Reliability is not profitability. Protocols like Across and Stargate optimize for uptime and latency, but these metrics are decoupled from user economic outcomes. A fast, cheap transaction that fails or front-runs the user destroys more value than a slow, successful one.
cross-chain-future-bridges-and-interoperability
Blog
Why Relayer Performance Metrics Must Be Tied to Economic Outcomes
An analysis of why traditional relayer KPIs like uptime and latency are insufficient for securing cross-chain value. We propose a first-principles shift to economic outcome-based incentives, examining protocols like LayerZero, Axelar, and Wormhole, and the emerging intent-based paradigm.
introduction
THE INCENTIVE MISMATCH
Introduction
Current relayer metrics measure activity, not value, creating systemic risk for cross-chain applications.
The MEV threat vector expands with every new chain. A relayer's performance on Arbitrum does not predict its behavior on Blast or Base, where novel MEV opportunities emerge. This creates unpredictable execution risk for intent-based systems like UniswapX.
Evidence: In Q1 2024, over 30% of cross-chain arbitrage value was captured by searchers, not the protocols or end-users. This leakage is a direct tax enabled by misaligned relayer incentives.
key-trends
ECONOMIC MISALIGNMENT
The Flawed State of Relayer Incentives
Current relayer models reward activity, not outcomes, creating systemic fragility.
01
The Problem: Volume Over Value
Relayers are paid for submitted transactions, not successful user outcomes. This creates perverse incentives for spam and frontrunning, degrading network quality for all participants.
- Incentivizes MEV extraction at user expense
- Ignores finality risk (e.g., failed cross-chain swaps)
- Metrics like TPS are vanity, not value
~30%
Failed Tx Waste
0%
Outcome Tied
02
The Solution: Outcome-Based Slashing
Tie a significant portion of relayer rewards and staked capital to successful, timely execution of user intents. Failed settlements or excessive latency trigger economic penalties.
- Aligns profit with user success
- Creates skin-in-the-game for reliability
- Automates QoS enforcement via cryptoeconomics
10x
Uptime Required
-99%
Spam Reduced
03
Entity Case: UniswapX & Solvers
UniswapX's solver competition demonstrates intent-based, outcome-focused architecture. Solvers compete on net output for the user, not just fee collection, creating a natural market for execution quality.
- Payment for filled orders only
- Competition on net user outcome
- Punishes bad routing via lost future auctions
$10B+
Volume Filled
Best Price
Key Metric
04
The Problem: Centralized Points of Failure
Without economic penalties, relayers face minimal consequences for downtime or censoring transactions. This leads to reliance on a few trusted entities, contradicting decentralization goals.
- Creates systemic fragility (e.g., L2 sequencer outages)
- No cost to censorship
- Trust required where it shouldn't be
1-3
Dominant Relayers
High
Concentration Risk
05
The Solution: Verifiable Performance Oracles
Implement decentralized oracle networks like Chainlink Functions or API3 to objectively measure and attest to relayer performance metrics—latency, success rate, censorship—feeding directly into reward contracts.
- Objective, on-chain QoS data
- Removes subjective governance from scoring
- Enables automated rebates for SLA breaches
<500ms
SLA Enforced
100%
On-Chain Proof
06
Entity Case: Across & UMA Optimistic Oracle
Across Protocol's bridge uses UMA's optimistic oracle to verify correctness and trigger refunds for slow fills. This bonds relayers to a specific service level, making latency a direct financial variable.
- Slow fills penalized via oracle-verified claims
- Economic security for speed
- Model applicable to any state attestation
~2 min
Avg. Fill Time
Optimistic
Verification
RELAYER PERFORMANCE
Metric Mismatch: Traditional KPIs vs. Economic Reality
Comparing traditional operational metrics against economic KPIs that directly impact user and protocol value.
| Key Performance Indicator | Traditional Metric (Vanity) | Economic Metric (Value) | Example: Top-Tier Relayer |
|---|---|---|---|
Speed Measurement | Median Latency (< 1 sec) | 95th Percentile Latency (< 3 sec) | 2.8 sec |
Cost Efficiency | Gas Price Paid (Gwei) | User's Total Effective Cost (Basis Points) | 15 bps |
Reliability Signal | Uptime SLA (99.9%) | Settlement Failure Rate (< 0.1%) | 0.05% |
Throughput | Max TPS (10,000) | Capital Efficiency (TVL / Daily Volume) | 8.5x |
Security Posture | Audit Count (3) | Time-to-Finality with Economic Guarantees (< 5 min) | 4 min |
Revenue Model | Transaction Fee Revenue | Value Secured in Escrow (Protocol TVL) | $42M |
User Alignment | Number of Integrations (50+) | Fill Rate for Limit Orders (> 99%) | 99.4% |
deep-dive
THE INCENTIVE MISMATCH
The First-Principles Case for Economic Slashing
Current relayer reward systems measure technical uptime, not economic outcomes, creating a fundamental misalignment with user value.
Slashing must be economic. Relayer incentives are broken because they reward availability, not correctness or finality. A system like Chainlink's OCR punishes downtime, but a relayer can be 100% online while delivering stale prices or censoring transactions. The penalty must directly correlate to the user's lost economic value, not just a binary uptime check.
Technical metrics are insufficient. Measuring TPS or latency creates a perverse incentive for relayers to prioritize low-value spam over high-value, complex transactions. This is why EigenLayer's cryptoeconomic security model ties slashing to protocol-specific faults; a generic uptime SLA fails to capture the actual cost of failure for applications like Across Protocol or Stargate.
The evidence is in failed arbitrage. If a cross-chain arbitrage bundle arrives 5 seconds late due to relayer latency, the opportunity and user funds are gone. The current messari-style reporting on 'network health' misses this entirely. The slashing penalty must be a direct function of the proven financial loss, creating a skin-in-the-game mechanism that pure technical monitoring cannot replicate.
protocol-spotlight
ECONOMIC ALIGNMENT
Protocols Pushing the Frontier
Leading relayers are moving beyond basic uptime to tie performance directly to user and protocol outcomes.
01
The Problem: Pay-for-Availability, Not Results
Traditional staking rewards relayers for being online, not for optimal execution. This creates misaligned incentives where a relayer can be "live" but still lose users money through poor MEV capture or slow fills.
- Misaligned Incentive: Rewards for uptime, not for securing best price.
- User Impact: Users bear the full cost of suboptimal execution (slippage, failed trades).
- Systemic Risk: Encourages a race to the bottom on hardware costs, not execution quality.
0%
Outcome-Based
100%
Uptime-Based
02
The Solution: Succinct's Proof of Timely Execution
Succinct's ZK light client relayers must submit validity proofs of state updates within a specific time window to earn rewards. Performance is cryptographically enforced.
- Cryptographic Enforcement: Rewards are gated on proof submission latency.
- Economic Outcome: Faster finality directly improves capital efficiency for apps like Across and Chainlink CCIP.
- Protocol Benefit: Creates a verifiable SLA, allowing dApps to select relayers based on proven historical performance.
< 4 min
Proven Latency
ZK-Guaranteed
Execution
03
The Solution: Across v3 & the Anoma Stack
Across uses a solver-based model where relayers compete on cost, bundling user intents. Payment is for filling the intent, not for being available. This is a direct application of Anoma's intent-centric architecture.
- Intent-Based Pricing: Relayers are paid a fee only upon successful fill, aligning profit with user success.
- Competitive Outcome: Drives down costs and improves fill rates through open competition among solvers.
- Architecture Shift: Moves from passive message passing to active fulfillment of user intents, as seen in UniswapX and CowSwap.
Solver-Based
Model
Fill-Driven
Fees
04
The Problem: Opaque MEV Extraction Erodes Trust
Relayers with access to order flow can extract value via frontrunning, sandwich attacks, or poor auction participation, directly harming end-users. This is a critical failure of economic alignment.
- Hidden Tax: Users unknowingly pay ~5-50+ bps in extracted MEV.
- Trust Assumption: Users must trust the relayer's black-box execution logic.
- Protocol Risk: Applications built on LayerZero or Wormhole inherit this opacity, damaging composability.
5-50+ bps
Hidden Cost
Opaque
Execution
05
The Solution: SUAVE & Encrypted Mempools
SUAVE (Single Unified Auction for Value Expression) decentralizes the block building market. It allows users to express preferences (e.g., "no frontrunning") and relayers to compete in a transparent auction for order flow.
- User Sovereignty: Preferences are enforced via cryptography, not trust.
- Relayer Competition: Rewards are tied to winning auctions and fulfilling expressed intent, not exploiting it.
- Ecosystem Outcome: Creates a liquid market for block space that aligns relayer revenue with explicit user demands.
Auction-Based
Revenue
Preference-Enforced
Execution
06
The Future: Performance Bonds & Slashing for SLOs
Next-gen relay networks will require staked bonds that are slashed for missing Service Level Objectives (SLOs) like latency, censorship resistance, or fill-rate guarantees.
- Skin in the Game: $1M+ bonds create direct financial liability for poor performance.
- Measurable SLOs: Slashing is triggered by objective, on-chain metrics (e.g., >10s latency).
- Endgame: Transforms relayers from infrastructure providers into economically-aligned guarantors of network performance.
$1M+
Performance Bond
On-Chain SLO
Slashing Condition
counter-argument
THE REAL-WORLD GAP
Counterpoint: The Practicality Problem
Abstract performance metrics are meaningless unless they directly impact user costs and protocol revenue.
Latency is a business metric. A relayer's transaction ordering speed directly determines a user's slippage on a DEX like Uniswap. Slow finality on a bridge like Across or LayerZero translates to arbitrage losses for the user, not just a technical stat.
Throughput must create revenue. A system claiming 100k TPS is useless if its fee market is broken. Validators and relayers, like those in the Cosmos ecosystem, optimize for profit, not theoretical capacity. Empty blocks are a failure.
The benchmark is economic finality. The only metric that matters is how quickly a user's asset is usable for its intended yield. This is why Solana prioritizes sub-second confirmation for its DeFi primitives over theoretical peak TPS.
Evidence: Arbitrum Nitro's 2M TPS capacity is a marketing figure; its practical constraint is the L1 data posting cost, which directly defines the economic model for sequencers and, ultimately, user transaction fees.
takeaways
ALIGNING INCENTIVES WITH OUTCOMES
TL;DR: The New Relayer Incentive Framework
Current relayer models pay for activity, not results. This framework ties rewards directly to measurable economic value delivered to users and protocols.
01
The Problem: Paying for Hops, Not Happiness
Legacy fee models reward relayers for moving bytes, not for optimal execution. This creates perverse incentives for latency, poor route selection, and MEV extraction at user expense.
- Misaligned Goals: Relayer profit ≠user best execution.
- Hidden Costs: Users pay for inefficiency and slippage.
- Protocol Risk: Poor finality can destabilize apps like Aave or Compound.
~15%
Slippage Variance
500ms+
Latency Tax
02
The Solution: Outcome-Based Slashing & Rewards
Shift from gas reimbursement to a bonded model where relayers are penalized for poor performance and rewarded for value-add. Metrics include finality speed, execution price vs. benchmark, and liveness.
- Skin in the Game: Require a $10M+ bond slashed for failures.
- Positive Sum: Share MEV savings (e.g., via CowSwap, UniswapX) with users.
- Verifiable Proofs: Use ZK proofs or optimistic verification for claims.
99.9%
Uptime SLA
+30%
User Savings
03
Protocols as Stakeholders: The Across & LayerZero Model
Leading cross-chain systems are moving to curate relayer sets based on performance data. This creates a competitive marketplace where the best operators win.
- Data-Driven Curation: Use on-chain analytics to score relayers.
- Dynamic Fee Auction: Fees adjust based on demand and historical performance.
- Sybil Resistance: Identity and reputation replace anonymous pools.
10x
Operator Churn
$5B+
Secured Volume
04
The Endgame: Relayers as Execution Co-Processors
The final evolution abstracts relayers entirely. Users submit intents; a competitive network of solvers (like Anoma, SUAVE) bids to fulfill them optimally. The relayer role dissolves into specialized execution.
- Intent-Centric: User declares 'what', not 'how'.
- Atomic Composability: Cross-chain swaps settle in one atomic action.
- Efficiency Maximized: Market forces drive cost to theoretical minimum.
-90%
User Complexity
~1s
Cross-Chain Finality
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