sell anonymized behavioral data to protocols via Ocean Protocol-like mechanics.
- Superior Risk Models: Lending protocols (like Aave) use holistic user graphs for underwriting, not just isolated collateral.
- Investor Edge: VCs gain unprecedented visibility into real user engagement and retention metrics for portfolio dapps.
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.