Spark has shelved its consumer app indefinitely and is shifting its focus to stablecoin infrastructure for other businesses, arguing that a more fragmented market will create room for an intermediary layer.
The protocol, part of the Sky ecosystem formerly known as MakerDAO and developed by Phoenix Labs, now wants to supply liquidity and yield to apps users already rely on instead of building a direct retail relationship of its own.
In an interview with CoinDesk, Spark CEO Sam MacPherson said the stablecoin ecosystem is "about to become more fragmented." PayPal has PYUSD, Circle has USDC, Tether has USDT, Robinhood has joined the Global Dollar (USDG) alliance while building its own chain, and Stripe and Coinbase have formed the OpenUSD (OUSD) alliance. As liquidity spreads across more tokens and networks, Spark’s thesis is that those systems will still need to connect, and it wants to sit in the middle as the transport layer.
Spark abandons the direct consumer push
MacPherson said Spark made a key decision last year: it stopped pursuing the consumer app that had been intended to compete more directly with Coinbase, PayPal and Robinhood. In November last year, he described the app as paused rather than canceled. He now calls it indefinitely paused and said that was "absolutely the right decision."
His view is straightforward. Consumer apps are extremely hard to compete in. Instead of owning the end-user relationship, Spark is adopting a B2B2C model, providing yield and liquidity infrastructure to products that already have distribution.
Robinhood Earn becomes a live test of the model
MacPherson pointed to Robinhood Earn as a working example of that strategy. The product offers about 7% APY on USDG deposits, with funds routed into an on-chain vault on Morpho curated by decentralized advisory firm Steakhouse Financial. That vault allocates capital across three lending markets: Ethena’s USDe, Maple’s syrupUSDG and Spark’s spUSDG.
The product attracted more than $200 million in deposits within 24 days of launch, according to the report. Spark is only one piece of the stack. Morpho provides the credit network and Steakhouse curates the vault. Still, MacPherson said the structure shows the model can work.
"Robinhood is large scale. We expect this to grow into the billions of dollars," he said. For Spark, that creates a way to reach retail deposits without owning an app or the customer relationship.
Uniswap v4 stablecoin layer processed $1.5 billion in 30 days
Spark is also building what MacPherson described as a stablecoin FX layer on Uniswap to help institutions move between different stablecoins. The protocol has shifted about $150 million into Uniswap v4 liquidity pools for USDS against USDT and PYUSD.
MacPherson said the system accounted for about 30% of Uniswap’s stablecoin-to-stablecoin swap volume and processed about $1.5 billion over 30 days.
The mechanism behind it is a Uniswap v4 hook called DualPool. Idle liquidity is parked in Spark vaults to earn yield and pulled into the pool only when swaps need to be executed, with settlement taking place within a single block. Spark has also signed infrastructure agreements directly with issuers. The report said Spark worked with PayPal last year to improve PYUSD liquidity in competition with USDT and USDC.
Institutional expansion continues in a weaker DeFi market
The strategy is unfolding during a difficult stretch for DeFi. Spark’s annual revenue has fallen from about $80 million during the bull market to roughly $20 million now.
Its bitcoin-backed OTC lending business issued through Anchorage has about $260 million in outstanding balances and about $400 million in cumulative originations. The target is $1 billion by year-end, which would require outstanding balances to roughly quadruple within half a year. MacPherson said market conditions have "slightly reduced demand."
Some of that demand comes from borrowers such as bitcoin miners, who, in his words, need capital to keep operating in both bull and bear markets. He said the bottleneck is more about onboarding speed.
Spark Prime, its hybrid prime brokerage business that combines on-chain and off-chain services, has about $20 million in loans outstanding and remains in what MacPherson described as an intentional beta phase. He added that most major crypto funds are onboarding and that discussions with traditional finance firms are increasing. Part of that, he said, is tied to interest in crypto-native equities trading on venues such as Hyperliquid.
The protocol is also seeking credit ratings from S&P and Moody’s, along with assessments from crypto-native groups such as Credora, to help institutional risk teams evaluate Spark before approving it as a counterparty.
MacPherson described the current environment as "one of the easier bear markets," citing fundamentals, adoption and regulatory clarity, and said the institutional side is ready.
A $3 trillion on-chain payments bet for 2030
MacPherson sees payments as the catalyst that could turn stablecoin fragmentation into transaction volume. With the GENIUS Act set to take effect next year and the Clarity Act potentially moving forward, he said on-chain payments could reach $3 trillion by 2030.
"It looks like nothing is happening," he said. "Then suddenly a lot of things happen at once."
That leaves Spark positioning itself as a provider of rails and liquidity services. Its core bet is that fragmentation creates a valuable middle layer. But if issuers keep liquidity inside their own networks, that role becomes harder to justify.