Which Restaking Services Will Be Eliminated When AVS Revenue Is Insufficient?
It will eliminate AVSs that cannot prove they have 'paying customers'. The direct consequence of insufficient AVS revenue is that restaking services shift from 'locked mining' back to business fundamentals—a AVS that long-term relies on token issuance subsidies rather than genuine service fees will be judged by the market as lacking product-market fit (PMF) and ultimately eliminated.
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！
The current state of the entire AVS layer is: security supply is extremely abundant (leading AVSs often lock millions of ETH), but genuine demand for paid services is extremely weak. This is not an issue for any single AVS; it's structural.
1. First, Understand the AVS Revenue Structure: Subsidies vs. Real Revenue
The 'yield' you earn from staking actually comes from two completely different pockets.
Situation A: Token inflation subsidies (currently the vast majority) AVSs are not profitable in the early stages, so they issue their own tokens to pay stakers and operators. DeFiLlama data shows that the EigenCloud protocol has annualized revenue of about $0, but annualized incentives of about $53.62 million. The 30-day total fees are about $1.06 million, and 30-day incentives are about $1.02 million. The vast majority of yields come from 'printing money', not from 'earning money'.
Situation B: Real service fee revenue (currently almost zero) Services provided by AVS (data availability, oracles, ZK proof generation) charge end users and then distribute part of the revenue to stakers and operators. Currently, for most AVSs this revenue is near zero. Among leading AVSs, EigenDA has locked over 4 million ETH, but actual slashing cases are zero; its security constraint mechanism remains largely on paper.
Key data: Currently, the EigenCloud ecosystem has over 20 AVSs and over 200 operators, but real paid usage scenarios are still in their infancy. This means the vast majority of AVSs are in a state of 'having a security budget but no commercial conversion'.
2. Which AVSs Are Most Likely to Be Eliminated — Three High-Risk Types
Based on the actual operation of the AVS layer, the following three types are most likely to be eliminated when revenue is insufficient.
Type 1: Application AVS, lacking clear payers A typical example is Cyber. It provides an additional security layer for appchains via MACH AVS, locking about 3.49 million ETH, but the project's progress has clearly slowed, with its whitepaper stuck at Q1 2025. The question is: who will pay for 'extra security' on an ongoing basis? If this cannot be answered, once subsidies stop, nobody will use it.
Type 2: Heavy on security, light on demand AVS This type of AVS positions itself as providing 'high-grade economic security', but the reality is that only a very small number of highly security-sensitive infrastructures actually need Ethereum-level security costs. The demand for shared security itself is shrinking—the number of new public chains and appchains continues to decline, and development resources are concentrating in mature ecosystems, naturally shrinking the demand base for providing general security for new chains.
Type 3: Computational AVS, but performance is the core competitiveness Take Lagrange as an example; it provides decentralized computation services for ZK proofs, locking about 3.05 million ETH. Restaking security plays a 'stability backstop' role, not a decisive advantage. Its long-term competitiveness depends on performance, latency, and unit cost. If ZK applications do not expand, or if proof services cannot form a sustainable paid market, even massive restaking security is useless.
3. Underlying Logic: The Incentive for AVS to Leave EigenLayer
If an AVS truly succeeds and generates sufficient real revenue, what will it likely do? Most likely, detach from EigenLayer.
The reason is straightforward: a successful AVS needs to share revenue with ETH/EIGEN stakers, which would cause a net drain on its protocol. Rather than continuously giving away money to external stakers, it would rather detach from the restaking system, retain more revenue, and accumulate more value for its native token. Just like many dApps choose to become their own L2 or AppChain to capture more fees and MEV.
This means that AVSs that can genuinely generate revenue will, in the long run, likely not remain within the restaking system. Those that remain are either still burning subsidies or simply can't earn money.
4. Data Validation: Estimated Annualized Yield of AVSs
According to data from August 2024, EigenDA had locked about 3.64 million ETH and about 71.4 million EIGEN tokens (worth about $247 million at the time). The corresponding data availability competitor Celestia's annualized data availability revenue for the same period was only about $20,000. Even if EigenDA captured the entire market, relative to its billions in staked capital, the annualized return would be shockingly low.
For eOracle's competitive market, the oracle leader Chainlink's price data revenue in the first half of 2024 was about $36.7 million. Even if eOracle captured the entire market share, the annualized return on its $7.8 billion in restaked capital would be less than 1%. And this doesn't yet account for the structural selling pressure caused by EigenCloud's own 7% annual inflation rate.
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！
5. What You Can Do Now — How to Judge Whether an AVS Has a Future
Check the AVS's public revenue data: On platforms like DeFiLlama, check its protocol revenue and fee data. If 'Protocol Revenue' stays at zero for a long time, it means there are no real paying customers yet.
Check if slashable assets are zero: If an AVS has locked millions of ETH but slashable assets remain zero for a long time, it means its security constraints are only on paper.
Determine whether the AVS has payers beyond token subsidies: Ask three questions—who pays? Is the payment predictable? If token incentives stop, will users stay?
How to confirm you have correctly made the judgment:
Next time you consider participating in an AVS's restaking, first check its '30-day fees' data. If fees are close to zero while incentives are massive, it means it's currently surviving on external support. Look at its whitepaper update date—if it hasn't been updated for over a year, the team's progress may have slowed. If both of these signals appear at the same time, this AVS has a high probability of being eliminated when revenue falls short.
