When Zero Network announced it was closing last month, the crypto community's reaction was telling. Not a shock. Not concerned. Just a weary nod.
The closure joined a growing list of struggling rollups and came amid renewed debate about whether Ethereum's sprawling layer‑2 ecosystem has become too crowded. At the same time, Ethereum creator Vitalik Buterin has urged developers to rethink the network's long‑term scaling roadmap, while several major projects have shifted away from marketing themselves as general‑purpose blockchains and toward more focused applications in payments, stablecoins and tokenized assets.
The question the industry keeps asking is the same one it's been avoiding for two years: did the L2 ecosystem grow too fast?
Too Many Chains. Not Enough Reasons
The honest answer, according to people building inside the ecosystem, is yes, but with an important nuance.
Ben Fisch, co‑founder and CEO of Espresso Systems, put it plainly: "There were way too many general‑purpose layer twos, which frankly don't make sense as a product, because there's no reason to have many, many versions of the same thing."
The data backs him up completely. Today, activity across Ethereum's layer‑2 ecosystem remains heavily concentrated among a handful of networks. Base and Arbitrum alone account for more than 80% of layer‑2 DeFi total value locked, according to DefiLlama data.
That concentration tells a harsh story for everyone else.
Smaller Chains Are Quietly Bleeding Out
While Base and Arbitrum hold the market's attention, the rest of the L2 landscape is quietly deteriorating.
Over the past six months, networks including Linea, World Chain, Starknet and Mantle have all seen declining bridge deposits. Linea's deposits, for example, fell from $976 million in November 2025 to $367 million in May 2026, a decline of more than 60%.
Losing over half your deposited value in six months isn't a growth problem. It's an existential one.
The Paradox Nobody Expected
Here's the uncomfortable irony at the heart of this: launching a layer‑2 has never been cheaper or easier, yet succeeding at one has never been harder.
Ethereum's Dencun upgrade, introduced in 2024, dramatically reduced the cost of posting rollup data to Ethereum through blobs. According to Messari research, data availability costs now represent only a small fraction of operator expenses for many OP Stack chains.
"From an operator perspective, it is definitely cheaper to run an L2 today," said Alice Hou, a former research analyst at Messari. "The economics of launching an L2 have become easier, but the real challenge is still generating enough sustained demand to make the network worth operating."
The barriers to creating a blockchain keep falling. The barriers to attracting users keep rising. That gap is where most L2s are dying.
The Shift: From Infrastructure to Applications
The teams that are surviving aren't just building another chain. They're building for specific use cases with existing distribution.
Fisch pointed to asset managers launching tokenized money‑market funds, stablecoin issuers and tokenized deposit platforms as examples of businesses that have clear reasons to operate on‑chain. For those firms, a dedicated layer‑2 can offer lower costs, greater control and more predictable performance than deploying directly as a smart contract.
Coinbase's Base is the clearest example of this working in practice, an exchange leveraging 100 million existing users to build network effects from day one.
What Actually Qualifies You to Build an V2
Hou framed the right question clearly: "The question should not be, 'Can this company launch an L2?' It should be: 'Does this business already have enough distribution, financial activity and ecosystem synergies to make an L2 meaningfully useful?'"
The L2 boom is over. The L2 consolidation has begun. What survives won't look like hundreds of competing general‑purpose chains. It'll look like a smaller group of focused networks tied to real businesses, real users, and real demand. Everything else is just expensive infrastructure waiting to be shut down.
