Spending on Visa-issued crypto cards jumped 525% in 2025, rising from $14.6 million in January to $91.3 million by December, based on Dune Analytics data. The scale of that increase signals a clear reduction in onboarding and checkout friction that moved digital assets closer to routine payment behavior.
Activity was not evenly distributed across the category, with EtherFi’s Visa-backed card reaching $55.4 million in net spending and Cypher recording $20.5 million. The concentration suggests early demand clustered around a small number of high-volume integrations that captured most of the initial consumer flow.
What’s Actually Improving in the Payment Funnel
The pattern reads less like a single breakthrough and more like end-to-end refinement across the conversion and checkout path. The numbers point to cumulative product improvements that reduced drop-off points rather than a one-feature story.
Programs that convert crypto at the point of sale, support multiple currencies, and attach rewards saw outsized traction in the same period. By reducing steps per operation and decision points at the moment of payment, these designs make completion more likely when users are trying to check out quickly.
From a product-engineering lens, the drivers are straightforward and highly repeatable: on-cart crypto conversion that removes manual off-ramping steps, rewards and cashbacks that reframe everyday spending value, multi-currency compatibility that reduces wallet switching, and staking or yield add-ons including programs that advertised SOL staking rewards. Taken together, these elements reduce touchpoints and lower perceived friction while preserving the familiar “card-rail” experience.
What to Watch as Crypto Cards Scale
Visa’s role shows up as an enabling layer across distribution and infrastructure, with partnerships cited at more than 70 crypto platforms, expanded stablecoin support across four blockchains, and a dedicated stablecoin advisory team for banks and merchants. Stablecoin-linked card volume across Visa and Mastercard networks was cited at $322 million, reinforcing that tokenized rails are becoming meaningful in card economics.
Incentives also shaped adoption dynamics, with some programs advertising cashbacks of 8–10% and staking-linked yields promoted by specific offerings. That mix of rebates and crypto-native utility can materially influence user decision heuristics at checkout.
. @Visa continues its expansion into crypto, steadily increasing spend volume through crypto cards such as @gnosispay, @ether_fi cash, @Cypher_HQ_, @AviciMoney, @Exa_App, @MoonwellDeFi card, and others.
Looking at the analytics for 6 crypto cards on Visa, we can see rapid… pic.twitter.com/Z5JzpBggI9
— Alex (@obchakevich_) January 4, 2026
A Polygon researcher on X (handle: obchakevich) tied the growth to Visa’s broader strategy, saying the figures show fast adoption and underscore crypto and stablecoins as strategically important to Visa’s payments ecosystem. The takeaway from that framing is that cards are being positioned as an integration layer, not just a niche product.
Regulatory and operational challenges remain part of the reality, including variability in user experiences across issuers and transaction handling issues. Scaling convenience will only hold if permission transparency and reconciliation flows are robust enough to sustain trust and operational control.
The highest-leverage work stays consistent: optimize the on-ramp, minimize steps per operation, and make permission and settlement states transparent, as participants look toward continued growth into 2026 alongside longer-range market forecasts. Execution quality will depend on reducing friction without losing clarity on state, permissions, and settlement outcomes.
Investors, issuers, and wallet teams are now watching whether stablecoin rails, reward mechanics, and merchant acceptance hold up in real usage and reporting. For UX teams, priorities center on A/B testing conversion modals, tightening transaction-state visibility, and reducing cognitive load so elapsed operation time and error rates fall materially.