Restaking has plenty of capital, but little revenue
The restaking concept was compelling. A user stakes ETH to secure Ethereum and earn a yield. That same capital can then be “restaked” to secure other services, such as oracles or data availability networks, in exchange for additional compensation.
The promise fit perfectly with the period when DeFi was still looking for yield everywhere. The problem emerged later: demand for this additional security never grew as quickly as the amount of money deposited.
At its peak, EigenLayer had reached around $19.7 billion in TVL. The restaking sector still held $10.02 billion on September 8, 2026, but had generated just $99,977 in fees over seven days. For an equivalent amount of secured capital, CoinDesk estimates that traditional liquid staking generated around 53 times more revenue.
DefiLlama’s current snapshot remains harsh: around $10.16 billion in TVL, against just $46,414 in fees over seven days and less than $3,000 recorded as protocol revenue during the period. Methodologies may vary, but the gap between locked capital and money actually captured remains enormous.
There is plenty of security available. Few customers appear willing to pay enough for it.
Ether.fi is already turning the page
The clearest symbol comes from ether.fi.
At launch, deposits on the platform were automatically restaked through EigenLayer. In August, ether.fi removed this mechanism from weETH, its main liquid token. Less than 1% of its assets were still restaked, according to documentation cited by CoinDesk, while its remaining technical links with EigenLayer are set to disappear.
Mike Silagadze, ether.fi’s CEO, explains the decision through a fairly simple combination: little additional yield and more risk.
Since April 2025, slashing has effectively been active on EigenLayer. This mechanism allows a service to penalize an operator that fails to meet certain conditions, potentially causing it to lose part of its committed capital. Eigen Labs itself presents the feature as a way to make economic commitments genuinely enforceable.
The additional yield must therefore compensate for a risk that has become concrete.
This equation becomes difficult to defend when revenue stagnates. It carries even more weight in a DeFi sector already facing a growing number of hacks and smart contract incidents.
Ether.fi is not abandoning all of its activities. Instead, the company is moving closer to a crypto neobank model, with staking, a card, borrowing and treasury products. Its own accounts show why: in the second quarter of 2026, fees linked to its card generated around $3.14 million in gross profit, while EigenLayer restaking rewards accounted for $2.87 million.
Smaller protocols earn almost nothing
For other players, the figures are even more modest.
Renzo, Kelp, Swell, Puffer Finance and Bedrock generated around $953,350 in gross profit in the second quarter of 2026 combined. Three quarters earlier, they had reached $2.18 million. Puffer generated just $21,590 for the quarter, while Swell brought in around $22,370.
Kelp illustrates the problem well. In the second quarter, the protocol reported $9.43 million in gross revenue. Much of that amount, however, simply represented staking rewards passed on to users. Once costs were deducted, its gross profit fell to around $415,000. EIGEN rewards, meanwhile, appeared at $460,600 in both revenue and costs: the money did not remain with the protocol.
In other words, a large TVL does not necessarily constitute a business.
That is probably the most interesting lesson from this episode. DeFi is increasingly being forced to demonstrate something beyond its ability to attract billions through points, airdrops and rewards. This shift aligns with Ethereum’s move toward more institutional use cases such as RWAs, where the revenue question rests more heavily on services that are actually purchased.
Restaking is not dead. EigenLayer still secures billions of dollars and continues to develop its market for verifiable services. But the first phase is over: the capital arrived well before the customers capable of paying for it.
After the rush for TVL, restaking is entering a much more ordinary period. It now has to make money.
