On‑Chain Markets, Ranked: Proven Cashflows, Mid-Sized Realities, and Story‑First Speculation
(as of July 24, 2025)TL;DR
- Only a few categories are truly proven at scale today: stablecoins, core DeFi (DEXs, lending, liquid staking), and blockspace (L1/L2 fees & sequencer revenue).
- RWAs (especially tokenized Treasuries), prediction markets, NFTs, MEV/orderflow auctions, and blockchain gaming are real but still mid-sized — they have identifiable users/fees, but cyclicality, regulation, and fragmented liquidity keep them from “escape velocity.”
- Data marketplaces (e.g., DIMO, Hivemapper, Vana, Grass), decentralized AI/GPU markets, DePIN, carbon/ESG tokens, web3 social, identity/ZK, DA/proof markets are story-first: valuations often discount massive TAMs while actual cash revenues are tiny or unproven.
- To avoid “$1B market cap for penny-per-bottle-top payouts,” apply a discipline checklist: Who pays today? Is there a transparent revenue line? Does the token actually capture it? Can the unit economics survive once token emissions stop?
I’ll sort everything that exists (or is being pitched) into three buckets, then go deep on each:
- Proven & Scaled — real users, recurring revenues, regulatory clarity improving.
- Real but Mid-Sized — genuine PMF pockets, but not yet massive or structurally fragile.
- Speculative / Story‑First — exciting narratives, minimal real revenue and/or unclear token value capture.
For every category, I’ll note approximate market size/volumes (as of July 24, 2025), how well-tested it is, the on-the-ground reality, and key risks.
Numbers move quickly — treat them as ranges, not gospel.
1) Proven & ScaledStablecoins (payments, settlement, collateral)- Order of magnitude today: ~$250B+ in circulating supply.
- Why it’s proven: Stablecoins are crypto’s de facto unit of account, powering CEX/DEX liquidity, remittances, and on-chain money markets.
- Reality: Growing regulatory clarity (MiCA in the EU, stablecoin bills in the US) plus real world usage = likely 10x TAM over the decade.
- Risks: Regulatory treatment (bank-like capital rules?), concentration in a few issuers, blacklist/custodial risk, and interest margin competition.
- Order of magnitude: > $100B TVL; trillions in annualized DEX & perps volume.
- Why it’s proven: Clear PMF for crypto-native leverage, trading, and collateralized USD liquidity.
- Reality: TVL isn’t revenue. Fee take is much smaller and often passed to LPs/validators. Still, recurring fees exist and are transparent.
- Risks: Smart contract & oracle failures, regulatory pressure on perps/securities, MEV extraction eroding UX.
- Order of magnitude: Ethereum L2s alone are doing ~$100M annualized in sequencer profits (fluctuates with activity).
- Why it’s proven: Blockspace is the product; the market pays to settle and prove state. As more users roll up, more fees get routed through sequencers.
- Risks: Fee compression via competition/modularity (cheaper DA layers), decentralizing sequencers (less rent to capture), and vertical integration.
- Order of magnitude: Low-to-mid tens of billions across all RWAs; tokenized T‑bills/MMFs ≈ high single billions.
- Why it’s real: A simple legal wrapper to port yield into crypto. Institutions are comfortable with the asset; infra (transfer agents, KYC) is maturing.
- Reality: Much of it lives in permissioned walled gardens. The big unlock is when this collateral becomes natively composable in DeFi without legal hairballs.
- Risks: Regulatory fragmentation, off-chain trust dependencies, and unclear value capture for protocol tokens vs. licensed issuers.
- Order of magnitude: ~$100M+/yr on Ethereum-scale chains (directionally).
- Why it’s real: There’s real, repeatable money in ordering transactions. Many networks/protocols now formalize auctions around this.
- Reality: Most value accrues to validators/builders, not tokens. “MEV tokens” rarely have direct fee claims.
- Risks: Enshrined PBS, pre-confirmations, and protocol-level mitigations could compress margins.
- Order of magnitude: Single-digit billions in market cap; low single-digit billions in yearly sales.
- Why it’s real: Persistent, if niche, user base. Infra & IP deals are getting more professional.
- Reality: A fraction of 2021 mania. Royalties are under pressure, liquidity is thin, and volumes are highly cyclical.
- Risks: Regulatory scrutiny (when marketed as investments), illiquidity, and recurring revenue models (royalties, fees) aren’t guaranteed.
- Order of magnitude: Hundreds of millions to low billions in monthly volume during peak events (e.g., elections), much less otherwise.
- Why it’s real: Clear product-market fit when the world cares. Polymarket has shown people will bet on everything.
- Reality: Extremely event-driven. Liquidity evaporates between macro catalysts.
- Risks: Gambling/securities law, oracle manipulation, and thin long-tail markets.
- Order of magnitude: Millions of daily wallets interacting; very low direct on-chain spend/user.
- Why it’s real: User engagement is real; wallets + mini-apps have lowered friction.
- Reality: Most economies are subsidized by tokens. Few games sustain themselves through fun-first retention plus on-chain sinks.
- Risks: App store policies, the “fun > token” challenge, and shallow sinks that turn economies into mercenary airdrop farms.
These may become huge — but right now, the cashflow is negligible and/or token value capture is hand-wavy.
Data Marketplaces (consumer, mobility, browsing, training data)Examples: Grass, DIMO, Vana, Hivemapper
- Today’s reality: Micropenny payouts, limited verified buyer demand, and lots of token emissions. Some tokens trade at hundreds of millions of market cap on essentially untested unit economics.
- Why it could be big: AI/enterprises need verifiable, consented, and provenance-rich data. On-chain rails can automate rev-sharing and rights.
- What must go right: Real enterprise buyers with budgets, robust privacy and governance, and positive unit economics after incentives.
Examples: Render, Akash, io.net, Gensyn
- Reality: Capacity and token prices grew, but revenues are small compared to FDVs.
- Promise: AI demand is astronomically large; decentralized infra can arbitrage idle capacity and resist censorship.
- Risks: SLAs/QoS, centralized buyer inertia, data residency/compliance, and potential race-to-zero margins.
Examples: Helium (post-migration), Hivemapper, WeatherXM
- Reality: Many show low real revenue vs. token subsidies.
- Promise: If token incentives bootstrap moat-y, cash generative networks, upside is massive.
- Risks: Chicken-and-egg demand, heavy capex, regulatory/licensing headwinds, data quality verification.
- Reality: Weak volumes, credibility issues (double counting, verification scandals).
- Promise: On-chain MRV, instant settlement, and transparent registries could modernize a broken market.
- Risks: Standards fragmentation, regulator skepticism, and long sales cycles with enterprises.
- Reality: Sub-1M MAUs territory with experimental monetization.
- Promise: Native ownership, composability, and new creator monetization rails.
- Risks: Competing with Web2 incumbents, spam/Sybil resistance, and unclear token value capture.
- Reality: Essential primitive, tiny direct revenue.
- Promise: Required for RWAs, compliance, and payments.
- Risks: UX, regulation, and tokens that don’t actually collect fees.
- Reality: Strong narrative; valuations assume massive rollup proliferation.
- Promise: If thousands of app-rollups exist, DA and sequencing are real utility businesses.
- Risks: Fee compression, shared sequencers, L1/L2 vertical integration, and moats that collapse to cost-plus pricing.
Many “new” markets distribute tokens to supply (data, compute, hardware, engagement) before there’s material, recurring demand from paying customers. This bootstraps a network, but often:
- The buyer side is speculative or subsidized (no recurring enterprise budgets yet).
- The revenue line is opaque or trivial relative to the FDV.
- Token holders don’t have a direct claim on future cashflows (governance ≠ cash).
- When token emissions slow, supply collapses — exposing a fragile network without sustainable unit economics.
This doesn’t mean these ideas won’t work. It does mean you should price them like options on TAM + execution, not like cashflowing businesses.
A Discipline Checklist for Evaluating Any On‑Chain MarketUse this every time you see a shiny new “X on-chain” pitch:
- Who pays, today? Name the customers. Are they recurring?
- Transparent fees/revenues? On-chain or auditable?
- Token ↔ Revenue linkage? Is there a fee switch, burn, rev-share, or is it just governance?
- Unit economics post-incentives: If token emissions stop, does the network sustain itself?
- Regulatory clarity: Is it a security, gambling product, money transmitter, or commodity?
- Substitution risk: Could Web2 infra + Stripe do it cheaper/faster?
- Composability moat: Is being on-chain actually an advantage (collateral, programmatic flows)?
- Liquidity depth & churn: Can users exit without nuking prices? Are there real sinks?
- Supply-side stickiness: Will suppliers stay if token price drops 80%?
- Event cyclicality: Does it only work during elections/bull markets?
- Data/quality oracles: Who ensures correctness (data, measurements, model outputs)?
- Credible decentralization roadmap: Or will rent-seeking central coordinators capture the surplus?
If you’re actively investing/building, track:
Proven & Scaled
- Stablecoins: Circulating supply, treasury backing, interest income captured by issuer vs. token.
- Core DeFi: Fee revenue (not TVL), DEX spot/perps volume, liquid staking market share, liquidation performance in stress.
- Blockspace/L2s: Fees, sequencer revenue, DA costs, decentralization of sequencers, effective cost/tx.
Real but Mid-Sized
- RWAs: On-chain AUM by asset class, breadth of integrations with DeFi, settlement latency, redemption flows.
- MEV/Orderflow: Builder/validator revenues, share captured by protocols/tokens, impact of PBS/MEV-burn.
- NFTs: Royalty realization, secondary market liquidity depth, non-speculative utility adoption (gaming, IP).
- Prediction Markets: Open interest, spread depth on major events, jurisdictional reach.
- Gaming: dUAW vs. on-chain spend/user, retention after incentives, % revenue from gameplay vs. speculation.
Speculative / Story‑First
- Data Markets: $ paid by real buyers per active contributor, churn, verified enterprise contracts.
- Decentralized GPU/AI: $ revenue vs. token emissions, QoS/SLA metrics, enterprise repeat usage.
- DePIN: Revenue/user vs. token rewards, coverage quality, paying demand growth without incentives.
- Web3 Social: MAUs, ARPU, spam/Sybil mitigation metrics, dev ecosystem building on top.
- Identity/ZK: Paid verifications, integrations into RWAs/compliance rails.
- DA / Provers: Cost/byte or cost/proof trends, competition from integrated L1/L2 stacks.
- Stablecoins become a regulated global settlement layer for fintechs, exchanges, and RWA platforms. Winners capture interest spreads + fees.
- RWAs graduate from walled gardens into composable collateral, bringing institutional liquidity on-chain.
- L2 economics compress as DA options expand, sequencers decentralize, and shared sequencers emerge — volume up, margin down.
- Prediction markets surge episodically, but sustained growth requires friendlier jurisdictions and institutional market makers.
- Gaming that hides the chain (wallet abstraction, gas sponsorship) might finally crack mainstream, but on-chain value capture will be thin unless there’s a real sink.
- Data/AI/DePIN will sort into a few genuine utilities (with net-positive cashflows) and many token-subsidized experiments that fade.
- ZK infra & DA layers become boring utilities — huge in volume, priced at cost-plus with value accruing to whoever aggregates demand or bundles services.
Most of the real, repeatable cashflows in crypto today come from:
- Dollarizing the internet (stablecoins)
- Trading/leverage & collateralized USD liquidity (core DeFi)
- Selling blockspace (L1s/L2s)
Everything else is either catching up (RWAs, prediction markets, NFTs/gaming) or selling a story whose unit economics haven’t been proven without token subsidies.
On‑Chain Markets, Ranked: Proven Cashflows, Mid-Sized Realities, and Story‑First Speculation was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.