Indexing is broken for intent-based architectures. Current systems like The Graph query final on-chain state, but miss the critical off-chain negotiation and fulfillment logic of protocols like UniswapX and CowSwap.
account-abstraction-fixing-crypto-ux
Blog
The Future of Blockchain Indexing: Tracking UserOps, Not Just Transactions
Current indexing infrastructure is built for EOAs and misses the intent-driven, batched, and sponsored transactions of smart accounts. This is a critical data blind spot.
introduction
THE PARADIGM SHIFT
Introduction
Blockchain indexing must evolve from tracking simple transactions to interpreting complex user intents.
UserOps are the new atomic unit. A single user intent, expressed as a UserOperation in ERC-4337, fragments into dozens of transactions across solvers, fillers, and bridges like Across and LayerZero.
The value is in the flow, not the state. Tracking the lifecycle of an intent—from submission to settlement—reveals solver competition, cross-chain liquidity routing, and real user demand that raw transaction logs obscure.
Evidence: Over 60% of swaps on CowSwap are settled via off-chain solvers, creating a multi-transaction flow that traditional indexers fail to reconstruct as a single economic event.
thesis-statement
THE USEROP PARADIGM
The Core Argument
The next generation of blockchain infrastructure must index user intents and their fulfillment paths, not just atomic on-chain transactions.
Indexing user intents is the new frontier. Current indexers like The Graph track transaction logs, which are low-level state changes. The future is tracking abstracted user operations (UserOps) from ERC-4337 account abstraction and cross-chain intents via protocols like UniswapX and Across.
Transactions are a subset of intents. A single user intent to 'swap ETH for ARB on Optimism' may spawn transactions across Ethereum, a bridging solution like LayerZero, and a DEX aggregator. Legacy indexers see three disjoint events; a UserOp indexer sees one logical operation.
The data structure changes. Instead of from, to, value, the primary objects become sender, intent, fulfillmentPath, and successState. This shift enables analytics for bundler profitability in ERC-4337 and solver competition in CowSwap.
Evidence: The mempool for UserOps (the "alt mempool") and intent-based flows now represents over 15% of major DEX volume. Protocols that index this layer, like Blocknative's Mempool Explorer for intents, capture the actual user journey.
key-trends
THE FUTURE OF BLOCKCHAIN INDEXING
Key Trends Breaking Legacy Indexers
Legacy indexers built for simple transfers are failing to capture the complexity of modern user interactions, from account abstraction to cross-chain intents.
01
The Rise of the Intent Graph
Users don't want to sign transactions; they want outcomes. Legacy indexers track TX hashes, but the future is tracking user intents across solvers, fillers, and bridges.
- Key Benefit: Enables discovery of MEV capture and solver competition.
- Key Benefit: Provides a true map of user demand flow, not just settlement.
UniswapX
Pioneer
CowSwap
Solver Network
02
Account Abstraction Fragmentation
Every AA stack (ERC-4337, Starknet, zkSync) has its own UserOperation mempool. A universal indexer must aggregate these to track the real user activity layer.
- Key Benefit: Unified view of gas sponsorship and session key usage.
- Key Benefit: Critical for auditing Paymaster economics and security.
ERC-4337
Standard
10+
Mempools
03
The Cross-Chain State Problem
A user's state (NFTs, tokens, positions) is now spread across 5+ chains. Legacy indexers are chain-siloed, creating a broken user profile.
- Key Benefit: Enables omnichain reputation and credit systems.
- Key Benefit: Essential for cross-chain intent execution and tracking (e.g., LayerZero, Axelar, Wormhole).
$10B+
Bridged Assets
Across
Intent Bridge
04
Real-Time is Non-Negotiable
DeFi and gaming require sub-second data. Legacy indexers with multi-block finality delays are obsolete for live applications.
- Key Benefit: Enables high-frequency on-chain strategies and real-time dashboards.
- Key Benefit: Reduces arbitrage latency from ~12s to <1s.
<500ms
Target Latency
12s
Legacy Lag
05
Modular Data Pipelines Over Monoliths
A single indexing stack cannot optimize for all data types (logs, traces, states, intents). The future is specialized decoupled pipelines.
- Key Benefit: Cost efficiency by only processing needed data.
- Key Benefit: Independent scaling for high-volume event types (e.g., NFT transfers vs. DeFi swaps).
-70%
Compute Waste
The Graph
Facing Pressure
06
Proving, Not Trusting, Data Integrity
Relying on RPC providers or centralized indexers for state is a security risk. The next standard is cryptographically verifiable query results.
- Key Benefit: Auditable data provenance for institutional use.
- Key Benefit: Enables trust-minimized oracles and on-chain proofs.
zk-Proofs
Endgame
EigenLayer AVS
Potential
THE INDEXING PARADIGM SHIFT
EOA vs. Smart Account: A Data Model Comparison
A data model comparison highlighting the fundamental differences between indexing Externally Owned Accounts (EOAs) and Smart Accounts (ERC-4337). This dictates the complexity and capability of future blockchain data infrastructure.
| Data Model Feature | Traditional EOA | ERC-4337 Smart Account | Implication for Indexers |
|---|---|---|---|
Atomic Unit of Execution | Transaction (tx) | User Operation (UserOp) | Must track a new, higher-level abstraction. |
Sender Abstraction | Single EOA address | Smart Account address (can be a contract) | Indexing logic must handle contract-based origin. |
Fee Payment Entity | Signer (msg.sender) | Separate (can be paymaster, bundler, or account) | Fee flow tracking requires multi-entity graph analysis. |
Transaction Validation Logic | Fixed (ECDSA sig) | Custom (validateUserOp function) | Indexers must execute custom logic to understand validity. |
Execution Trigger | Direct on-chain tx | Indirect via Bundler & EntryPoint | Requires tracking mempool (alt mempool) and a new actor (Bundler). |
Gas Sponsorship Model | Self-funded only | Paymaster-sponsored (gasless) | Must index new payment flows and token allowances for paymasters. |
Batch Execution Support | No (native) | Yes (via handleOps) | Single on-chain tx can represent multiple user intents; requires unpacking. |
Session Key / Automation | No (requires separate tx) | Yes (native via validateUserOp logic) | Enables stateful user sessions; indexing must track active permissions over time. |
deep-dive
THE DATA GAP
The Anatomy of a Blind Spot
Current blockchain indexing fails to capture the intent-driven, multi-chain reality of user operations, creating a critical data vacuum.
Indexers track transactions, not actions. This is the fundamental flaw. A user swapping ETH for USDC on UniswapX via a permit2 signature and a fill-or-kill intent executes a single on-chain transaction, but the indexer sees only the final settlement. The user's intent, the off-chain auction, and the cross-domain routing logic are invisible.
ERC-4337 and intents create abstraction layers. The rise of account abstraction and intent-based architectures like UniswapX and CowSwap decouples user commands from on-chain execution. Indexing the resulting UserOperation or settled transaction provides a distorted, incomplete view of user behavior and protocol flow.
The multi-chain execution is a black box. A user's action often involves a cross-chain intent relayed via Across or LayerZero. Current indexers, tied to single chains, cannot reconstruct the unified user journey. They see isolated settlement events, not the cohesive cross-chain operation.
Evidence: The MEV supply chain. Over 90% of DEX volume on Ethereum is routed through private order flow. Indexers tracking only the public mempool and final blocks are blind to the off-chain auction mechanics and searcher-builder relationships that define modern execution. This data gap distorts metrics like true price impact and protocol efficiency.
protocol-spotlight
THE FUTURE OF BLOCKCHAIN INDEXING
Who's Building the New Stack?
The next wave of dApps requires real-time data on user intents and cross-chain states, not just historical transactions.
01
The Graph's Substreams: Real-Time Intent Streams
Substreams enable firehose access to raw blockchain data, allowing indexers to build custom data pipelines. This is critical for tracking User Operations (UserOps) from AA wallets and intents across chains like Ethereum and Arbitrum.\n- Sub-second latency for event streaming\n- Enables intent-specialized subgraphs for protocols like UniswapX\n- Foundation for cross-chain state proofs and MEV analysis
<1s
Latency
1000x
Throughput
02
Goldsky & The Indexer Trilemma
Goldsky solves the indexer trilemma: speed, correctness, and cost. Their managed service provides sub-second finality for indexed data, essential for real-time dashboards and intent-based applications.\n- ~500ms end-to-end data delivery\n- SQL and GraphQL APIs for complex UserOp queries\n- Powers frontends for Across Protocol and LayerZero explorers
500ms
Finality
-70%
Dev Time
03
Covalent: The Unified Data Layer
Covalent provides a single API spanning 200+ blockchains, abstracting away chain-specific complexities. This is the backbone for wallets and dashboards that need a unified view of user activity and asset flows across fragmented ecosystems.\n- No-code historical and real-time data queries\n- ERC-4337 UserOp support for Account Abstraction analytics\n- Used by Rainbow Wallet and CoinGecko for portfolio tracking
200+
Chains
1 API
Unified
04
RPC 2.0: From Simple Queries to Complex States
Next-gen RPC providers like Alchemy and QuickNode are evolving into verifiable state providers. They move beyond eth_call to offer APIs that can prove the state of another chain, a prerequisite for secure cross-chain intent execution.\n- Enhanced APIs for smart account simulation\n- Bundler services integrated with RPC endpoints\n- State diffs for tracking intent fulfillment across Optimism and Base
Zero-Lag
State Sync
10k+ RPS
Scale
05
Hyperbolic: The Intent-Centric Database
Hyperbolic is building a parallelized OLAP database specifically for blockchain data. It enables complex analytical queries over petabytes of intent data, which is impossible with traditional indexing architectures.\n- Columnar storage optimized for aggregate UserOp analysis\n- Sub-second queries across entire chain history\n- Enables MEV research and intent market design analytics
PB Scale
Data
100x
Query Speed
06
The Problem: Indexers Can't See Across Chains
Current indexing is chain-siloed. An intent to swap on UniswapX that routes through Across on Ethereum and executes on Arbitrum creates an unobservable workflow. This fragmentation breaks user experience and composability.\n- No atomic view of cross-chain user journeys\n- Impossible to audit intent fulfillment guarantees\n- Fragmented liquidity and pricing data hinders CowSwap-like solvers
Siloed
Data
Broken UX
Result
counter-argument
THE MARKET REALITY
The Skeptic's View: Is This Premature?
The infrastructure for tracking user intents is being built before the dominant use case for it exists.
The market is not ready. The primary use case for intent-centric indexing is cross-chain activity, but the dominant user experience remains single-chain. Protocols like UniswapX and CowSwap are early experiments, not mass-market products.
Infrastructure precedes demand. This is a classic crypto pattern. The ERC-4337 standard for account abstraction is live, but the killer app for bundlers and paymasters is still hypothetical. We are building the observability layer for a ghost town.
The data is too noisy. Indexing raw UserOperations without a standardized semantic layer creates uninterpretable data. A user's intent to swap on Aave/Compound is buried in low-level calldata that current indexers like The Graph cannot parse.
Evidence: Daily active wallets on Arbitrum and Optimism exceed 300k, but cross-chain messaging volume via LayerZero and Axelar is a fraction of that. The bridge is built, but the traffic isn't there yet.
takeaways
THE INTENT-CENTRIC SHIFT
TL;DR for Builders and Investors
Blockchain's next infrastructure war will be won by indexing user intent, not just final state.
01
The Problem: Transactional Indexing is Obsolete
Current RPCs and indexers like The Graph track on-chain state, missing the critical pre-execution intent layer where ~30% of DeFi volume now originates via solvers and fillers.
- Blind to UserOps: Can't track cross-chain intents, failed bundles, or MEV opportunities.
- Reactive, Not Predictive: Data arrives post-settlement, useless for real-time intent matching.
- Fragmented View: No unified feed for intents from UniswapX, CowSwap, Across, and native AA wallets.
-30%
Visibility Gap
500ms+
Latency Lag
02
The Solution: Intent-Aware RPC & Indexer
A new infrastructure layer that ingests, classifies, and routes user intents in real-time, becoming the search engine for pre-chain activity.
- Universal Mempool Feed: Aggregate intents from public mempools, private channels, and solvers.
- Intent Graph: Map relationships between users, dApps, and solvers across chains like Ethereum, Solana, Arbitrum.
- Predictive Analytics: Surface fillability scores and optimal routing paths for builders.
10x
Data Scope
<100ms
Intent Latency
03
Build: The New Primitive is an Intent Oracle
Forget basic price feeds. The killer app is a verifiable oracle for intent state, enabling new dApp architectures.
- Conditional Execution: Build dApps that trigger only when specific intent patterns emerge.
- Solver Competition: Create transparent markets for intent fulfillment, reducing costs.
- Cross-Chain UX: Enable seamless LayerZero-style experiences by tracking intent lifecycle across all domains.
$1B+
Market Opportunity
0
Current Providers
04
Invest: Follow the Solver Stack
Value accrual is shifting from L1s to the intent fulfillment layer. The indexing stack that powers solvers will capture rent.
- Vertical Integration: Indexers that also offer solver services (like Flashbots SUAVE) have inherent advantage.
- Data Moats: Real-time intent graphs create unassailable barriers to entry.
- Monetization: Fee-sharing from routed volume and premium API access for hedge funds.
50%
Margin Premium
100x
TAM Multiplier
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