Switching Processors Requires Top Engineers
Founder of startup card issuing platform on the competitive dynamics of card issuing
The real moat in issuer processing is not the API, it is the risk of touching the live transaction engine underneath a scaled card program. Once a fintech like Cash App is pushing large dollar volume through one processor, changing that core system means rebuilding ledger logic, authorization rules, network links, and bank operations with near zero downtime. That makes big customers hard to win away, even when they have pricing leverage over the incumbent.
-
Issuer processors sit on the critical path of every swipe. They decide in milliseconds whether a card works, update balances, route messages to Visa or Mastercard, and keep records aligned with the sponsor bank. That is why the work is described as a distributed systems problem, not a simple vendor integration.
-
The migration cost is organizational as much as technical. Large programs already have engineers, ops teams, and compliance workflows built around one processor. Research on Cash App and Marqeta shows the incumbent gains stickiness from proven scale and reliability, while the customer mainly uses its bargaining power to push pricing down.
-
This also explains why newer issuers like Lithic and Highnote mostly target new builds and the long tail, not rip outs of mature enterprise programs. The easier opening is with startups launching a first card program, while Marqeta and Galileo keep an advantage where uptime, custom workflows, and transaction history matter most.
The market is moving toward a split structure. Enterprise incumbents will keep the largest scaled programs because switching is painful, while newer platforms win where customers are launching from zero and want faster setup. Over time, the best challengers will expand by surrounding the processor with better tooling, but the core engine will remain the hardest piece to dislodge.