The market leader in crypto cards offers 0% cashback.

That one fact should be enough to suggest the some framework for analyzing this market is wrong. If this were a cashback race, the leader wouldn’t look like this.

The real competition is happening somewhere else and not every player is fighting the same battle.

What history taught us? This Fight Has Been Fought Before

In 1958, Bank of America launched the BankAmericard in California, later renamed Visa. For the next two decades, the competition wasn’t about who gave the most back. It was about whose card worked in more places. By 1966, BankAmericard was accepted at 61,000 merchants across 42 states. That same year, a group of New York banks formed the Interbank Card Association, which became Mastercard. Through the 1960s and 70s, dozens of regional card networks thrived in their own territories, then disappeared one by one—not because their products were bad, but because in a winner-take-most market, limited coverage isn’t a defensible position.

Cashback didn’t arrive until 1986, when Sears launched the Discover Card. By then, credit cards had been around for nearly thirty years and the infrastructure battle was long over. Discover invented cashback and grew because of it, but couldn’t hold its position on cashback alone. It was acquired three times and ended 2025 as a Capital One subsidiary.

Crypto cards are running through both phases at once, compressed from decades into years. The infrastructure fight isn’t over. The cashback fight has already started. Most players are competing in the second phase while thinking they’re in the first.

Three Types of Cards, Three Different Fights

Every crypto card on the market carries the same label, but they’re solving fundamentally different problems.

The market share estimates below are based on on-chain traceable transaction volumes from the leading players@RedotPay,@KASTxyz,@ether_fi,@useTria, @joinkarta are meant as a structural approximation, not precise figures.

Type 1: Dollar Access (~75–80% of traceable volume)

These users aren’t crypto enthusiasts. They live in places with currency instability or capital controls, and what they need isn’t cashback—it’s a way to hold and spend dollars. Crypto cards are currently the fastest path to a functional dollar account. Redotpay dominates this category. Karta is cutting into the same user base through Telegram distribution. KAST sits across categories: its ACH and Fedwire rails are essentially extending the US banking system to people without American accounts, which is the same underlying need as a Bangladesh user holding USDT—just packaged as a freelancer tool.

Type 2: Crypto Liquidity (~15–20% of traceable volume)

These users are already inside crypto and need a way to spend what they have. ETH restakers don’t want to liquidate their positions but need spending power. Web3 freelancers get paid in USDC and need to live on it. @ether_fi, @Nexo, and @AviciMoney are all solving variations of this problem for slightly different user profiles.

Type 3: Commodity Competition (low single digits)

These players are competing on the same dimensions—zero-fee marketing, cashback rates, multi-chain support—none of which have any structural barrier to replication. Cypher, Holyheld, MetaMask Card, and Gnosis Pay all live here. The competitive dynamic is fees compressing toward zero, margins thinning, and no compelling reason for users to stay. This is the post-1986 credit card market: feature convergence, no structural differentiation, and the players without a distinct position eventually get acquired or disappear.

The key implication of this framework: competition within a type is real; competition across types barely exists. Redotpay isn’t threatened by EtherFi. EtherFi isn’t threatened by Cypher. The actual threats come from stronger players targeting the same user, or traditional institutions entering from outside.

Why the Leaders Are Winning, and Where They’re Exposed

Using $10M in monthly volume as a threshold for April 2026, only five players clear the bar: Redotpay ($413.8M), KAST ($74.7M), EtherFi ($65.7M), Karta ($27.7M), and Tria ($15.6M). For each one, the analysis runs on two tracks: the strength of the moat, and the cost for users to leave—because those aren’t always the same thing.

Redotpay

$413.8M in monthly volume and 75% market share came not from superior product design but from being first to build agent networks and local-language communities in emerging markets. Bangladesh, India, Egypt, Nigeria—users in these markets didn’t compare cashback programs. They found something that worked and stayed. Switching costs are real: leaving means going through KYC again, rebuilding local top-up channels, and accepting uncertainty in markets where alternatives are thin.

The logic mirrors BankAmericard’s early expansion—winning through presence, not features. A California user with a BankAmericard in the 1960s wouldn’t switch to a better-designed card if the new one wasn’t accepted anywhere. Redotpay’s position in emerging markets is roughly that right now.

The real risk: A significant portion of Redotpay’s core markets run on regulatory tolerance, not regulatory authorization. In Bangladesh, its use hasn’t received official approval from the central bank. The more immediate threat is Western Union, which announced a USD Stable Card in 2026 targeting the same markets. What Western Union brings isn’t a better product—it’s 175 years of global agent infrastructure and licensing depth. Regulatory authorization takes years. Brand trust takes a decade. These aren’t things Redotpay can buy its way to quickly.

KAST

Monthly volume went from $16M in November 2025 to $75M by January 2026. The core reason is simple: KAST is one of the few platforms offering real ACH and Fedwire access to non-US users. A designer in Taiwan can receive a US client payment in USDC into a KAST virtual account, then spend or withdraw from it—something that normally requires a US bank account.

The reason competitors haven’t caught up yet isn’t that the technology is hard to copy. It’s that the banking relationships behind those rails take time to build: compliance reviews, legal frameworks, the risk appetite of the partner bank. That process typically takes six to twelve months. KAST has a head start, but the window isn’t permanent.

The real risk: Switching costs are low. KAST’s rails are rented, not owned. Users stay because it works, not because leaving is painful. The bimodal Trustpilot rating—3.3 overall, 53% five-star versus 43% one-star—suggests the product works well when it works and fails badly when it doesn’t. Once a more stable competitor offers the same rails, churn will be as fast as growth was. This is the Discover dynamic: the innovation is copyable, users aren’t deeply locked in, and the eventual outcome is absorption by something with wider distribution.

Etherfi

The 3% cashback is funded by restaking yield on $5.8B in protocol TVL—not by cross-subsidy from trading fees or marketing budget. Users stake ETH into the protocol and draw against it as a credit line. The ETH keeps appreciating, the yield keeps running, and they can spend without triggering a taxable sale. No other crypto card runs on this structure.

Switching costs are the highest of any player in this group. Leaving EtherFi means unstaking, waiting out the unlock period, and moving assets elsewhere. That friction is a genuine moat—closer to American Express’s position in the corporate travel market than to any of the commodity players. Amex didn’t win by offering the best general-purpose card; it served a specific, high-value segment well enough that they had no strong reason to leave.

The real risk: Restaking yield isn’t fixed—it tracks ETH staking market conditions, and if it compresses, the cashback either shrinks or becomes loss-making. The more immediate constraint is geography: EtherFi Card is only partially available in the US, with New York, California, Texas, and Florida all excluded while the SEC and CFTC continue rulemaking on DeFi credit products. EtherFi hasn’t disclosed a timeline for full US availability. Those four states are exactly where the “hold ETH, borrow against it, avoid capital gains” use case is strongest. Without them, the card is missing its most natural market.

Karta and Tria | Directionally Interesting, Unverified

Karta ($27.7M, up 325% in six months) and Tria ($15.6M) together represent less than 10% of Redotpay’s single month. Treating them as peers to the top three would misrepresent the market. But they point to two genuinely interesting bets.

Karta is a wager that Telegram becomes the primary financial interface for Web3-native users. Zero top-up fees and native Telegram distribution eliminate most of the friction that slows other card programs down. The structural problem is that binding a financial account to a social messaging platform creates security vulnerabilities—SIM-swapping and phishing risk—that can’t be fixed with product iteration.

Tria’s chain abstraction direction is right: let users spend from any chain without manual bridging or holding gas tokens. The execution problem is specific and self-inflicted—charging users for a card tier before their KYC is approved, then struggling with refunds when applications are rejected. This is the Diners Club risk: correct thesis, insufficient execution, and the possibility of becoming a footnote that later builders get credit for pioneering.

What to Watch

The competitive landscape isn’t static. Three variables will determine who’s gaining ground and who’s losing it over the next twelve to twenty-four months. They have different levels of certainty and different time horizons, which affects how much weight to put on each one.

Highest certainty, nearest term: Regulatory tightening

Europe has a hard deadline: MiCA requires all CASPs to hold full authorization by July 2026 or stop operating. Most commodity-tier players don’t have it. The second half of 2026 is the first real market clearing event. The parallel to the 1970s regional card networks is direct—they didn’t lose because their products were worse, they lost because they couldn’t qualify to compete at the next level. In emerging markets, the timing is less predictable, but the direction is the same. Redotpay’s core operations in Bangladesh and India are the most important regulatory risk to track on an ongoing basis.

Medium term, largest impact: How fast traditional institutions move

Western Union’s Stable Card, Wells Fargo’s WFUSD trademark filing, Citi exploring stablecoin infrastructure—these signal that the competitive structure of the dollar access market is changing. This isn’t the first time a large incumbent has used distribution reach to reshape a card market. In 1977, Citibank mailed millions of unsolicited credit card applications to consumers across the US. It didn’t win through product superiority. It won because its scale and regulatory capability were simply out of reach for competitors to replicate. Western Union’s entry into stablecoin cards follows the same logic. The important asymmetry: traditional institution entry is an existential threat for dollar access players, and barely relevant for crypto liquidity players. ETH restakers aren’t going to use Western Union. That divergence is probably the single most important variable for assessing where Redotpay and EtherFi end up five years from now.

Longest term, sets the ceiling: Infrastructure consolidation

Rain ($1.95B valuation), Reap ($600M acquired by Kraken), BVNK ($1.8B acquired by Mastercard). The capital markets are pricing infrastructure, not consumer cards. The logic is the same as BankAmericard spinning out of Bank of America to become Visa: the valuable thing wasn’t any individual card, it was the network every card had to run on. Consumer card margins are bounded by whoever sits beneath them in the stack. This plays out over three to five years, but the acquisition prices have already made the answer clear.

Credit cards took thirty years to move through three competitive phases—coverage, cashback, ecosystem lock-in. Crypto cards are running all three simultaneously, compressed into a few years.

The framework for assessing who wins has never been about who offers the highest cashback rate. How hard is it for users to leave—that’s the individual question. Whether what’s been built is owned or rented—that’s the structural question. When consolidation arrives, whether a player gets acquired or replaced—that’s the market question. The answers to all three, taken together, are what actually determine where each player stands in this fight.

Bulls quote that number as proof of a boom.

Bears point out that on-chain analysts flag most of the volume as wash trading and infrastructure testing, with organic activity still tiny.

Here’s the thing: for anyone deciding where to build or invest, that debate matters less than it seems. Whether today’s volume is 20% real or 80% real, it’s a rounding error against the $33T that stablecoins moved in 2025 and either way, the same question decides everything: which path does agentic payments actually take from here?

Because it isn’t one bet. It’s three distinct paths - Replace, Integrate, Create with different winners, different risks, and a different role for stablecoins in each. Most of the noise on this topic comes from people arguing about different paths without realizing it.

A few facts frame all three. The rail demonstrably works: 76% of x402 activity sits below the ~$0.30 fixed cost of a card transaction — commerce that card rails cannot price at all — and USDC settles 98.6% of it. The human-facing side of agent commerce is real but card-native: ChatGPT Instant Checkout and Google’s Universal Commerce Protocol run on traditional rails. And the whole category is gated on autonomy that’s still being built.

So: the rail works, demand is embryonic, the narrative is ahead of the data.

Now the map.

Path 1: Replace

The hypothesis: agents are cost-optimizers. They’ll notice the 2–3% merchant fee on card rails and systematically route around it.

I don’t buy it. Two reasons, one small and one fatal.

The small one: the math is less dramatic than the headline. The famous 2–3% mostly isn’t the networks’ money — interchange flows to issuing banks; the network’s own take is closer to 12 basis points. Even if agents rerouted 1% of Visa’s ~$17T, Visa itself loses roughly $200M against ~$40B in revenue. The real exposure sits in the issuer interchange pool and the rewards programs it funds — painful for banks, survivable for rails.

The fatal one: someone has to be liable. When an agent buys the wrong flight with your money, you don’t want the cheapest rail — you want someone to call. A stablecoin transfer is final: no chargeback, no issuer, no institution obligated to make you whole. And that’s not a missing feature crypto will eventually ship. Recourse requires a balance sheet that voluntarily absorbs losses — and the moment you bolt one onto a crypto rail, you’ve rebuilt the card network with extra steps. The incumbents understand this perfectly: Amex now backs both consumer and merchant when a registered agent errs; Visa shipped agent identity verification at checkout. No regulator has assigned liability for autonomous purchases, and the incumbents are converting that unsolved question into their moat.

So my read: Replace doesn’t happen anywhere a human bears the loss. What survives of it is pure machine-to-machine flow — sub-cent, nothing to dispute, no one to refund — and that isn’t a frontal assault on cards at all. That’s Path 3’s territory.

Checkpoints — that is, what would prove me wrong: a major agent platform routing meaningful consumer volume off-card; a credible liability framework that makes irreversible payments safe for humans; issuer rewards economics visibly compressing.

Path 2: Integrate

The hypothesis: stablecoins don’t fight the card networks — they become the settlement plumbing underneath them. Spend stays on cards; settlement moves on-chain.

The evidence is a progress report, not a forecast. Visa expanded its stablecoin settlement pilot to nine blockchains, explicitly positioning itself as a common settlement layer across them. Mastercard acquired BVNK and Stripe bought Bridge. The networks are jointly backing a new stablecoin platform. Stablecoin-funded card programs are already live in 18 countries — users spend stablecoins at 175M+ acceptance points while merchants settle in local fiat.

But “most certain” is not “risk-free,” and two risks deserve more airtime than they get.

Regulatory: settlement concentrating on stablecoins means concentrating on stablecoin issuers — that 98.6% USDC share is a systemic dependency on one company’s reserves and regulatory standing. And jurisdictions are diverging: if any major market decides to treat stablecoin settlement as shadow banking or an AML bypass, Integrate doesn’t die — it fragments by geography, which kills much of its point.

Technical: card settlement today is batch and forgiving. Machine commerce wants real-time. Chain finality, fee volatility, and cross-chain bridging are tolerable for T+0 batch settlement, but unproven for high-frequency clearing at card-network scale. The nine-chain abstraction play is itself a tell: the networks don’t believe any single chain is ready to carry this alone.

Checkpoints: the share of card settlement volume actually clearing in stablecoins; whether the networks’ agent protocols default to stablecoin settlement; bridge-layer M&A continuing at premium prices; a second issuer reaching scale (de-concentration from Circle).

Path 3: Create

The hypothesis: the biggest prize isn’t redirecting existing payments — it’s transactions that never existed. A research agent paying $0.005 to a niche data API it discovered thirty seconds ago. GPU time metered by the second. Agent-to-agent task outsourcing. No procurement process and no card rail could ever price these; they’re not displaced volume, they’re new TAM. And there’s a path-dependency kicker: agents won’t shop like tourists, they’ll trade like businesses — and relationships born on stablecoins stay on stablecoins, because flows born on-chain have no legacy rail to return to.

The verification problem is a cold start. Create faces a brutal chicken-and-egg: who opens the first pay-per-call endpoint, and who accepts half a cent from an anonymous agent with no relationship and no recourse? Supply won’t come without agent demand; agents can’t demand what nobody sells. (That manufactured x402 volume from earlier? Read charitably, it’s this problem made visible — when organic demand doesn’t exist yet, participants simulate both sides at once.)

And the demand gate is autonomy itself: only about a quarter of enterprises run truly autonomous agents today, maybe half by 2027. Until the human leaves the loop, the loop pays with the human’s card.

Checkpoints: autonomous (not supervised) deployment numbers; non-crypto companies opening pay-per-call endpoints; the organic share of agent-payment volume rising; a real billing-and-credit layer emerging for machine commerce.

The whole framework in one table

These paths aren’t mutually exclusive — the likeliest future is all three running in parallel at different speeds: Integrate compounding first (it’s underway), Create building slowly underneath, Replace skirmishing at the issuer-pool margins for years.

One test cuts through most stablecoin debates along the way: does this flow have a trusted intermediary everyone can accept? Where yes, stablecoins compete on cost and speed — and may win or lose on the merits. Where no — cross-ecosystem, sub-threshold, machine-speed, no human to sign up for anything — they’re frequently the only rail that exists on day one. Both situations are real. They’re just different markets.

Where to look

Directions, not picks:

The control layer. Every path — card-settled or chain-settled — needs the same thing: scoped spending authority for agents, with limits, permissions, and audit trails. The killer-product question is behavioral, not technical: how much money will a human actually delegate to an agent, and under what guardrails? Whoever answers that defines the product category. Rail-agnostic, which makes it the safest exposure to the entire theme.

The billing layer. Metering, reconciliation, statements, credit, arbitration for machine commerce. The rail works; the financial operations around it don’t exist yet. This is the most acknowledged gap in the entire stack — including by the people building the rails.

The bridge layer. Stablecoin↔fiat orchestration is crowded and getting acquired fast — but two ten-figure exits already proved the buyer demand. Legible exits, compressed timelines.

The checkpoint discipline. The cheapest position of all: don’t pick a winner yet. Every checkpoint above is public and observable. Define in advance what you’re watching for, and the market will tell you which path is compounding.

If you build, build what makes the real volume grow: the billing, the authorization, the trust plumbing that finally lets a human leave the loop.

The interesting question was never whether agents will pay with crypto. It’s which transactions exist only because they can — and right now, someone still has to build the reasons.

Why This Is Different

Crypto didn’t invent the perpetual swap. What it did was turn the prime broker into a protocol.

For fifty years, prime brokerage looked like this: 24/7 leveraged trading across asset classes, cross-margin between equities, rates, FX, and commodities, securities lending, total return swaps. It’s the layer that makes serious institutional trading possible — and it’s gated. To access it, you need $100M+ AUM, a relationship at Goldman, JPMorgan, or Morgan Stanley, and a recognized jurisdiction. Worldwide, maybe a few tens of thousands of entities qualify. Bill Hwang ran Archegos with roughly $10B of equity backing $100B+ of gross stock exposure on this stack — and when it blew up in 2021, the prime brokers ate more than $10B in losses.

That entire leverage primitive was unavailable to 99% of the planet.

In 2016, BitMEX productized the perpetual swap. The structure wasn’t new — total return swaps existed in the 1980s, CFDs in the 1990s. What crypto did that nobody had done before was make it permissionless, anchored to an index price rather than a dealer’s quote, and cleared through a shared margin pool instead of bilateral ISDA agreements. Funding rates replaced negotiated financing legs. Smart contracts replaced counterparty paperwork.

The result: the prime brokerage stack — at least the part that mattered for leveraged speculation — went from “you need to be a hedge fund” to “you need a wallet with USDC.”

For eight years, this primitive has only served BTC, ETH, SOL, and the long tail of crypto assets. That alone produces $200B+ in daily volume. Now it’s about to cross into the actual large markets: equities, commodities, FX, rates. That’s the bet.

Why Now — Not 2022, Not Later

Three pieces of infrastructure reached production-grade after 2025. None of them is enough on its own. All three together is what turns RWA perps from a PoC demo into a venue that can actually scale.

Oracles finally got good enough

Crypto-native pricing is relatively well-solved for major pairs. RWAs are harder. Where does TSLA price come from after the bell? How do you mark gold between LBMA fixes? Who’s the source of truth on EUR/USD over the weekend? When NVDA does a 10-for-1 split overnight, how does the oracle handle the corporate action without nuking the liquidation engine?

Three years ago, oracle delays of 10–30 seconds were normal for anything outside the top crypto pairs. You can’t run a liquidation engine on TSLA with a 10-second delay, let alone 30.

Now it’s different:

• Pyth Network pulls first-party data directly from CBOE, Coinbase, Binance, Virtu Financial, and 125+ exchanges and market makers — sub-second updates, no aggregator middleman

• Chainlink shipped dedicated RWA infrastructure through its SmartData suite, including NAV and Proof of Reserve feeds

• RedStone built modular pull-based oracles for the long tail

None of this existed in usable form three years ago.

Stablecoin collateral hit institutional scale

To run institutional-grade cross-margin perps, you need a stablecoin pool deep enough to absorb billions in positions and arbitrage. In 2020, USDT and USDC combined were under $20B. By Q1 2026, the total stablecoin market sits at ~$315B, with the two of them holding ~84% market share. Stablecoin issuers are now collectively one of the largest holders of U.S. Treasuries.

A few billion in stablecoins doesn’t get you there. A few hundred billion does.

Tokenized RWAs finally have ground truth

Pure synthetic perps technically don’t need tokenized RWAs. But the ecosystem around them does. Arbitrageurs need tokenized underlyings to run cash-and-carry trades that pull perp prices back to fair value. DeFi protocols need them as collateral. Regulatory stories need a physical settlement option somewhere in the stack.

The numbers:

• BlackRock’s BUIDL crossed $3B AUM in early 2026 across nine blockchains — and is now trading on Uniswap via UniswapX

• Ondo Finance hit $1.6B in tokenized treasury TVL in September 2025; their Global Markets platform has processed billions in cumulative tokenized U.S. stock and ETF volume, and they’re launching tokenized U.S. equities on Solana

• Dinari became one of the first SEC-registered broker-dealers cleared to issue tokenized U.S. equities

Two years ago, none of these existed in production form. Today they collectively support a $10B+ on-chain RWA layer. That’s the ground truth a perp venue needs to anchor to. Without it, the whole derivatives layer is floating.

Why waiting is losing

Derivatives venues are extreme winner-take-most markets. Once liquidity concentrates, latecomers can’t pull market makers or onboard users without a structural edge — a new asset class, a new jurisdiction, a categorically better UX.

Hyperliquid took roughly 24 months to lock in dominant position in crypto perps: $185B+ in 30-day volume, ~44% of the perp DEX market, and the only major perp DEX still gaining share in 2026. RWA perps will follow the same compression on the same timeline. By the time the thesis is obvious, you can’t build into it.

The decision happens now, not later.

What This Actually Does for the Market

1. Filling the price discovery gaps

The lazy framing is that “traditional markets are only open 30% of the time.” Markets aren’t a single thing — you have to break it down by asset class.

Single-name equities are the worst. Liquid price discovery on TSLA, NVDA, MSTR happens for 32.5 hours a week — the New York session from 9:30am to 4pm Eastern. That’s 19% of total time. Pre-market, after-hours, and the new overnight sessions exist, but liquidity is too thin to anchor a real reference price — spreads widen by an order of magnitude and large orders move prices on their own. The remaining 81% has no trustworthy single-name price anywhere on the planet. This is the single biggest gap in modern markets, and it’s where RWA perps can plausibly own pricing.

Commodities and FX are better covered on weekdays — CME Globex runs ~23 hours, FX interbank runs 24/5. The real dead zone is the 48-hour window from Friday 5pm ET to Sunday 5pm ET, when every futures market and every FX desk is shut. Geopolitical news tends to break exactly during this window. TradFi has no venue to react until Sunday evening.

The point isn’t that RWA perps replace the NYSE closing auction. They won’t. The point is they own structural gaps the NYSE can’t be open during — Asian-hours single-name trading, the Friday-to-Sunday weekend window, earnings vacuum periods.

CME BTC futures didn’t become the reference price for Bitcoin — spot still holds that role. But academic research shows them leading spot during U.S. trading hours, and they became the venue institutions actually use for BTC exposure. RWA perps walk a similar path, but across equities, commodities, FX, and rates simultaneously.

This is a complement to traditional markets, not a substitute. A complement can coexist with NYSE and CME. A substitute gets crushed by them.

2. Prime brokerage for anyone with a wallet

Historically, a real prime brokerage relationship requires $100M+ AUM, a Goldman or JPM contact, and a recognized jurisdiction. Worldwide, maybe a few tens of thousands of entities qualify.

RWA perps make the same primitive — cross-margin, leveraged, 24/7, across BTC and equities and commodities and FX — available to any wallet holding a stablecoin. The user base goes from tens of thousands of credentialed institutions to tens of millions of wallet holders globally.

Four concrete cases:

A retail user in Lagos with $500 in USDC who wants to lever long NVDA can access tokenized stocks through Robinhood EU or xStocksfi on Solana — but those are spot. Not leveraged, not cross-margined. To get real leverage on a single name, they’d need an offshore brokerage account, a SWIFT relationship, and full KYC. RWA perps collapse that into one wallet.

A crypto-native treasury that wants to short U.S. equities to hedge market beta has options — CME E-minis, SPY puts, inverse ETFs — but each requires off-ramping USDC into USD, opening a broker account, and managing margin in a separate pool. RWA perps eliminate the off-ramp and unify the margin.

An Asian family office reacting to Middle East geopolitics on a Saturday afternoon has nowhere to go in TradFi until Sunday 5pm ET. That 48-hour window is real, and it’s exactly when geopolitical news tends to break. RWA perps own it.

A small quant fund running a BTC perp / TSLA perp pair trade with $5M in capital can’t open a Goldman PB account and can’t get cross-asset net margin from Hidden Road or FalconX — their minimums start in the eight figures. RWA perps are native to this — same wallet, same margin pool, both legs, no minimum.

The pattern across all four: traditional alternatives exist, but they’re fragmented across separate accounts, separate jurisdictions, separate margin pools, and separate KYC regimes. RWA perps collapse all of it into one execution layer accessible from one wallet. Not zero-to-one for any single use case — zero-to-one for the combination.

What’s Already Happening

This isn’t pure prediction. Protocols are shipping, and the data is starting to come in.

Hyperliquid is the king of crypto perps and the largest single variable in this thesis. $185B+ in 30-day volume, ~44% of perp DEX market share, FDV around $40B. If they decide to turn on RWAs tomorrow with their existing liquidity, market maker relationships, and UX, the race is essentially over before it starts. As of now they’re still focused on crypto perps — though HIP-3, their permissionless perp framework, is already powering oil and silver markets at the margin. Watching them is mandatory.

Ostium Labs, on Arbitrum, is the cleanest pure-play in the market right now. They started with commodities (gold, oil, sugar) and FX majors (EUR/USD, GBP/USD, USD/JPY) and have been adding equity indices and 22 single-name U.S. equities. Numbers as of April 2026: ~$6B in monthly volume, $213M in open interest with ~97% sitting in non-crypto pairs (gold alone carries $71M+ in OI), cumulative trade volume past $50B, TVL around $56M — the market making vault has grown roughly 30x over the past year. They raised over $27M total, including a $20M Series A in late 2025 led by General Catalyst and Jump Crypto, with Coinbase, Wintermute, GSR, Susquehanna, and angels from Brevan Howard, Bridgewater, and Two Sigma. Founded by Harvard alumni out of Bridgewater, BlackRock, Coinbase. As pure-play RWA perp as anyone in the market.

Synthetix is the legacy synthetic asset protocol mid-pivot. They sunset Arbitrum in early 2025, then pivoted again in late 2025 to consolidate everything on Ethereum Mainnet with a new CLOB architecture — positioning 2026 as a comeback year built around multi-collateral margin and basis trade vaults. Volume dropped sharply during the transition, but a six-week trading competition season did $11B in volume and $4.5M in fees, proving demand is real. The question now is whether Mainnet CLOB can convert competition liquidity into sustained organic flow. Technical depth is real; distribution is a question mark.

The tokenization issuers — Backed, Dinari, Ondo — sit adjacent rather than head-on. Dinari has the first SEC-registered broker-dealer license for tokenized stocks; TVL is around $60M and growing. Ondo Global Markets has processed billions in cumulative tokenized U.S. stock and ETF volume and is launching tokenized U.S. equities on Solana. They all know that pure spot tokenization doesn’t capture the real fee pool, and they’re all either in conversations with perp venues or evaluating vertical integration internally.

And then there are stealth teams. I’ve heard of multiple groups pairing TradFi derivatives veterans with crypto infrastructure builders, working toward this category. Hybrid teams like these are usually where category winners come from. None are ready to talk publicly yet — which itself tells you the market hasn’t fully formed.

The race is wide open. That’s what makes the next 12–18 months matter.

Where This Thesis Breaks

If you’ve read this far and the path looks too clean — you’re right. Three angles I think most plausibly kill it:

Regulators step in hard. The SEC has not really softened on tokenized equities. Europe’s MiFID, Hong Kong’s SFC, Singapore’s MAS could all tighten rather than loosen permissionless derivatives access. The reason Dinari getting a broker-dealer license was an event is precisely because that path is so narrow. If 2026 or 2027 brings a major enforcement action — say, the SEC going after an offshore RWA perp venue — institutional adoption freezes for two or three years. This is the most likely thesis-killer.

Liquidity bootstrap never works. The market makers who provide liquidity in TSLA and gold and EUR/USD (Citadel Securities, Jane Street, Virtu) are not the firms making markets in BTC and ETH (Wintermute, GSR, Amber). If cash-and-carry arbitrageurs can’t profitably close the basis between tokenized RWA spot and the perp, prices drift, funding rates blow out, and retail leaves. The venue dies. Ostium at $6B/month is a good start — but it needs to reach $50B/month to cross the institutional threshold.

TradFi takes it from above. CME has been moving — extending hours, working with Securitize and DTCC on settlement experiments. 24X National Exchange has the SEC’s first 24/5 exchange license. Bullish, EDX, and IBKR are all running on-chain settlement pilots. If any one of them ships permissionless access plus on-chain settlement — even 90% of the way — the narrative gets pulled back into a TradFi-onchain hybrid story. Crypto-native players lose the venue.

Who Wins

Three things have to be production-grade simultaneously:

Oracle and risk engineering. Sub-second multi-source pricing, weekend marks, funding rate design, liquidation engines, insurance funds. None of this can be outsourced.

Liquidity architecture. Orderbook design, market maker recruitment and incentives, cold start. Harder than crypto-native perps because the underlying liquidity providers in RWAs (traditional MMs) and crypto MMs are two different communities — bridging them is real work.

Distribution to institutions. The customer is not retail degens. It’s family offices, prop funds, crypto treasuries. The distribution motion has to look like prime brokerage, not Robinhood.

If the thesis is right, the winner of the RWA perp category over five years is at Hyperliquid-tier valuation. Hyperliquid’s $40B FDV is a useful anchor. Perps are the largest fee pool in crypto. The eventual RWA perp winner takes the cross-margin volume of two worlds — RWAs and crypto — so the addressable market is larger by definition. Tens of billions of FDV is the right order of magnitude.

Your Turn

Three things I’m not sure about — I’d like to hear what you think:

Who’s the most likely winner? If there’s a project on your radar that I missed, name it.

Who moves first? Retail in emerging markets, crypto-native treasuries, or institutional cross-asset desks? Different answers imply very different go-to-market strategies.

Where does this break that I’m not seeing? I covered regulators, liquidity bootstrap, and TradFi reverse engineering. What’s the fourth risk?

Leave a comment. I read them all.

Nothing in this piece is investment advice.

Each quadrant is facing a completely different challenges. Here’s how I’d break them down.

Crypto-Native, No Traction

This is the most crowded quadrant, and it’s also where the most capital is being destroyed.

These are the teams still pitching decks that reference TVL numbers they once helped inflate during a previous cycle, without being able to articulate why it worked the first time. They’re asking for $20M, $30M, sometimes $200M valuation on a utility token and a roadmap, claiming the token has “clear use cases” because it’s required to pay fees or vote on governance.

The data on this approach is brutal. Of 118 tracked token generation events in 2025, 84.7% traded below their launch price, with the median token down 71% on fully diluted valuation. A number of the most hyped “native DeFi L1” launches of the cycle saw their TVL drop by more than 90% within the first year, with tokens tracking the same path. The AI-adjacent token cohort averaged -50% returns for the year, with several of the top 2024 performers giving back 80%+ from their peaks.

The pattern is consistent. Initial traction came from users who want quick profit instead really like you product. Tokens priced on narrative, with no revenue or user retention to justify the valuation, bled out in 2025. Heavy emissions revealed that on-chain activity was mostly mercenary.

What this quadrant needs to internalize: a token’s long-term value comes from the team’s ability to generate revenue and return capital to holders not from contrived utility that forces users to spend it. Regulation still prevents anyone from saying “the token is equity” out loud, but empirically, that’s the only model that works. Everything else is a cyclical trade at best.

If you’re here, the honest move isn’t another token launch. It’s getting back to fundamentals: who is your actual user, what would they willingly pay for, and how do you capture a piece of that?

Crypto-Native, With Traction

This quadrant is full of teams that built something real years ago, usually during a previous cycle and have quietly been earning decent revenue on trading, lending, or swap fees. The team is small, cash flow covers payroll, the product works.

Sounds good? But they also have their challenges to overcome.

But most of them issued a token in an earlier era, and now they’re stuck with a structural problem: the revenue exists, but the token has no mechanical claim on it. Some of the largest products in the market process tens of hundred millions of dollars in monthly volume, yet their tokens captured zero direct value for years. No matter how good the revenue/profit is, the market doesn’t really trade the token on a consistent multiple and the market is pricing in a lot of expected growth, not current economics.

The buyback debate is the other half of this quadrant’s story. Some protocols committing to weekly fee-funded buybacks in early 2025 saw 40%+ price appreciation in the following month. Others running automated, fee-funded buyback programs cumulatively repurchased north of $1B in tokens over seven months, with single-day buybacks hitting nearly $4M. DeFi buybacks across the sector totaled ~$2B in 2024-2025.

Buybacks sound like an answer. Sometimes they are. But for teams in this quadrant without overflow revenue, buying back tokens is just burning future runway to defend a price that may not hold anyway. The harder and better question is whether you can grow a second revenue leg that’s not tied to crypto volatility. Because if TradFi-oriented competitors build better distribution into institutions while you’re still living off altcoin traders, your moat turns into infrastructure commodity pricing fast.

TradFi-Oriented, No Traction

This group ballooned in 2024-2025. Custody tooling, compliance middleware, tokenization rails, on-chain FX, institutional settlement, all genuinely useful. All expensive to build. All with enterprise sales cycles measured in quarters, not weeks.

The problem isn’t the product. It’s the math. Founders raised $15-30M on the premise that institutions were coming, but onboarding even one Tier-1 bank client can take 12-18 months and requires compliance infrastructure that eats a year of burn before the first dollar of revenue lands.

The good news is that the exit environment for this quadrant is unusually healthy. Crypto M&A hit a record $8.6B in 2025, with 140+ VC-backed crypto companies acquired 歐a 59% year-over-year jump. Several of the biggest deals were incumbents paying up, in the high hundreds of millions to multiple billions, for distribution, licenses, and enterprise relationships in derivatives, trading infrastructure, and payments rails.

If you’re in this quadrant, the unsentimental playbook is: manage valuation and cash runway like your life depends on hitting a meaningful M&A outcome, because it does. Don’t price yourself out of the acquirer pool. Don’t burn through 24 months of runway chasing one enterprise logo. Build complementary partnerships with bigger players who might eventually want to own you.

TradFi-Oriented, With Traction

The winners of the current regime.

Tokenized real-world assets grew from $5.5B at the start of 2025 to $18.6B by year-end, a 3.4x increase in twelve months. The largest tokenization platforms are now processing billions in institutional liquidity, with the market leader commanding roughly 20% share and backing one of the largest tokenized treasury funds in the world at nearly $3B AUM.

These companies aren’t trying to convince anyone that crypto is the future. Their institutional customers already decided. The game now is straightforward enterprise sales: win more banks, more asset managers, more issuers; build alliance structures so that when an institution buys one of your products, they naturally buy three more from your partners; squeeze unit economics on the compliance and custody stack that you already built.

If the team is a pure service provider, it becomes a classic enterprise software fight, sales velocity, net retention, integration depth.

The main risk for this quadrant isn’t competition from crypto-natives. It’s that incumbents, the big asset managers and global banks, eventually build their own rails and bypass the startups that helped them get comfortable on-chain. The window is real, but it’s not infinite.

The four quadrants look different on the surface, but they’re all navigating the same underlying shift: the market is maturing.

That doesn’t mean narrative is dead. Institutions chase hot themes too — anyone who watched semiconductor and AI valuations in the past two years knows that. But in a maturing market, narrative alone has a shorter half-life. It can still launch you. It can’t sustain you.

Tokenized equity is having a moment. You can hold tokenized Apple and Tesla on Ethereum. Robinhood launched tokenized U.S. stocks for European users last year. The narrative is that equity is finally moving on-chain, that TradFi is using blockchain for settlement.

It isn’t.

Tokenized equity is a receipt. Not a settlement.

When you buy a “tokenized” equity, the token on-chain represents a claim on shares sitting in a custodian’s account, usually inside the DTC nominee system. The blockchain records the transfer of that claim. The actual settlement, the legal transfer of ownership, still happens on the same DTCC rails that existed before any of this. The token is a digital wrapper. The real asset never moved chains.

There’s a fundamental difference between a blockchain representation of an existing security, where the off-chain record stays the legal source of truth, and native equity settlement on-chain, where the blockchain itself is the share register and on-chain records carry full legal ownership. Every headline you’ve read is about the first thing. The second thing doesn’t exist yet, anywhere, at scale.

The 98% problem nobody talks about

The real reason you can’t just swap blockchain in for DTCC has nothing to do with regulation.

Before any actual settlement happens, DTCC runs a netting process across every trade in the system. You bought $1B of Apple, someone else sold $900M of Apple. In a gross settlement world, both trades settle individually and $1.9B moves. In DTCC’s netted world, only the $100M difference settles. Their netting efficiency is roughly 98%. Only about 2% of gross notional value actually needs to move as cash and securities.

Blockchain is gross settlement by default. Every transaction settles on its own. There’s no netting layer.

Replace DTCC’s clearing function with a public blockchain tomorrow, regulatory problems magically solved, and the capital required to run the system jumps roughly 50x. That’s not a policy problem. That’s math.

The T+1 transition in May 2024 is instructive here. Shaving one day off settlement took years of industry coordination, and NSCC’s clearing fund still dropped $2.4B (20%) just from that change. Moving to real-time gross settlement on a blockchain isn’t a faster version of what we have. It’s a structurally different system.

Then there’s the corporate actions problem

Even if netting gets solved, native equity on-chain runs into a wall that tokenized equity conveniently sidesteps.

When Apple pays a dividend, DTC credits the nominee accounts, brokers receive and pass through to investors. Voting works the same way. Broadridge processes over 90% of U.S. proxy votes through a centralized pipe. Tokenized equity ignores all of this because the on-chain token is just a claim and the corporate actions flow through the same pipes as always.

Native equity on-chain would need to handle dividends, voting, rights issues, tender offers, and mergers entirely through smart contracts, with legal enforceability across every jurisdiction where investors hold the stock. No protocol has solved this. There isn’t even a credible roadmap for most of it.

So what does TradFi actually use blockchain for next?

Not equity settlement. Not perpetuals. Collateral.

Right now, institutional collateral, T-bills, money market funds, bonds, sits locked in siloed custodian accounts. Moving collateral between counterparties takes hours. Every minute it isn’t in position is capital sitting idle. If that T-bill is an on-chain token, it moves in seconds, 24/7, without settlement windows.

This is already happening. JPMorgan’s Tokenized Collateral Network is live and Fidelity International is using it. DTCC unveiled its Collateral Appchain in April 2025. BlackRock BUIDL getting accepted as margin collateral on Binance is the first real proof of concept for programmable institutional collateral, not a crypto story.

The collateral use case moves first because you’re not touching the legal record of who owns Apple. You’re making an already-owned asset move faster. Lower legal risk, immediate P&L benefit for banks, smaller regulatory footprint. The path forward is roughly this:

Tokenized money market funds are here now. Programmable collateral and real-time margin follow in the next one to two years. Intra-day repo on-chain after that. Bond issuance natively on-chain is already starting with EIB and the World Bank. Native equity settlement is a decade-plus away and requires everything else to work first.

What to watch

AUM numbers aren’t the signal. Infrastructure milestones are.

When BUIDL gets accepted as margin at CME, not a crypto exchange but a regulated derivatives clearinghouse, that’s when tokenized assets enter the real clearing system. When DTC’s tokenization services go live in 2026, that’s regulators formally recognizing on-chain entitlements in U.S. market structure. When the first company issues equity directly on-chain with no custodian twin and no token wrapper, that’s the starting gun for the decade-long transition.

None of these have happened yet.

The bottom line

Tokenized equity is the bridge. Collateral mobility is the first road on the other side. Native equity settlement is the destination, visible on the horizon but separated from here by legal rewrites, structural redesigns, and a coordination problem spanning every major financial market in the world.

The BlackRock headlines are real. The revolution is still being built, one collateral token at a time.

Back in 2009, the era internet is super hot. I had watched projects burn millions on user acquisition only to see them vanish when the hype faded.

In crypto, we chase TVL active addresses but retention is what separates survivors from ghosts. The secret sauce? Retention. It’s not just a buzzword; it’s the multiplier that turns a protocol into a powerhouse.

Why Retention Matters?

Imagine two projects, Project A and Project B, both acquire 1,000 users during a hype cycle. Each user contributes roughly $100 to TVL and generates $1 in fees.

Here is the difference 5 months later:

• Project A (High Churn): Retains 100 users.

• Project B (High Retention): Retains 200 users.

At a glance, 100 users difference seems small. But look at the gap: Project B automatically generates 2x the value of Project A without spending a dime on new marketing.

Now, scale this up to a Tier-1 protocol level with 1 million users:

• The TVL Gap: The difference explodes to $10,000,000

• The Profit Gap: That’s an extra $100,000 in pure profit fees sitting in the treasury.

Project B wins simply because its bucket doesn’t leak.

I dug into cohort retention for top protocols like Raydium, Orca, Pump.fun Uniswap, Aerodrome, and Meteora. The goal: See if they’re actually keeping users after the big acquisition pushes.

The Reality Check: Who Is Actually Holding? 🕵️‍♂️

Theory is nice, but on-chain data is brutal. I pulled the cohort analysis for the current market leaders to see who fits the “Compounder” profile versus the “Churner” profile.

I looked at two key metrics:

1. The “Plateau”: How quickly does the churn stop? (Earlier is better, it means users get “hooked” fast).

2. M5 Retention: How many users are left after 5 months?

Here is the breakdown based on on-chain data.

The data paints a clear picture: The market isn’t uniform. It splits into radically different types.

• Ecosystem First Mover: (@Raydium & @AerodromeFi, 24-31%): First mover in an ecosystem. Tied to chain health. Operations focus: Build on-chain dominance.

• Gateway (@Uniswap&@PancakeSwap, 15-18%): Casual users via aggregators mean slower plateaus. Operation focus: Brand and scalable acquisition.

• Utility (@MeteoraAG, 30-50%): Bets on efficiency and “sticky” tools.

• Casinos (@Pumpfun, <5%): Rides waves but risks burnout.

Apparently, there are two extreme retention cases - Meteora & Pump.fun

1. Meteora: High Retention as a Productivity Engine

On the high end, Meteora shows how smart design creates a “moat” you can’t cross. Their Month 5 retention hits 51%, insane for not only DeFi but a lot of web2 applications.

Why? It’s the DLMM (Dynamic Liquidity Market Maker) model.

1. Active Retention Magic: Users don’t just dump liquidity and dip. They manage “price bins” and dynamic fees to optimize yields. This turns LPs into “workers” building daily workflows.

2. The Switching Cost Hack: Once you master bins, switching to a simpler DEX feels like downgrading your phone. Result: Early plateau at Month 3

To understand the magnitude of this, we have to compare it to the segment leader: @Raydium.

Raydium is the first mover in Solana Ecosystem. Yet, looking at the 2023/24 cohorts, Raydium’s retention stabilizes around 30%. Meteora holds 51%. (Both of their 2025 retention rates were BAD <5%)

Raydium Retention (2023-2024)

Meteora Retention (2023-2024)

That ~20% gap is the “Productivity Premium.”

• Raydium Users: Passive parking. They rely on market beta (general chain volume).

• Meteora Users: Active earning. They create their own alpha through management.

How does this retention gap translate to the balance sheet? With the same amount of TVL, Meteora generates more fees than Raydium.

• Meteora: Generates ~$1.54 in fees for every $1 of TVL.

• Raydium: Generates approximately$0.42 in fees for every $1 of TVL.

To generate $1 million in trading fees, Raydium requires a substantial army of liquidity (~$2.4M TVL). Meteora hits the same goal with a lean “special ops” team (~$650k TVL).

While Meteora wins on capital efficiency, Raydium remains the king of Gross Profit.

Raydium effectively taxes the “Longtail Asset” (Memecoin) market via pool creation fees and a hard protocol cut. They own the “Casino,” and the house always wins.

Meteora, conversely, acts as an “LP Paradise.” They distribute nearly all fees to users to fuel that 51% retention. This makes Meteora a high-performance engine for users, even if the protocol takes a smaller slice of the pie for now.

But here is the kicker: The “Quality per Address.”

Raydium captures the mass market, but its data is diluted by millions of “one-and-done” bots and tiny retail dust. Meteora’s user base is far more concentrated:

• Higher Economic Value: Each Meteora address carries significantly more “gold content”, higher TVL per wallet and higher fee contribution.

• The Power of the Few: While Raydium requires a massive army of liquidity to scale, Meteora’s lean “special ops” team of professional LPs creates far more impact per capita.

Meteora can eventually capture more value by adjusting the Take Rate, but their ultimate challenge is doing so without breaking that beautiful 51% retention curve.

2. Pump.fun - Low Retention as a Traffic Casino

Flip side: @Pumpfun is the ultimate churn machine, with just 9% Month 5 retention and no plateau in sight.

1. Churn by Design: It’s a “traffic funnel” thriving on fresh gamblers, not repeat players. 98% of tokens flop; when the pump dies, users leave.

2. Product Pitfalls: No deep engagement, just click, launch, hope. Bull market cohorts (tourists chasing 10,000% APY) crater to sub-10% by Month 12.

3. Operational Reality: High but volatile fees. It needs a constant “firehose” of new blood via marketing. When Solana cools, revenue cliffs appear.

Navigating 2026: Why Stickiness Wins the Next Cycle

In the next cycle, the protocols dominating the landscape won’t necessarily be the ones with the biggest launch pads, but the ones with the flattest retention curves.

The logic is simple: Acquisition is just a cost; Retention is the business.

Instead of spending resources on infinite acquisition to fill a leaky bucket, the winning strategy focuses on analyzing churn to understand why users leave. Optimizing for the “obsessed few” yields far better operational results than chasing the “flighty many.”

The users who come for airdrops will eventually leave when the points stop. Building a product for the users who stay turns a protocol from a giveaway venue into a sustainable business.

The Retention/TVL ratio is the new litmus test.

MetaDAO has introduced several new mechanisms that are worth discussing.

1. Rewarding Point Farmers with Allocations, Not Airdrops

We’ve all seen it: A project airdrops tokens to point farmers, and they dump 100% at the TGE, sending the price into a death spiral.

In the Ranger Finance ICO, they let the point holder have the right to buy instead of an airdrop. This “Right to Buy” is a direct middle finger to that cycle:

• Filtering the Noise: By requiring real capital, it ensures that the people receiving tokens are actual participants, not just sybil bots.

• Commitment Required: Point holders don’t get free tokens; they get a dedicated allocation pool that they have the right to purchase.

Filtering the Noise: By requiring real capital, this ensures that the people receiving tokens are actual participants, not just sybil bots.

2. The Buy-back Wall

This is the project’s most discussed feature. With a minimum raise of $6M, every dollar committed beyond this threshold is funneled into a dedicated buy-back support mechanism.

If the team can secure commitments exceeding the initial target, the team gains two strategic advantages:

• Price Defense: A “Buy Wall” designed to defend the token price at or near the ICO level for the first 90 days.

• Liquidity Management: Post-launch price stability is often a significant challenge. By treating the token as a dynamic product, the team uses this extra buffer to mitigate volatility. While this requires a higher Fully Diluted Valuation (FDV) for the team to unlock their own allocations, they are intentionally sacrificing immediate upside for the capital necessary to ensure a smooth market entry.

3. Allowing VC-backed projects to launch with MetaDAO

This looks like a good adjustment to me. For the first time, a project with VC backing is launching on MetaDAO. I see this as a huge upgrade of the platform, but it’s not without its trade-offs.

• The Good: This opens the door for more deal flow. Instead of projects being extorted by expensive CEX listings or launching on a DEX with zero leverage, MetaDAO provides a legitimate third path. It’s a win-win for everyone.

• The Bad: Limited transparency remains a concern. While public records and news history identify the investors, the specific entry costs remain undisclosed. Although the 24-month vesting period is a positive sign, the lack of data regarding their cost basis makes it impossible to determine if the valuation is equitable.

A few adjustments would make it even better

Happy to see MetaDAO keeps involving itself to become a more comprehensive token launching venue, a few things I feel it can improve

Before committing capital, it is essential to examine the project’s core fundamentals. I think it would be great for MetaDAO to include this information in future project ICOs:

• Limited Information about how the business is doing: Ranger is currently generating approximately $200k–$250k in cumulative revenue. For a DEX aggregator on Solana, these figures reflect early-stage “startup” traction rather than established market leadership.

• Team information: While the founder has a visible presence online, the official documentation remains surprisingly thin regarding the broader team’s profiles and credentials. Everyone can build something, but who the builder is makes a huge difference.

What’s the final commitment of Ranger ICO and its FDV 180 days after TGE? Make a guess.

Everyone is talking about buybacks lately. It’s not always working.

In the early days of DeFi, buybacks were sold as a magic trick to pump prices. The logic seemed simple: a project takes its profits, buys its own tokens off the market, and—voila—demand goes up, supply goes down, and the price moons.

But in 2025, reality hit hard.

Projects spent millions on buybacks, yet their token prices crashed.

Why?

1. The Expectations Trap: What Should the Token Holder Expect? Are You Amazon or Apple?

It isn’t just about whether a company is “growing” or “stable.” It is about communication.

Before anything else, a project needs to send a clear signal: “Who are we?”

Every project makes an unspoken promise to its community. The market sets the token price based on that promise. If actions don’t match the story, the market gets confused. And when markets get confused, prices drop.

Let’s look at the classic examples:

• The Amazon Promise: Investors buy Amazon because they believe in future dominance. They don’t want dividends. They expect every spare dollar to be reinvested into future growth.

• The Apple Promise: Investors know the hyper-growth phase is over. They expect safety. They want buybacks and dividends because that is how a mature cash cow rewards its holders.

The Problem in DeFi:Most crypto projects pitch themselves as the “next big thing,” but they often fail to paint a clear picture of how the business will grow or what it will look like at scale.

Not all projects should be graded the same way. The goal is to accumulate a group of token holders that matches the business status.

Most crypto projects are still in the “Amazon” phase. Investors expect massive growth, new tech, and user acquisition. But then, without warning, a project might pull an “Apple” move and start aggressive buybacks.

Because nothing was communicated beforehand, holders interpret the buyback in their own way—usually the wrong way.

It screams to the market:

“We told you we were building the future, but actually... we don’t know what else to do with this cash. We ran out of big ideas.”

The Lesson:Buybacks aren’t for everyone. It comes down to defining the project’s position. Is it a high-growth builder? Or a yield-generating mature protocol?

• If the promise is growth, spend on growth.

• If the promise is value, return the cash.

Observations: When actions contradict the story, the market punishes the price.

2. The Math Problem: The Scoop vs. The Faucet

This was a common pattern in 2025: projects spending heavily on buybacks while simultaneously releasing huge amounts of new tokens.

A buyback only helps the price if tokens are being removed faster than new ones are created. If a project prints new tokens faster than it buys old ones back, the buyback effectively becomes exit liquidity for early investors or the team selling their unlocked tokens.

Think of token supply like a bucket of water:

• The Faucet: New tokens flowing in (inflation, unlocks, rewards).

• The Scoop: Buybacks (the project trying to remove tokens).

• The Water Level: The Token Price.

If the faucet flows faster than the scoop can remove water, the bucket still overflows—and the price drowns.

Case Study:

@Jupiter

($JUP) in 2025

• The Scoop: The project spent over $70 million on buybacks (only absorbing a small fraction of newly unlocked tokens)

• The Faucet: Unlocks and airdrops increased the circulating supply by roughly 150%. Hundreds of millions of new JUP hit the market every month.

• The Result: The scoop was too small for the massive faucet. The flood of new tokens drowned out the buyback. The price dropped nearly 89% from highs. By early 2026, discussions surfaced about redirecting funds to growth instead, acknowledging that the buyback had little impact on the price.

Case Study:

@AAVE

($AAVE)

AAVE demonstrated the correct math.

• The Math: Circulating supply remained relatively stable

• The Result: Because the “faucet” was tight, their “scoop” was effective. With strong revenue growth and controlled supply, their buybacks supported price stability and outperformed many peers.

The Reality Check: Data from 2025 shows a clear trend. Spending money on buybacks implies nothing unless the net supply actually drops or flattens.

Only protocols with flat supply and real business growth posted positive returns against Bitcoin. For the rest, even massive buybacks were overwhelmed by inflation.

The Observation: The size of the buyback (the scoop) is irrelevant if the new supply (the faucet) isn’t controlled first. Without fixing the supply issue, a buyback is often just “theater”, making holders feel good while providing an exit ramp for everyone else

3. The Valuation Reality: Don’t Buy the Top

A buyback is simply an investment. And like any investment, the price you pay determines if it’s a good deal.

During a hot market, tokens are expensive compared to the fees they earn. If a project buys back tokens, then they are buying at a premium.

When the market inevitably cools down, the price drops. People often look at the chart and say, “See? The buyback failed.”

But that misses the point. If buying the dip is risky, buying the top is a guaranteed failure. The project just bought the token when it was overpriced.

This is crucial right now because the market is changing. We are seeing the FDV to Revenue ratio dropping across the board. The market is becoming stricter and pricing tokens lower than before.

FDV/Revenue Ratio (Outliers > 80 removed)

• Buying High ($JUP): The project bought back heavily when the price was at its peak. When the market cooled down, they realized they had spent millions buying “expensive” tokens right before a drop.

• Buying Low (AAVE): They bought back when the token was cheap compared to how much money they were making. This added real value because they retired shares at a discount.

The Observation: Buying back tokens when they are most expensive usually isn’t a smart strategy. It effectively burns treasury cash for little return.

Conclusion

We can no longer treat buybacks as a magic button to pump prices; they are serious capital allocation decisions that require mathematical rigor, not just marketing flair.

For a buyback to actually create value, three stars must align:

1. Identity: The project must be mature enough to return capital rather than reinvest it in growth (The Apple vs. Amazon test).

2. Mechanics: The “scoop” must be larger than the “faucet.” Buying back tokens against a backdrop of high inflation is simply incinerating treasury funds.

3. Valuation: The purchase must happen at a discount, not a premium. Buying the top isn’t returning value; it’s providing exit liquidity for sellers.

Moving forward, builders and investors need to stop obsessing over the act of buying back tokens and start focusing on the efficiency of it. If the math doesn’t work, keep the cash. Build the product. The market is finally smart enough to know the difference between a value-add mechanism and expensive theater.

If there’s a view you would like to discuss, DM me on X.com

Aave’s Governance “Victory” is a Warning Sign: The Lab Model is Broken

Aave just “won” a governance vote: 55% rejected handing brand assets to the DAO. Stani Kulechov posted a conciliatory thread. Everyone moved on.

But here’s the uncomfortable truth most holders are ignoring: That vote didn’t fix anything. It just confirmed that Aave’s deepest risk isn’t Morpho, Fluid, or any other competitor—it’s the Lab + Foundation model itself.

The Paradox of Success

The protocol is objectively crushing it:

• $40B+ Total Value Locked (TVL)

• $140M DAO revenue this year alone

• GHO, V4 roadmap, cross-chain liquidity, RWA push, and a mobile app on the horizon.

Aave is the strongest lending protocol in DeFi. Yet holders keep asking: “Where’s my share of the upside?” Because the current setup funnels value the wrong way.

The Problem: Legacy Structures in a Mature Market

Labs hold near-total autonomy over integrations (like CoWSwap), frontends, fee routing, and brand ownership. That was fine in 2020—teams needed speed and legal cover.

But five years later, with real revenue flowing, semi-independent Labs acting as profit centers breeds distrust. True alignment isn’t polite forum posts; it’s structural and needs to be coded.

The Solution: From “Hired Devs” to Accountable Stewards

Labs must evolve. Every major decision should pass one test: Does it maximize long-term value for $AAVE holders?

This doesn’t mean micromanaging code. Just as Jensen Huang doesn’t poll Nvidia shareholders on every partnership, Labs need operational flexibility. However, major economic shifts and brand decisions require a “shareholder-style” veto or approval.

The Blueprint for Alignment:

1. Tie Incentives to Metrics: Link Lab rewards to revenue growth, TVL expansion, and milestones that boost long-term token value.

2. Operational Transparency: Deliver dashboards that holders can actually trust.

3. On-Chain Mechanisms: Stop relying on “trust me” messages. Lock alignment into the protocol.

Aave is a Growth Story, Not a Utility Stock

This isn’t about draining the treasury with premature buybacks or dividends. DeFi is still a land grab; payouts kill growth.

Value capture means real upside exposure. Holders should influence scale and strategy, while Labs take big swings at institutional adoption and marketing. Amazon shareholders waited years for dividends and got richer because they were betting on conquest. Aave holders should do the same.

The Path Forward

To avoid bleeding momentum, Aave must:

• Tie meaningful Labs incentives to AAVE performance metrics.

• Streamline the Foundation to pure compliance—no more middleman blame-shifting.

• Treat the DAO like real shareholders: Give Labs freedom for day-to-day execution, but set clear thresholds for value-impacting decisions.

Stani’s promise of “clearer economic alignment” is a start. But promises fade. Mechanisms endure.

What do you think?

Are you ready for real structural alignment, or are you fine with the status quo? Let’s discuss in the comments.

$AAVE #Governance #DeFi #Crypto #Tokenomics

Would you like me to generate a specific “TL;DR” summary for the Substack email preview or a list of suggested tags for the post settings?

Mostly posting my thoughts on X.com: Do follow me on @chingtsengtw to get the latest updates.

The Utility Era: A Necessary Detour

In crypto’s early days, we went out of our way to highlight every possible “utility” for our tokens. Why? To avoid being labeled securities.

The regulatory environment was immature and hostile, so we invented endless creative uses: governance voting, staking rewards, platform access, in-app payments.

The narrative was always the same: “As users consume or lock tokens for these utilities, supply shrinks, demand grows, price moons.”

But let’s be real. Does that logic actually hold?

Have you ever seen a traditional company push shareholders to spend their stock on daily transactions, expecting the share price to soar as a result? It’s inherently contradictory. If you truly believe the price will rise, why would anyone burn or give away their tokens?

This worked in crypto’s infancy because holders and users were the same people, early adopters wearing both hats.

Now, as we pursue true mass adoption, we need to evolve. Let token holders be shareholders: focused on value appreciation and governance. Let users be users: engaging with the product without needing to own tokens.

Not every Tesla shareholder drives a Tesla, after all. Separating these roles is the key to real scale.

The Shift Is Happening — But We’re Not There Yet

Things are finally moving. Regulations are maturing (MiCA in Europe, clearer SEC frameworks in the US), and tokenized assets are gaining real traction.

Projects like MetaDAO are tackling basics, such as requiring community votes to move funds. It’s progress, but it’s still not how real companies work. Effective governance doesn’t mean every token holder has to vote on every single decision. In practice, that leads to low participation, endless debate, and eventual paralysis. We’ve seen this pattern repeat across many DAOs.

In mature systems, shareholders hold ultimate power, but they don’t exercise it constantly. Instead, they delegate wisely: electing trusted representatives and defining clear boundaries upfront. Token governance can evolve in the same direction.

Token holders can vote to elect a small council or board, people they trust to handle day-to-day operational decisions such as product direction, partnerships, budget allocation, hiring, and marketing. This group acts efficiently on behalf of the collective, while remaining accountable through transparency and regular reporting.

However, when decisions touch major economic or control interests, issuing large amounts of new tokens, changing core protocol rules, taking on significant debt, or selling the project, broader approval should be required. These moments justify supermajority thresholds or direct votes by token holders.

Traditional mechanisms like drag-along and tag-along rights can even be encoded directly into smart contracts. Once a high threshold (for example, 75%) approves a sale, drag-along provisions automatically bring all holders into the transaction, preventing small minorities from blocking value creation. At the same time, tag-along rights ensure smaller holders participate on equal terms.

Key safeguards can also be pre-coded. Team tokens can vest automatically over multiple years. Spending limits can be hard-capped unless explicitly overridden by token holders. These constraints reduce both abuse and ambiguity.

Most importantly, ultimate power still rests with token holders. If an elected council underperforms or loses trust, holders can vote to remove and replace them, just as shareholders do with boards in traditional companies.

This isn’t about giving up control. It’s about making control practical and effective. Token holders retain authority where it matters most, delegate execution to capable representatives, and preserve the final backstop to intervene when needed.

That’s how we move from fragile, vote-on-everything DAOs to robust, scalable governance systems, ones that can actually endure, adapt, and grow.

The traditional Finance are also moving assets to on-chain with proper governance too

This shift isn’t happening in isolation. As of late 2025, tokenized equities are finally gaining real traction, with traditional finance migrating assets on-chain under robust, compliant governance structures.

A standout example is Superstate (@SuperstateInc) a fintech firm bridging traditional securities with blockchain infrastructure. Through its Direct Issuance Programs, launched in December 2025—SEC-registered public companies can now directly issue new tokenized shares on Ethereum and Solana, using the same CUSIP, voting rights, and economic terms as their traditional counterparts. Operating as a registered transfer agent, Superstate enables instant stablecoin settlements, real-time shareholder registry updates, and programmable controls, all without intermediaries.

For issuers, this means significantly lower costs (bypassing the 2–3% fees and delays of traditional ATM offerings), direct access to global investors, and tapping into the ecosystem’s nearly $200B in stablecoins.

For investors, it delivers equitable access to shares at real-time market prices, full economic and governance rights, plus the ability to trade 24/7 or use them as on-chain collateral.

Superstate recently partnered with @Backpack to bring these tokenized equities to a centralized exchange on Solana, allowing eligible non-U.S. users to trade SEC-registered stocks alongside crypto assets in a unified, compliant environment.

At the Convergence of TradFi and Crypto

We’re at a pivotal convergence point between traditional finance and tokens. THE DAO’s chaotic experiment is over. Lessons learned. With maturing tech and regulation, we can build practical, automated, inclusive Coin Governance.

Tokens aren’t just utilities. They’re shares in the digital economy. Let’s treat them that way. Separate holders from users and unlock true mass adoption.

Governance as code. Faster. Fairer. Truly global finance.

Micropayments feel like a natural fit for blockchains. Traditional payment systems were never designed for $0.0003 transactions or thousands of transfers per second. They rely on fixed fees, multi-day clearing cycles, and dispute windows calibrated for human behavior, not autonomous agents. Blockchains, by contrast, offer global reach and programmable transfers with low marginal costs. So when people imagine agents paying each other in tiny increments, they intuitively reach for crypto rails. But this assumption holds only until you realize that micropayments aren’t really about the fee structure. They’re about settlement guarantees, and this is where everything breaks.

A Layer 1 chain can’t finalize transactions fast enough or cheaply enough for real-time machine-to-machine payments, so the industry naturally shifts the conversation toward Layer 2 networks. L2s feel instant: a sequencer accepts your transaction, the wallet UI updates immediately, and the user believes the payment is done. But that instant UX is not real finality. It’s only a pre-confirmation. The sequencer still needs to batch your transaction, submit it to L1, wait for inclusion, and only then does it become economically irreversible. For human payments, this delay is acceptable. For an AI agent operating on millisecond time frames, it creates a structural mismatch. The agent receives the service instantly, but the payment only becomes final sometime later. If the batch is censored, delayed, reordered, or fails, the provider has already delivered something that cannot be taken back.

In other words, the business action happens immediately, but the financial finality lags behind, and the gap between the two is where all the risk sits. Once you acknowledge that gap, a question emerges that most people in this space have been avoiding: who actually carries the risk during this window? If a sequencer goes down, or the batch is reverted, or the user runs out of funds after consuming the service, someone must absorb the loss. But nobody wants to say who.

Right now, the implicit answer is: no one clearly does. Service providers are expected to deliver instantly and hope that the settlement eventually finalizes. Sequencers implicitly act as if they’re underwriting payments, but they aren’t regulated, capitalized, or insured like a real financial institution. And the idea that every user should post large collateral deposits completely defeats the purpose of micropayments. The system works only when nothing goes wrong, which is exactly the kind of assumption you cannot make in real-world commerce, especially when AI agents operate autonomously at high speed.

This is why micropayments still don’t work in production settings. Not because the technology “isn’t ready,” but because the economic model behind finality is unresolved. Agents can move faster than blockchains can settle, and in that gap lives unpriced credit exposure, default risk, and unclear accountability. The industry doesn’t need a faster chain; it needs a coherent answer to a simple but unavoidable question: Who guarantees payment when the agent has already consumed the service but the settlement is still pending?

Until that question is answered, whether through insured pre-confirmations, restaking-backed guarantees, sequencer collateralization, or an entirely new trust model, AI-agent micropayments will remain stuck in demos and prototypes. The vision is compelling. The mechanics are broken. And the missing piece isn’t throughput or fees or UX. It’s the oldest problem in finance resurfacing in a new form: finality without a clearly defined guarantor isn’t finality at all.

Let me know your thoughts and the projects you think that is solving this problem.

The reason crypto is indispensable isn’t ideological — it’s architectural.

1. Machines can’t run on credit.

The fiat stack was built for humans — authentication, KYC, chargebacks, and manual dispute loops. It assumes the buyer has an email, an identity, and patience.

AI agents can sign, but not KYC. They speak in cryptographic proofs, not passports. Fiat rails still depend on human-centric checkpoints — all of which break at machine speed.

Crypto wallets fix this at the primitive layer. Signatures replace identity; tokens replace intermediaries. x402 lets agents just pay, triggered by an HTTP header, with ERC-20 transfers verified within the same flow. It’s not UX — it’s protocol design.

2. Micropayments collapse under fiat economics.

Card networks weren’t designed for micro-transactions. Their 2–3% fees and fixed settlement costs break down below a dollar. A $0.001 query would lose money before it even clears. That’s why the web ossified around ads and subscriptions: friction made precision pricing impossible.

x402 restores true atomic micropayments to the web. On L2s like Base, transfers typically cost under a cent — low enough to meter every API call or compute cycle. “Payment Required” finally means what it should: pay exactly for what you use — no more, no less.

3. Settlement becomes a feature, not a liability.

Traditional rails treat settlement as a batch process — T+2 days, reconciled by banks and acquirers. Crypto compresses that window from days to seconds — and more importantly, makes every outcome cryptographically verifiable.

No chargebacks, no reversals — just deterministic state updates. The trade-off is intentional: machine commerce favors finality over forgiveness. That’s what enables autonomous transactions — agents can trust code, not counterparties.

4. Programmability changes the game.

Card networks can’t express conditional logic at the transaction layer. You can’t tell Visa to “pay only if the query succeeds” or “stream $0.01 every second.” Crypto can.

x402 turns payments into programmable primitives — composable, inspectable, automatable. That’s the difference between digital payments and internet-native money.

5. Tokenized cards aren’t a bridge; they’re ballast.

Putting credit cards onchain doesn’t fix the architecture. It still relies on centralized issuers, fiat settlement, and identity-linked trust. That model is structurally incompatible with a world where APIs and agents transact peer-to-peer, without permission or intermediaries.

Tokenized cards extend fiat’s reach — but they don’t make it autonomous. They replicate the old logic of control inside a new wrapper.

Crypto isn’t an add-on for x402. It’s the substrate that makes it possible.

HTTP 402 finally has a foundation that matches its intent — an open, programmable economy where computation and value converge.

x402 doesn’t make payments cheaper. It makes them native — native to the web’s protocol layer, where value moves with the same atomicity as information.

For the latest updates, follow me on X.

We are thrilled to announce our investment in Marcelo Ruiz de Olano, the visionary CEO of karpatkey, who is redefining treasury management in DeFi. By providing non-custodial treasury management solutions and infrastructure,  karpatkey has already garnered the trust of major DeFi protocols like AAVE, Uniswap, and Balancer. Here’s why we decided to back Marcelo and his exceptional team.

Marcelo Ruiz de Olano: A Visionary with a Story of Resilience and Innovation

At AppWorks, we seek founders with the vision, determination, and execution ability to address critical challenges in their fields. Marcelo Ruiz de Olano not only possesses these traits but his deep expertise in both finance and decentralized systems makes him uniquely qualified to revolutionize treasury management in decentralized finance (DeFi). His ability to navigate complex financial environments and his commitment to building transparent, secure solutions have positioned him as a transformative leader in this space.

Marcelo’s journey is a testament to resilience and adaptability. Born in Buenos Aires, Argentina, he often felt like an outsider, fueling his desire to seek new opportunities and a sense of belonging beyond his home country. During college, Marcelo longed to participate in an exchange program to explore life outside Argentina, but his parents refused to support him, dismissing it as a desire to party abroad. Determined, he found a way to finance himself and moved to Spain to fulfill his ambition.

However, the aftermath of the 2008 financial crisis made opportunities scarce. Without any financial backing from his family or the government, Marcelo faced tough challenges and was on the verge of proving his parents’ doubts right. Yet, he persevered, supporting himself through the economic downturn and overcoming the cultural differences between Argentina and Spain, despite the shared language 

Upon returning to Argentina, Marcelo joined Chevron, where he refined his decision-making skills and financial modeling on major capital projects exceeding $100 million. This role significantly deepened his understanding of global finance. By the age of 28, Marcelo had mastered the art of arbitraging stocks, bonds, and commodities, generating enough cash flow to achieve financial independence. He left the corporate world to travel and explore new opportunities. His journey eventually led him to cryptocurrency during the 2017 boom. The subsequent ICO bust became a pivotal moment, solidifying his belief in the transformative potential of decentralized finance.

Under Marcelo’s leadership, assets within karpatkey’s DeFi treasury network have skyrocketed from $300 million to over $1.8  billion in less than 3 years. His vision of creating a non-custodial, transparent treasury management system earned the trust of industry giants like Uniswap and AAVE. Marcelo’s personal and professional journey is a testament to his relentless drive to innovate and lead in the world of decentralized finance.

karpatkey: Empowering DeFi with Transparent, Non-Custodial Treasury Management and Advanced Automation Tools.

karpatkey has quickly established itself as a leader in DeFi treasury management, transforming how top organizations manage their financial assets. Originally formed in 2020 to manage the Gnosis treasury, karpatkey has since expanded its services to provide sophisticated financial solutions for renowned DeFi protocols such as AAVE, Balancer, and ENS.

karpatkey is trusted by the DeFi community due to its commitment to transparency and security. The company’s non-custodial approach allows DAOs to retain full control over their assets—a critical feature in today’s trust-driven ecosystem. karpatkey also offers advanced automation tools for risk management, ensuring that clients’ assets are managed with precision and care.

karpatkey: Redefining Finance with Innovative DeFi Solutions and Seamless Institutional Access.

karpatkey’s future vision includes expanding its product suite to serve both the DeFi and traditional finance sectors. One of the upcoming initiatives is the launch of actively managed open-ended funds, designed to give institutional investors access to blue-chip DeFi tokens and yield-generating opportunities.

karpatkey is also preparing to introduce a DeFi Exchange Traded Fund (ETF), providing institutional investors with a secure and regulated gateway to DeFi yields. This marks a significant step toward bridging traditional finance with DeFi, reinforcing karpatkey’s role as a pioneer in decentralized financial management.

Regardless of the token's intended use, startups commonly face the problem of what happens if the product doesn’t have PMF.

Reflecting on history, a startup's journey is filled with pivots and adjustments. It is rare for a startup to go public and never change its business from its initial idea or design. But in contrast, most token functions are set in stone from day one, even before the product is launched, and seldom changes. Whether the token's design has flaws or the startups haven't allowed enough flexibility for future adjustments, the token can easily become worthless if "Plan A" doesn't work. 

Most token fundraising structures in the market pose a significant problem: tokens don’t have the flexibility for future adjustments. Since the token is tied to the original product design rather than the team, tokens can easily become worthless if the product fails to find and achieve product market fit. The token also becomes obsolete if the product gains initial traction but fails to maintain momentum. Furthermore, If founders move on to the next product, token holders are essentially abandoned. 

Delaying the initial token launch until the product gets PMF might be an option for startups. This move can lower the risk of startups launching a dead token on the market and being blamed by the market for the token not working. 

Decentralization is something a lot of people care about,  but that doesn’t mean founders have no responsibility after the token launches. Letting the community decide is not an excuse a founder should use when a product is not working. Certain levels of centralization might give founders the flexibility to make adjustments to the product. 

The most important thing a web3 startup should know is that launching a token is not going to help your startup get product market fit sooner. It takes time and effort. Keep it up!

Follow me on twitter @chingtsengtw

Subscribe now