Before the article — one flag: the parsed content you supplied contains zero information points. Every field in the stage-two report is N/A, and the report itself states that any filled-in detail would be invented. So I cannot honestly claim the piece below is "based on the parsed content" — there is nothing to base it on.
What follows is therefore clearly labelled as original analytical feature writing, not a report on a specific source article. If you paste the original text (or a non-empty information-point list), I can rewrite it as a genuine derivative piece with proper attribution and quoted specifics.
Fee compression, modular design, shared security and regulatory clarity are rewriting the economics of public blockchains — and the winners are unlikely to be the ones the last cycle priced in
An analytical feature on the structural forces reshaping blockchain infrastructure. Figures are directional and illustrative of well-documented industry trends rather than point-in-time measurements; readers should verify current data before acting.
1. The quiet collapse of the fee premium
For most of the industry's history, the central investment thesis for a layer-one blockchain rested on a single, simple mechanism: block space is scarce, demand for it grows faster than supply, and therefore the native asset accrues value through fee burn, staking yield, or both. The logic was elegant, easy to communicate, and almost entirely dependent on one assumption — that blockspace would remain expensive.
That assumption has been steadily dismantled over the past several years. The introduction of dedicated data-availability sampling and blob-carrying transactions reset the cost of posting data to a layer-one from a variable expense into something closer to a rounding error for most rollups. Combined with a broader migration of activity to execution environments that batch thousands of transactions into a single on-chain commitment, the marginal cost of transacting has fallen by orders of magnitude on most major networks. The result is a structural change that is easy to celebrate as a user-experience win and much harder to reconcile as a business model.
The consequences ripple outward. When fees collapse, the burn mechanism that underpinned a large share of the "scarce asset" narrative produces far less scarcity. Staking yields that were partly funded by priority fees compress toward the issuance rate. And the security budget of a proof-of-stake network — the total annualized compensation paid to validators — becomes a subject of genuine debate rather than a settled matter. A network that is cheap to use is a network that is cheap to attack relative to the value it secures, unless the asset price or the issuance schedule compensates.
This is not a temporary condition awaiting a demand recovery. It is the predictable endpoint of a decade of engineering effort aimed precisely at making block space abundant. The industry succeeded. Now it has to live with the consequences.
The most visible adjustment has been a reframing of where value actually accrues. If execution is cheap and abundant, then execution itself is not the scarce resource. What remains scarce is credible neutrality, deep liquidity, reliable settlement guarantees, and — increasingly — the right to order transactions. Maximal extractable value, once treated as an unfortunate byproduct to be minimized, has become the dominant revenue line for many infrastructure operators. That shift has profound implications for how these networks are governed, how they are regulated, and who ultimately captures the surplus.
There is a second-order effect that receives less attention. When transaction fees are negligible, the economic relationship between a network and its users changes character. Users no longer pay for the network in any meaningful sense; they pay the application layer, or the sequencer, or the solver, or nobody at all. The network becomes a public utility funded by inflation and by a small number of high-value actors — market makers, arbitrageurs, and MEV searchers — whose activity subsidizes everyone else. That is a stable arrangement only for as long as those actors remain present and profitable.
2. Modularity's second act
The modular thesis — separate execution, settlement, consensus and data availability into specialized layers — was the defining architectural argument of the previous cycle. It has now moved from whiteboard to production, and the results are more nuanced than either its advocates or its critics predicted.
Data availability was supposed to be the hardest problem and the most defensible business. Instead, it became the most commoditized. Multiple well-funded teams shipped DA layers within a short window, each competing on throughput, cost and cryptographic guarantee. The competitive dynamic was predictable: when several providers offer a functionally similar service and switching costs are low by design, price converges toward marginal cost. Blob space on the dominant settlement layer set a reference price that no independent DA layer could undercut profitably at scale, and the market split into a small number of large users and a long tail of chains whose data volumes are immaterial.
The survivors in DA have largely repositioned. Rather than selling raw capacity, they sell guarantees: cryptographic assurances that a light client can verify availability without downloading the whole dataset, or legal and social commitments that data will be retained for a defined period. That is a real product, but it is a compliance-and-cryptography product, not a throughput product, and it commands a different kind of margin.
Settlement, by contrast, has proven stickier than expected. Rollups that anchor to a large, liquid, credibly neutral base layer inherit its security assumptions and its network effects. The cost of settling elsewhere is not primarily technical — it is the loss of composability with everything already there. This is why, despite years of predictions about settlement fragmentation, the majority of economically significant rollups continue to post their proofs and their state roots to a single dominant chain. The moat is not technical; it is gravitational.
The layer that has genuinely gained strategic importance is sequencing. Whoever sequences a rollup sees its transactions first. They can extract MEV, they can censor, they can offer preconfirmations, and they can sell priority. As execution margins compress, sequencing becomes the profit center. This has produced a wave of experimentation — shared sequencers, based sequencing that delegates ordering back to the base layer, and encrypted mempools that attempt to remove the informational advantage entirely. Each option trades off revenue against credible neutrality, and no consensus has emerged on the correct balance.
The unresolved question in modular architecture is where the accountability sits. When a rollup fails, users file complaints against the rollup operator. When the DA layer fails, they file complaints against a different team. When the settlement layer reorgs, a third party is responsible. Splitting the stack into four vendors has clear engineering benefits and a clear governance cost: nobody owns the whole system, and therefore nobody owns the failure. Regulators and institutional users have begun to notice, and it is shaping which architectures are considered acceptable for regulated activity.
3. Rollup economics after the subsidy era
Rollups were sold to the market as businesses. They would collect fees from users, pay a smaller amount for base-layer data and proof generation, and pocket the difference. For a while, that arithmetic appeared to work because fees were high and data costs were non-trivial but manageable.
Two things broke the model. The first was the collapse in data costs, which removed a cost line but also removed the justification for charging users a premium. The second was competition. Dozens of general-purpose rollups launched within a compressed window, each with a token, each with an incentive program, each competing for the same finite pool of liquidity and users. The result was a subsidy war in which rollups paid users — through token emissions, points programs and yield boosts — more than they collected in fees. The gap was funded by treasury sales and venture capital.
That model is now under visible strain. Token prices across the sector have compressed, which raises the real cost of emissions, which forces a choice between reducing incentives and extending runway. Teams that chose to extend runway by shrinking incentives have watched activity migrate to whichever chain is currently paying the most. The churn reveals how much of the reported growth was rented.
The teams that have navigated this transition best share a common trait: they stopped describing themselves as chains and started describing themselves as products. A rollup that exists to host a specific application, or a specific class of application, can charge for the value it delivers rather than trying to win a generic throughput contest. A general-purpose rollup competing on transaction cost alone is competing in the one dimension where its product is most easily substituted.
Proof generation is the other variable. Zero-knowledge proofs are computationally expensive, and the cost falls on the operator. Hardware acceleration and improved proving systems have driven that cost down materially, but it remains a real expense that optimistic rollups avoid entirely by accepting a challenge period. The trade-off — capital efficiency and fast finality against ongoing proving costs — is not settled, and different applications reasonably choose different answers. Payments infrastructure needs fast finality. A long-tail application with low transaction volume does not.
A third pressure is emerging from shared sequencing and based rollup designs, which route ordering back to the base layer. These designs reduce the operator's ability to capture MEV, which is exactly the point, but they also reduce the operator's revenue. A rollup that adopts based sequencing is making a credible-neutrality commitment with a measurable price tag. Whether the market rewards that commitment with enough volume to offset the lost revenue is one of the more consequential open questions in the sector.
4. Restaking and the price of shared security
Restaking was the most capital-efficient idea of the last cycle and remains the most difficult to reason about. The premise is straightforward: an asset already staked to secure one network can be simultaneously committed as economic security to additional services, allowing those services to bootstrap without issuing their own token or building their own validator set.
The mechanism works. The question is what it actually buys. Economic security is not the same thing as security. A restaked protocol can claim that a large amount of value is at risk if its operators misbehave, but that claim is only as strong as the slashing conditions and the willingness of operators to accept them. If an operator is restaked across twenty services and one of them triggers a slashing event, the operator's entire stake is impaired. The rational operator, aware of this, either refuses the risky service or prices the risk accordingly — which pushes the riskiest services toward exactly the operators with the weakest risk management.
That correlation is the structural weakness. Restaking does not create new security; it reallocates existing security across more claims. The aggregate amount of capital at risk does not grow. What grows is the number of parties with a claim on that capital, and the number of scenarios in which a single adverse event cascades across unrelated protocols. During calm markets, the system looks robust. During stress, the correlation becomes visible all at once, which is precisely when it matters most.
The mitigations are real but partial. Operators can impose service-specific caps. They can require additional collateral for high-risk services. They can build reputation systems that price the risk of delegation to specific operators. Each of these adds sophistication to a market that is still thinly understood by the delegators who supply most of the capital. Delegators frequently cannot evaluate the slashing conditions of the services their stake secures, and the interfaces they use present restaking as a yield product rather than a contingent liability.
The genuinely valuable use case for restaking is narrower than the market narrative suggests: it is best suited to services whose failure is expensive but recoverable, and whose operations are simple enough to be verifiable. Light-client bridges, oracle networks, and data-availability attestations fit that description. High-value settlement, where failure is catastrophic and irreversible, does not, and the services that operate there will need dedicated security regardless of what the restaking markets offer.
The honest summary is that restaking has converted a question of protocol design into a question of risk management, and then handed that risk to retail delegators. That is not inherently wrong, but it is a meaningful transfer of responsibility that has not been matched by a transfer of information.
5. Intents, solvers and the invisible exchange
The user-facing experience of decentralized trading has improved dramatically without most users noticing the reason. The shift is from imperative to declarative: instead of specifying a sequence of on-chain steps, a user declares a desired end state — I want this asset, subject to these constraints — and a competitive market of solvers figures out how to achieve it.
This architecture, generally described as intent-based, is a genuine improvement in abstraction. It hides gas estimation, route selection, slippage management and cross-domain complexity behind a single signature. For a retail user on a mobile wallet, the difference between a raw AMM interaction and an intent-based swap is the difference between a tool and a product.
It also relocates the value capture. In a model where users submit intents and solvers compete to fill them, the solver sees the order before anybody else. The solver decides whether to route to an automated market maker, to a centralized exchange, to an off-chain counterparty, or to a private liquidity pool. The user receives the quoted outcome; the difference between that quote and the true best execution is captured by the solver. In a competitive market with many solvers, that spread should compress toward the cost of execution. In practice, solver markets are concentrated, because solvers need capital, infrastructure and reputation to compete, and those are not equally distributed.
The concentration is not automatically harmful. A concentrated solver market with transparent auction mechanics can deliver better execution than a fragmented one, because the winning solver is typically the one with the best access to liquidity. But concentrated solver markets create dependencies that are difficult to unwind. If a small number of solvers handle the majority of order flow, then those entities occupy a position structurally similar to a wholesale dealer network — essential infrastructure that users never see and cannot easily audit.
Encrypted mempools and threshold decryption attempt to address the informational advantage directly by making it impossible for anyone, including the solver, to view an order before it is committed. The trade-off is latency and complexity. A system where orders cannot be inspected until execution is a system where the solver cannot offer price improvement through off-chain routing, because the solver has no time to route. The design space is genuinely difficult, and the current market has not converged on a dominant answer.
What is clear is that order flow is now a first-class asset in the blockchain economy, comparable in importance to liquidity. Protocols that control order flow exercise influence far beyond their own transaction volume, and the regulatory treatment of that influence — is a solver a broker? a dealer? an exchange? — remains undefined in most jurisdictions. That ambiguity is a live risk for anyone building in the space.
6. Stablecoins as the actual product-market fit
Strip away the narrative cycles and one application has demonstrated sustained, compounding, real-world demand: dollar-denominated tokens used for payments and settlement.
The demand is not speculative. It comes from businesses that need to pay contractors in jurisdictions with slow or unreliable banking, from remittance corridors where the legacy fee is measured in double digits, from treasuries that want to hold dollar exposure without opening a foreign bank account, and from trading operations that need instant settlement across venues without pre-funding accounts in each. In each case the value proposition is concrete: faster, cheaper, and available outside the hours of the traditional banking system.
Issuer economics are therefore straightforward and unusually attractive. The issuer holds reserves — typically short-dated government securities and cash equivalents — and pays the holder nothing, while collecting the full yield on those reserves. In a high-rate environment that spread is enormous relative to the operating cost of the business, which explains why so many entrants have moved into the category. It also explains why the business is cyclical in an underappreciated way: when short rates fall, issuer revenue compresses sharply while operating costs do not, and the whole model looks different.
That asymmetry has driven the most important structural debate in the sector: whether stablecoin holders should receive a share of reserve yield. From the issuer's perspective, paying yield turns a float business into a margin business and invites securities-classification questions. From the holder's perspective, a token that pays nothing while the issuer earns the risk-free rate is a transfer of value that becomes harder to justify as competition intensifies and as tokenized money-market funds offer a yielding alternative. Regulators in several jurisdictions have moved toward restricting yield-bearing stablecoins, which would settle the question in favor of the issuers and push yield-seeking holders toward separate instruments.
Regulatory clarity has arrived unevenly but decisively. Frameworks that define reserve composition, redemption rights, disclosure obligations and permissible activities have moved stablecoins from a gray zone into a licensed perimeter in a growing number of markets. For large issuers this is a competitive advantage — compliance costs are fixed and therefore regressive, and a regime with strict reserve rules disadvantages smaller entrants. For the industry as a whole it is a normalization: stablecoins are becoming a regulated payments utility, not a crypto-native curiosity.
The unresolved competitive question is distribution. An issuer with a strong consumer app, a large marketplace, or a captive merchant network can push its stablecoin into circulation in ways a purely crypto-native issuer cannot. That favors technology companies with existing payment surfaces over blockchain-native teams, and it suggests that the long-run stablecoin market will look less like a protocol competition and more like a payments-network competition.
7. Tokenized treasuries and the gravity of yield
The tokenization of real-world assets has been a reliable conference topic for nearly a decade and a marginal business for most of that time. That has changed, though not in the way the early advocates imagined.
The growth has not come from tokenizing exotic assets — real estate, invoices, fine art. It has come from tokenizing the most boring instrument in finance: short-dated government debt. Tokenized money-market funds now represent a meaningful pool of assets, and their appeal is uncomplicated. They offer a dollar-denominated yield, daily or intraday liquidity, and the ability to be used as collateral inside on-chain credit markets without a banking intermediary.
This is genuinely useful, and it is also a distribution story rather than a decentralization story. The issuer is a regulated asset manager. The custody is with a regulated custodian. The transfer agent performs a compliance check on every transfer. The token is a wrapper around a traditional financial product, and the wrapper exists to provide programmability and 24/7 settlement, not to remove intermediaries. That is fine — it is what institutional users actually want — but it means the value of these products is captured largely by the issuers, not by tokenholders of whatever chain hosts them.
The strategic implication for blockchains is uncomfortable. If the highest-quality assets on a network are tokenized treasuries, and those assets generate their yield from off-chain government debt, then the network's economic activity is a thin veneer over traditional finance. That is not a criticism; it is a description of where the volume is. But it does mean that the health of the on-chain economy becomes correlated with the health of the traditional financial system, and specifically with the level of short-term interest rates. Tokenized treasuries are a spectacular product in a high-rate environment and a much weaker one when the risk-free rate approaches zero and the on-chain alternatives become more attractive by comparison.
The collateral question is where the real architectural work is happening. A tokenized treasury that can be pledged as collateral in an on-chain lending market, subject to haircuts and liquidation rules, is a bridge between two financial systems. Building that bridge requires solving problems that the crypto industry has historically avoided: legal enforceability of collateral claims in bankruptcy, valuation of instruments that do not trade continuously, and the interaction between on-chain liquidation mechanics and off-chain settlement cycles. These are unglamorous problems with long timelines, and they are the actual barrier to institutional adoption.
8. Regulation enters the framework phase
The most consequential change in the operating environment over the past several years is not technological. It is that major jurisdictions have moved from enforcement-led ambiguity toward written rules.
The shift matters because ambiguity was expensive. A protocol team that could not determine whether its token was a security had three options: spend heavily on legal structuring, restrict itself to jurisdictions with favorable treatment, or operate and hope. All three were costly, and the third carried existential risk. Written frameworks — however imperfect — replace an unbounded risk with a bounded one, and bounded risk can be priced, insured, and financed.
The frameworks that have emerged share a common architecture even where they differ in detail. They distinguish between the asset and the activity. A token's classification depends less on its intrinsic properties than on how it is sold, marketed and distributed. Intermediaries — exchanges, custodians, brokers — face licensing, capital and disclosure obligations. Issuers of stablecoins face reserve and redemption requirements. Decentralized protocols occupy an uncomfortable middle ground that the legislation has generally not resolved, leaving it to enforcement discretion and future rulemaking.
The consequence is a bifurcation of the industry into a regulated perimeter and an offshore remainder. Inside the perimeter, compliance costs are high and competition is limited to entities that can absorb them. That is not necessarily bad for incumbents — regulatory cost is a moat — but it does mean that the next generation of infrastructure will be built by better-capitalized teams, which favors consolidation.
Outside the perimeter, activity continues, but the terms of engagement have changed. Access to banking, to fiat on-ramps, and to institutional liquidity increasingly depends on demonstrable compliance. A protocol that cannot show who its users are, where its front-end is hosted, and how it handles sanctions exposure will find itself progressively excluded from the parts of the market where real money sits. Decentralization is a genuine architectural property, but it is not a legal shield, and the enforcement record of the past several years has made that clear.
The most underappreciated regulatory risk is not a ban. It is the slow accretion of compliance requirements around the points where crypto touches the traditional financial system — banks, custodians, payment processors, market makers. Each individual requirement is reasonable. Collectively they determine which applications are viable, and those decisions are being made by supervisors rather than legislators, with limited transparency and limited appeal.
9. Credit returns to DeFi, carefully
On-chain lending never went away, but its character has changed. The first generation was dominated by overcollateralized pools with algorithmic interest rates, efficient for liquid assets and useless for anything else. The current generation is more interesting and more dangerous.
The mechanism that has driven the change is the risk curator model. Instead of a single monolithic pool where depositors share all risk equally, curated vaults allow a specialist to define a strategy — which collateral, which borrowers, which liquidation thresholds — and to attract depositors who accept that risk profile in exchange for a share of the yield. Depositors can diversify across curators. Curators build reputations, and their ability to attract capital depends on their track record.
This is a genuine financial innovation, and it is also a re-creation of a structure that existed before the 2008 crisis. The curator is functionally a portfolio manager. The depositor is functionally an investor in a fund. The difference is that the curator operates without capital requirements, without a fiduciary standard enforceable in any obvious way, and without the disclosure regime that applies to a registered fund manager. When a curator's risk assessment is wrong, the loss falls on depositors who may not have understood what they were accepting.
The overcollateralization that protected early DeFi has not disappeared, but it has been stretched. Loan-to-value ratios on volatile collateral are tighter than they once were, which is prudent. But the introduction of restaked assets, liquid staking derivatives and tokenized treasuries as collateral introduces new correlation risk: assets that appear diversified may all decline together in a stress event, and liquidation mechanisms that rely on the existence of a liquid market may fail precisely when that market is needed most.
Undercollateralized credit — the genuinely transformative use case — remains small and institutionally intermediated. It requires identity, legal recourse, and a mechanism for enforcing claims against a borrower in a specific jurisdiction. Every one of those requirements is at odds with the permissionless ethos of the underlying infrastructure, and the teams building in this space have made pragmatic compromises: allowlisted borrowers, off-chain legal agreements wrapped around on-chain settlement, and credit assessment performed by entities that are not decentralized in any meaningful sense.
That is not a failure. It is what credit actually requires. The honest framing is that DeFi has discovered that the hard part of lending was never the smart contract; it was the assessment of whether a borrower will repay. Smart contracts made settlement cheap. They did not make credit judgment cheap, and the industry has spent several years rediscovering that distinction.
10. The developer and user picture
Metrics in this industry are notoriously easy to manipulate and notoriously hard to interpret. With that caveat, some patterns are robust enough to be worth stating.
Developer activity has shifted away from protocol-layer work and toward application-layer and tooling work. The number of teams building a new layer-one has fallen sharply, which is unsurprising given the economics described above. The number of teams building on existing infrastructure — wallets, payment applications, developer tooling, compliance infrastructure — has held up better, and the quality of the tooling has improved substantially. Deploying a contract, indexing its events, and building a front-end for it is now a matter of hours rather than weeks, and the frameworks that enable that are mature in a way they were not five years ago.
User activity tells a more complicated story. Headline wallet counts and transaction counts are inflated by incentive programs and by a small number of automated actors. The metric that matters — retained users who transact repeatedly without a subsidy — is difficult to measure and almost never disclosed. What can be observed is that the applications retaining users tend to be those with a clear non-speculative utility: payments, savings, trading, and in some markets, gaming. Applications whose only function is to generate yield from a token subsidy have not retained users once the subsidy ends, which was entirely predictable and predicted.
The user-experience improvement is real and underrated. Account abstraction has made it possible to pay fees in the asset being transferred, to recover a lost account through social mechanisms, and to batch multi-step operations into a single confirmation. Embedded wallets inside consumer applications have removed seed phrases from the critical path for a large class of users. These are not glamorous changes, and they do not generate headlines, but they address the single largest barrier to mainstream adoption, which was never scalability and always usability.
The remaining gap is trust and recovery. A user who loses access to an account in a traditional financial system has recourse. A user who loses access to a self-custodied wallet generally does not. Every serious attempt to solve this problem — social recovery, custodial delegation, multisignature arrangements — reintroduces some form of intermediary, and the industry has not yet found a design that provides both recoverability and credible self-sovereignty. This is the most important unsolved problem in consumer crypto, and it is not primarily a technical one.
11. What to watch
Every cycle generates its own set of signals, and most of them turn out to be noise. The following are the ones with genuine information content.
Fee revenue net of incentives. A chain's gross fees are meaningless if they are recycled back to users through token emissions. The relevant figure is fee revenue minus incentive spend, and it is disclosed by almost nobody. Where it can be estimated, it separates businesses from subsidy programs.
Sequencer decentralization progress. Promises are cheap. The observable indicators are whether a rollup has a functioning decentralized sequencer set with a real threshold for liveness, whether forced inclusion is available to users, and whether the operator can censor transactions unilaterally. A rollup that cannot answer those questions is a centralized service with a blockchain interface.
Stablecoin supply growth by corridor. Aggregate supply understates what matters. Growth in specific remittance corridors and in business-to-business settlement indicates genuine utility; growth concentrated in trading venues indicates leverage. The two have very different implications for durability.
Treasury runway across major protocols. Many treasuries were denominated in native tokens and have lost most of their value. A protocol with two years of runway at current prices and a year of runway at a fifty percent drawdown is in a materially different position from one that presents the same headline number. Runway disclosure normalized to a stressed price is the single most useful governance metric in the sector.
Unlock schedules relative to liquidity. Vesting cliffs continue to be a dominant driver of price action, and the relationship between unlock size and available liquidity matters more than the absolute size of the unlock. Projects that have restructured their schedules to align with liquidity conditions have generally fared better than those that did not.
Restaking concentration. The share of restaked capital delegated to the top handful of operators is a direct measure of correlated slashing risk. Concentration in that market is invisible in normal conditions and decisive in a crisis.
Regulatory perimeter expansion. The specific rulemakings to watch are those that touch the boundary between crypto and banking: custody rules, capital treatment for crypto exposures, payment-stablecoin reserve standards, and the treatment of decentralized protocols under market-structure frameworks. Each of these determines which applications can reach mainstream users.
Conclusion
The industry spent a decade building infrastructure on the assumption that blockspace would be scarce and therefore valuable. It solved the scarcity problem, and in doing so it invalidated the business model that funded the solution.
What is emerging in its place is less dramatic and more durable. Execution is a commodity. Settlement, order flow, compliance and distribution are the scarce resources. Stablecoins have found real product-market fit and are becoming a regulated payments utility. Tokenized treasuries have brought institutional balance sheets on-chain, along with their interest-rate sensitivity. Restaking has redistributed risk without creating security. Regulation has replaced ambiguity with a bounded, expensive, and largely workable set of rules.
The uncomfortable implication for investors is that the returns from here are more likely to accrue to businesses than to protocols. A protocol that charges nothing and pays for its own security through inflation is not a business; it is a public good with a token attached. The entities that will compound are those that control something genuinely scarce — a settlement network with real liquidity, an order-flow franchise, a licensed payments rail, or a distribution channel that reaches users who do not care about the underlying architecture.
That is a less exciting story than the one the last cycle told. It is also the one the technology has been pointing toward all along.