The Composability Question: Is Synchronous Atomic Composability a Solution Looking for a Problem?

tl;dr

• Synchronous atomic composability is not useless — it is misframed. The EEZ and similar projects position composability as the answer to Ethereum’s fragmentation problem. The data suggests the real problem is consolidation: markets are resolving fragmentation by killing weak chains, not by connecting them.
• For trading, product focus beats composability. Every DEX winner — Hyperliquid (70%+ perps OI), Solana ($117B spot), Aerodrome (50-63% of Base volume) — won by building excellent products on single chains. None needed cross-chain atomic execution.
• Cross-chain bridge volume confirms this. An estimated $18-20B monthly vs. $231B+ in intra-chain DEX volume (~8-9%). Users go where the products are and stay there.
• For non-trading DeFi, the case is real but narrow. Cross-chain liquidations, unified liquidity pools, and complex programmable finance workflows are genuine use cases where atomicity matters — but they represent the edge of DeFi activity, not the center.
• Asynchronous intents solve most of the remaining gap. ERC-7683 and similar standards handle ~90% of cross-chain UX needs. The EEZ addresses the remaining 10% — the question is whether that 10% justifies the engineering complexity.
• The EEZ lacks a product thesis. “Shared liquidity” is a feature, not a product. Without explicit economic terms for why L2s would opt in — what they gain, what they sacrifice, who subsidizes proving costs — the project risks repeating the IBC/Polkadot pattern: technically sound, economically hollow.
• The strongest version of the thesis: composability matters at the margins of DeFi infrastructure, but markets have already solved the trading problem through simpler means. Whether those margins grow depends on market structure evolution that has not yet occurred.

Table of Contents

1. Introduction: The Question Worth Asking
2. Section 1: The CEX-DEX Landscape – Where Volume Actually Lives
3. Section 2: DEX Winners – Product Focus Over Composability
4. Section 3: Cross-Chain Activity – The Bridge Volume Reality Check
5. Section 4: L2 Consolidation and the Zombie Chain Problem
6. Section 5: The EEZ and the Composability Paradox
7. Section 6: Agentic Trading – Separating Signal from Noise
8. Section 7: The Strongest Counterarguments
9. Section 8: Synthesis – What Markets Actually Need
10. Data Sources & Methodology

Introduction: The Question Worth Asking

There is an implicit assumption embedded in much of Ethereum’s scaling roadmap: that fragmentation across L2s is the core problem, and that restoring synchronous composability – the ability for smart contracts across chains to atomically interact within a single transaction – is the solution. It is the premise behind projects like Gnosis’s Ethereum Economic Zone (EEZ), announced March 29, 2026 at EthCC Cannes, co-funded by the Ethereum Foundation.

This report tests a different thesis: what if synchronous atomic composability is over-engineered for what markets actually need?

The claim is not that composability is useless. The claim is more specific: that the trading venues which have won market share did so through product focus on specific chains, not through cross-chain orchestration. That most L2 activity is retained through loyalty and network effects, not through the promise of interoperability. That the data, as it exists today, suggests markets function – and function well – without the kind of synchronous composability the EEZ promises.

This is a thesis worth stress-testing, because if it’s right, significant engineering effort and capital may be flowing toward a problem that markets have already solved through simpler means. And if it’s wrong – if there are domains beyond spot and perps trading where synchronous composability is genuinely load-bearing – then the thesis needs to be narrowed, not abandoned.

Let’s look at the data.

Section 1: The CEX-DEX Landscape – Where Volume Actually Lives

The first-order observation is straightforward: centralized exchanges dominate trading, and they do so without any composability at all.

CEX vs. DEX Market Share (January 2026)

CEX Volume Rankings (Aug 2025 - Jan 2026 Cumulative)

Combined CEX volume hit $5.61T in February 2026 – a 16-month low, split $1.50T spot and $4.11T derivatives. Even at a cyclical low, this dwarfs all DEX activity combined.

But the trend line matters more than the snapshot. DEX spot share doubled from 6.9% in January 2024 to 13.6% by January 2026, peaking at 24.5% in June 2025. DEX perps share grew 5x over the same period: 2.0% to 10.2%. The total perps market (CEX + DEX) reached $7.24T monthly in January 2026.

[CHART: Line graph — CEX vs. DEX market share trajectory, Jan 2024 to Jan 2026. Two series: spot (6.9% → 13.6%) and perps (2.0% → 10.2%). Highlight the June 2025 spot peak at 24.5%.]

This trajectory suggests something structural. DEX venues are not merely capturing overflow from CEX downturns – they are building persistent market share. The question is how they are building it, and whether cross-chain composability plays any role in that story.

The answer, so far, is no.

Section 2: DEX Winners – Product Focus Over Composability

The DEX venues that have gained the most ground share a common trait: they built excellent products on specific chains and let users come to them. None of them won through cross-chain composability.

DEX Spot Volume by Chain (February 2026)

DEX Spot Volume – Cumulative (Aug 2025 - Jan 2026)

[CHART: Horizontal bar — DEX spot volume by chain, February 2026. Solana ($117B), BNB (~$90B est.), Ethereum ($52B), Base (~$21B est.), Arbitrum (~$4.4B est.). Title: “Single-Chain Dominance: DEX Spot Volume by Chain.”]

Solana’s story is pure product-market fit. $117B in February 2026, led by memecoin trading infrastructure that offers sub-second finality and negligible fees. The majority of that volume — estimated at 60%+ — was memecoins. This is not a composability story – it is a speed-and-UX story. Solana’s DEX ecosystem did not need to compose with Ethereum or any other chain. It built something users wanted on a single execution environment. (The fragility is also real: Solana’s weekly volume collapsed 62% in three weeks from its $118.2B weekly peak, demonstrating how memecoin-driven volume can evaporate.)

Hyperliquid is the most striking case. It captured 70%+ of DEX perpetual futures open interest and reached $178B in monthly volume ($6.27B daily), with $1.59T in cumulative volume over just six months. It is the only DEX that ranks in the top 10 of all exchanges (CEX and DEX combined). Hyperliquid achieved this by building a purpose-built L1 with a custom order book – the exact opposite of the composability thesis. It now trades S&P 500 perps and oil commodities 24/7, expanding the addressable market of on-chain derivatives beyond crypto-native assets.

Its market share has declined from ~80% to ~38% as competitors like Lighter ($1.59B daily), Aster, and EdgeX emerge. But the competitive dynamic reinforces the point: these competitors are also building focused products on specific chains, not cross-chain composable systems.

DEX Perps Landscape

Uniswap remains dominant in spot across Ethereum and L2s – $0.54T cumulative over six months. Its strength is protocol ubiquity and liquidity depth on Ethereum mainnet, not cross-chain orchestration. Uniswap deploys on multiple chains, but each deployment is essentially independent.

Aerodrome (rebranding to Aero in Q2 2026 following its November 2025 merger with Velodrome) is perhaps the most instructive example for the composability debate. It accounts for 50-63% of all Base volume, having built a ve(3,3) flywheel that generates deep, sticky liquidity on a single L2. Aerodrome does not need to compose with Ethereum L1 or other rollups. Its liquidity model is self-reinforcing within its own chain.

The pattern is consistent: winners build focused products that create gravitational pull within a single execution environment. Users bridge to where the product is. They do not wait for composability to bring the product to them.

The Multicoin Thesis Shift – Composability as Table Stakes

Perhaps the most telling signal comes from Multicoin Capital, which has real skin in this game. Multicoin built its original Solana investment thesis around atomic composability as the killer affordance – the reason monolithic chains like Solana would beat modular architectures. Team members famously featured “composability” in their Twitter usernames. Kyle Samani’s keynote at the 2021 Multicoin Summit was dedicated to the concept. Their third Solana thesis, “Technical Scalability Creates Social Scalability,” and fourth, “The Hidden Costs of Modular Systems,” both foregrounded composability as a decisive competitive advantage.

Then something shifted. In January 2025, Multicoin published its fifth Solana thesis: “Internet Capital Markets.” In this iteration, atomic composability is mentioned just three times – each in passing, as a background property rather than a central argument. The bulk of the thesis focuses on execution quality, market microstructure (conditional liquidity), latency optimization, and total addressable market expansion against TradFi incumbents.

This is not a public retraction. Multicoin did not issue a mea culpa. But the reweighting is unmistakable: composability went from being the argument to being a property – from thesis centerpiece to table stakes. The firm that bet its reputation on composability as the killer feature of monolithic chains tacitly acknowledged that what actually wins markets is execution quality, not composability. The market taught them this. Hyperliquid, which has no composability with anything, was eating the perps market alive. Jupiter and Raydium were winning on speed and UX, not on the ability to atomically compose across chains. When the most prominent composability evangelist in crypto stops leading with composability, that is a data point worth registering.

A necessary caveat for intellectual honesty: Multicoin’s fifth thesis focuses heavily on competing with TradFi – equities, forex, commodities. Their de-emphasis on composability may partly reflect that TradFi incumbents simply do not care about it; when your competitive framing is “Solana vs. NYSE,” atomic composability between DeFi protocols is not the selling point. It is possible Multicoin still views composability as critical for crypto-native use cases while deprioritizing it in their TradFi-facing narrative. But even granting this, the shift is telling: the most commercially significant framing of Solana’s value proposition no longer centers composability. Whether that is because composability is less important or because TradFi does not need convincing of it, the practical result is the same – the market’s attention has moved to execution quality.

Section 3: Cross-Chain Activity – The Bridge Volume Reality Check

If cross-chain composability were critical to how markets function, we would expect bridge volumes to be large relative to intra-chain trading. They are not.

Bridge Volume Comparison

Major Bridge Protocols

[CHART: Stacked bar — Intra-chain DEX volume ($231B+) vs. cross-chain bridge volume (~$18-20B est.) for January 2026. Title: “Bridge Volume: ~8-9% of Intra-Chain DEX Activity.”]

Monthly bridge volume of an estimated $18-20B is roughly 8-9% of intra-chain DEX spot volume. This is not zero – bridges are used, and growing (CCIP’s 62% month-over-month increase is notable). But it suggests that the vast majority of trading activity happens within single execution environments. Users are not constantly shuffling capital across chains to access opportunities. They park liquidity where they trade and stay there.

The composition of bridge volume is also telling. DeBridge routes 40% of its volume through Tron USDT – this is stablecoin transfer infrastructure, not DeFi composability. Much of what bridges facilitate is simple asset movement, not the kind of synchronous contract-to-contract interaction that composability solutions like the EEZ would enable.

This is not to say bridges are unimportant. For a user who wants to move USDC from Arbitrum to Base to LP on Aerodrome, bridges are essential infrastructure. But that workflow is asynchronous. It does not require atomic composability. The user sends, waits, receives, and then interacts with the destination chain. ERC-7683 (cross-chain intents) and similar standards can make this flow faster and more user-friendly without requiring synchronous execution guarantees.

Section 4: L2 Consolidation and the Zombie Chain Problem

The L2 landscape is consolidating rapidly, and the consolidation pattern undermines one of the core motivations for synchronous composability.

L2 Market Share

[CHART: Pie chart — L2 transaction share. Base (60%+), Arbitrum (~20%), Optimism (~10%), All Others (<10%). Title: “L2 Consolidation: Three Chains, 90% of Activity.”]

The thesis that Ethereum needs synchronous composability presupposes a thriving multi-rollup ecosystem where value is meaningfully distributed across many chains that need to interoperate. The data tells a different story: three rollups capture ~90% of transactions, and the long tail is dying. 21Shares has stated publicly that “most Ethereum L2s may not survive 2026.”

Base alone accounts for 60%+ of L2 transactions and was the only profitable L2 in 2025, earning $55M. Its dominance is not driven by composability with other rollups – it is driven by Coinbase’s distribution, Aerodrome’s liquidity, and a coherent product strategy.

If L2 activity continues to consolidate into 2-3 dominant chains, the composability problem becomes less acute. The “fragmentation” that the EEZ aims to solve may resolve itself through market selection rather than protocol engineering. The $40B+ in value that Gnosis cites as “siloed across 20+ L2s” is increasingly concentrated in a handful of venues where it is already usable without cross-chain composability.

This does not mean the zombie chain problem is solved – it may simply be that the market is solving it by letting weak chains die.

Section 5: The EEZ and the Composability Paradox

On March 29, 2026, Gnosis’s Friederike Ernst and Jordi Baylina (Zisk) announced the Ethereum Economic Zone at EthCC Cannes. The EEZ enables synchronous composability between Ethereum L1 and connected rollups via real-time ZK proving. Smart contracts on connected rollups can call L1 contracts with the same guarantees as if they were on Ethereum itself. ETH serves as the default gas token with no additional bridging needed.

The founding members – Aave, Titan, Beaver Build, Centrifuge, xStocks – are serious. The structure is a Swiss non-profit, co-funded by the Ethereum Foundation. Ernst framed it directly: “Ethereum doesn’t have a scaling problem. It has a fragmentation problem.”

This is a credible project. But credibility does not resolve the core tension: intent-based standards like ERC-7683 already solve the vast majority of cross-chain UX problems through asynchronous flows. The EEZ addresses the remaining slice where synchronous guarantees are genuinely necessary. The question is whether that slice justifies the engineering complexity.

There is a reasonable case that it does – but primarily outside of trading. The use cases where synchronous composability is most compelling are:

• Cross-chain liquidations: A lending protocol on an L2 that needs to access L1 price feeds and execute liquidations atomically, without the latency risk of async bridges.
• Unified liquidity pools: An AMM that pools liquidity across L1 and multiple rollups in a single logical position, enabling deeper markets than any single chain can support.
• Programmable finance workflows: Complex multi-step DeFi transactions (flash loans that span chains, collateral that lives on one chain securing a position on another) that are infeasible without atomic guarantees.

These are real use cases. But they are not trading volume use cases. They are infrastructure and risk management use cases. The EEZ’s value proposition may be strongest in precisely the domain where the trading volume data is least relevant.

The Missing Product Thesis

There is a deeper critique of the EEZ that neither composability maximalists nor skeptics are articulating clearly: “shared liquidity” is a feature, not a product. The EEZ has no articulated theory of what products become possible — and economically viable — that do not exist today.

A product thesis would be specific: “Aave can atomically liquidate cross-chain collateral, reducing its risk premium by X basis points, enabling Y% better lending rates.” Or: “An AMM can pool liquidity across five rollups, achieving Z% less slippage than any single-chain deployment.” The EEZ makes neither of these claims. It says “fragmentation is the problem, composability is the solution” and stops there. That is an infrastructure thesis, not a product thesis.

Before building the technical architecture, the right first-principles question is: under what economic terms would it be rational for an L2 to sacrifice execution independence in exchange for shared state with L1? Specifically:

• What does the L1 get from an L2 joining the EEZ? Fees? MEV flow? Settlement demand for ETH?
• What does the L2 get? Shared liquidity? Credibility? Access to L1 protocols?
• What does each party give up? Sovereignty? Sequencer revenue? Execution independence?
• Who subsidizes the real-time proving costs?
• What happens when economic interests diverge — when an L2 wants to capture value that the EEZ framework routes to L1?

These are the questions that determine whether anyone actually opts in beyond the founding members. The Swiss non-profit structure is a governance answer, but governance without explicit economic terms is a committee, not a trade agreement.

We have seen this movie before. Cosmos IBC works beautifully as technology — and the cross-chain DeFi explosion it was supposed to enable never materialized because there was no economic framework for why value should flow between chains rather than concentrating on the best single chain. Polkadot’s parachains are technically sound and economically hollow. The pattern is consistent: solve the hard computer science problem, assume economic activity follows, watch as it does not.

The design question the EEZ has not answered is whether the technical architecture should be built around an understanding of trade agreements — mutual obligations, value flows, dispute resolution, exit terms — rather than around solving the hardest possible proving problem and hoping the economics sort themselves out. History suggests the economics do not sort themselves out. They get solved by products with specific users, specific margins, and specific reasons to exist — or they do not get solved at all.

Section 6: Agentic Trading – Separating Signal from Noise

Proponents of synchronous composability often cite a future dominated by autonomous AI agents as the ultimate catalyst for cross-chain execution – agents that need to atomically move capital across chains, compose protocols, and settle trades without human intervention. If this future materializes, it would be the strongest demand-side argument for synchronous infrastructure. The data, however, does not support this narrative.

AI agent tokens – exemplified by the Virtuals ecosystem – have collapsed. Virtuals tokens fell 86% from their all-time high ($5.07 to $0.72 as of reporting). Virtuals protocol revenue cratered from a $3.9M monthly peak.

Claims that “60-80% of crypto volume is AI-driven” conflate two very different things. Traditional algorithmic trading – market-making bots, arbitrage bots, MEV searchers – has existed since at least 2020. These are sophisticated programs, but calling them “agentic” is rebranding, not innovation. MEV bots have been extracting value on Ethereum since before DeFi Summer. The term “agentic” implies novel autonomous decision-making capacity. There is no verified evidence that meaningful trading volume comes from truly autonomous AI agents operating independently of human-defined strategies.

The token creation data tells its own story: 24.04 million tokens were created between January 2025 and January 2026. Pump.fun alone created 5.01 million. This is not agent-driven innovation – it is speculation infrastructure operating at scale, generating volume through token creation and immediate trading, overwhelmingly by human actors (or bots following human-defined parameters).

If agentic trading were producing material volume, we would expect to see it reflected in on-chain metrics distinct from bot activity. We do not. The agentic trading narrative, at present, is a category error: repackaging existing automated trading under a new label and layering a composability requirement on top that the underlying activity does not demand.

Section 7: The Strongest Counterarguments

The thesis – that synchronous composability is over-engineered for market needs – is strongest when applied to trading venues. But it has genuine weak points that deserve honest engagement.

Trajectory vs. Snapshot

DEX perps grew from 2.0% to 10.2% of the total market in two years. DEX spot doubled its share. If this trajectory continues, DEXs could capture 20-30% of trading within 2-3 years. At that scale, the infrastructure constraints that are invisible at 10% share – liquidity fragmentation, capital inefficiency across chains, latency in cross-chain settlement – may become binding. Synchronous composability might not be needed today, but the market of 2028 might look very different.

The Fragility of Single-Chain Winners

Hyperliquid runs on a single validator set. Solana has a documented history of network outages. The very product focus that makes these venues compelling also concentrates risk. A catastrophic failure on Hyperliquid’s chain would take 70%+ of DEX perps open interest with it. A multi-chain system with synchronous composability would, in theory, distribute that risk. This is not a current market need – it is a risk management argument for the future.

Non-Trading DeFi Requirements

Lending protocols, insurance markets, and structured products have different composability requirements than spot and perps trading. A lending protocol like Aave (notably an EEZ founding member) that operates across 12+ chains faces real fragmentation: liquidity is split, oracle latency introduces liquidation risk, and governance must be coordinated across deployments. Synchronous composability would allow Aave to operate as a single logical protocol across chains rather than a federation of independent deployments. This is a genuine problem that async intents do not fully solve.

Cross-Chain Arbitrage and MEV: The Market Efficiency Argument

This is perhaps the most technically sound counterargument and one this report would be incomplete without addressing. Retail traders and institutional flow do not need synchronous cross-chain composability to trade spot or perps. But market efficiency relies on arbitrageurs and MEV searchers to keep prices aligned across venues – and these actors are the primary beneficiaries of atomic execution.

A recent academic study of cross-chain arbitrage (September 2023 - August 2024) found 242,535 cross-chain arbitrage transactions generating $868.64 million in trading volume and $8.65 million in net profit. The study identified two execution methods:

The implication is clear: without atomicity, cross-chain arbitrageurs must either pre-position inventory on every chain (capital-inefficient) or accept leg risk – the possibility that one side of the trade executes while the other fails or slips during the 242-second bridge delay. Synchronous composability would eliminate leg risk entirely, allowing arbitrageurs to tighten spreads across the entire multi-chain ecosystem with single-transaction atomic execution.

However, the data also reveals a complication for composability advocates: the five largest arbitrage addresses execute over 50% of all cross-chain trades, with one address capturing nearly 40% of daily volume post-Dencun. Cross-chain arbitrage incentivizes vertical integration and concentrates economic power. Synchronous composability might make this more efficient, but it would also entrench already-dominant searchers rather than democratize access.

The honest framing: synchronous composability would improve market efficiency for sophisticated actors. Whether that improvement is worth the infrastructure cost, given that inventory-based arbitrage already handles two-thirds of the volume with 9-second settlement, is the real question.

The “90/10” Framework

The strongest synthesis may be this: ERC-7683 and similar intent-based standards solve 90% of the cross-chain UX problem. The EEZ addresses the remaining 10% – the edge cases where atomicity genuinely matters (liquidations, unified pools, complex DeFi workflows). The question is whether that 10% justifies the engineering complexity and whether it will expand as DeFi matures.

Section 8: Synthesis – What Markets Actually Need

Here is what the data tells us when we assemble it honestly:

For trading – spot and perps – product focus beats composability. The venues that have captured meaningful DEX market share did so by building excellent products on single chains: Hyperliquid’s order book on its purpose-built L1, Solana’s memecoin infrastructure, Aerodrome’s ve(3,3) flywheel on Base. None of them needed or used cross-chain atomic composability. Cross-chain bridge volume (~$18-20B monthly) is a fraction of intra-chain trading ($231B+ monthly). Users go where the products are and stay there.

For broader DeFi infrastructure, the picture is more nuanced. Lending protocols, structured products, and risk management systems face real fragmentation costs that trading venues can more easily absorb. Aave’s presence as an EEZ founding member is not accidental – they operate the kind of cross-chain DeFi infrastructure where synchronous composability has genuine utility.

L2 consolidation is resolving fragmentation through market selection, not protocol engineering. Three rollups capture ~90% of L2 transactions. The long tail is dying. The problem of “20+ L2s with siloed value” is increasingly a problem of zombie chains, not of thriving ecosystems that need better composability.

Agentic trading is not a credible driver of composability demand. Token valuations have collapsed, protocol revenue has cratered, and the claimed AI-driven volume is indistinguishable from traditional bot activity that has existed for years.

The CEX-DEX migration is structural but slow. CEXs still handle ~86-90% of volume. DEX gains are real but measured in percentage points per year, not paradigm shifts per quarter. At the current trajectory, DEXs have years before they face the kind of liquidity fragmentation that might make synchronous composability load-bearing for trading.

The Verdict

Synchronous atomic composability is not a solution looking for a problem. But it may be a solution looking for the right problem. The right problem is not trading – trading has found its solutions through product focus and single-chain execution. The right problem is the 10% of DeFi interactions where atomicity genuinely matters: cross-chain liquidations, unified liquidity provision, and complex programmable finance.

The EEZ is a serious project backed by serious institutions. The question is whether the market it serves – the intersection of multi-chain DeFi and synchronous execution requirements – is large enough to justify the investment. Right now, the data says that market is small. In three years, as DEX share grows and DeFi infrastructure matures, it may not be.

Friederike Ernst said Ethereum has a fragmentation problem, not a scaling problem. The data suggests a third possibility: Ethereum has a consolidation problem. The market is solving fragmentation not by connecting chains but by killing weak ones and concentrating activity on winners. Whether that consolidation leaves room for – or eventually demands – the kind of interoperability the EEZ provides is the question this market will answer in 2027 and beyond.

The strongest version of the thesis, after testing it against the data: For trading, product focus beats composability. For broader DeFi infrastructure, composability matters but asynchronous intents solve most of it. The remaining edge cases are real but small – and whether they grow depends on market structure evolution that has not yet occurred.

Data Sources & Methodology

Primary Data Sources

• CEX Volume Data: CoinGecko, The Block Data Dashboard, CryptoQuant (Feb/Mar 2026 snapshots)
• DEX Spot Volume: DeFiLlama API (api.llama.fi/overview/dexs), DefiLlama chain-level breakdowns
• DEX Perps Volume: DeFiLlama derivatives overview, Hyperliquid Stats, CoinGecko perp DEX rankings
• Bridge Volume: DeFiLlama bridges dashboard, deBridge analytics, LI.FI public metrics, Chainlink CCIP dashboard
• L2 Transaction Data: L2Beat, Dune Analytics L2 transaction dashboards
• Market Share Data: The Block Research, CoinGecko exchange rankings, Kaiko market share reports
• Token Data: CoinGecko token pricing, Dune Analytics token creation dashboards
• Gnosis EEZ: EthCC Cannes announcement, Gnosis official communications (March 29, 2026)
• Vitalik Composability Commentary: Vitalik Buterin public posts (June 2024)
• Multicoin Capital Thesis Evolution: “The Solana Thesis: Internet Capital Markets” (January 2025), compared against earlier theses including “The Hidden Costs of Modular Systems” and 2021 Multicoin Summit keynote
• Cross-Chain Arbitrage Research: “Cross-Chain Arbitrage: The Next Frontier of MEV in Decentralized Finance” (arXiv:2501.17335, ACM SIGMETRICS 2025)
• CEX Exchange Data (Supplementary): CoinLaw OKX Statistics 2025, CoinLaw Gate.io Statistics 2026, Bullish official announcements

Methodology Notes

• CEX vs. DEX market share percentages are calculated from aggregate volume data as reported by CoinGecko and The Block for January 2026.
• “Cumulative” volumes (Aug 2025 - Jan 2026) represent six-month rolling totals as reported by the respective analytics platforms.
• Daily volume figures are multiplied to estimate monthly totals where only daily data is available; these estimates are marked with “est.” in tables.
• Bridge volume comparisons use January 2026 monthly figures for both bridge and DEX volumes to ensure temporal consistency.
• AI agent token pricing and protocol revenue data sourced from CoinGecko and on-chain revenue trackers.
• All data retrieved and verified during March 2026. Markets move; these numbers represent point-in-time snapshots.
• Per editorial policy: no data points have been fabricated. Where data was unavailable for a specific metric, the gap is noted rather than filled with estimates.