The dominant payment gateways — Stripe, Adyen, Checkout.com, Worldline — were built for a particular kind of internet business. The American or European SaaS company. The DTC e-commerce brand. The SaaS-adjacent marketplace operating in one or two regions. They are excellent at what they were designed for. They are increasingly poor at everything else.
The shape of modern internet businesses has changed. The fastest-growing categories — AI startups, prediction markets, gaming and esports, creator platforms, cross-border remittance, Web3 — are precisely the categories the legacy gateways de-risk, decline, or hobble. The problem is not that the gateways are bad. The problem is that the world they assume no longer matches the world that exists.
Where traditional gateways fail
Five failure modes show up repeatedly across the customers we onboard.
1. Onboarding is hostile to scale-ups
A typical legacy onboarding requires 4–12 weeks of back-and-forth, manual KYB review, repeated requests for the same documents, and an underwriting team optimised to decline anything novel. A modern internet business cannot wait 12 weeks to start accepting payments. The cost of waiting is the cost of a missed cohort.
2. Reserves are punitive on high-velocity verticals
Gateways respond to risk with rolling reserves — withholding 5–25% of revenue for 90+ days. For a low-margin marketplace this is fatal: the cash never lands when it's needed. The reserve mechanism is rational from the gateway's risk perspective, but it is incompatible with the operational economics of most modern verticals.
3. Geographic patchwork
Most legacy gateways are excellent in their core regions and weak in their adjacent ones. Pix integration in Brazil. UPI in India. AANI in the UAE. NIBSS in Nigeria. PromptPay in Thailand. Getting these rails through one vendor — let alone with consistent pricing and SLAs — is rarely possible. Businesses end up stitching three or four gateways together and paying engineering overhead to maintain the integration.
4. T+2 / T+3 settlement is a working-capital tax
We covered the float economics in our piece on stablecoin settlement. The summary: every day of delay is real money out of working capital. Legacy gateways inherit the underlying rail settlement timing; they cannot make it faster without a different underlying model.
5. High-risk vertical limitations
Prediction markets, online gaming, regulated gambling, certain types of digital-asset business, adult content, and a dozen other categories run into either outright refusals or extreme reserve and pricing terms. Some of this is genuine compliance risk; much of it is risk-policy inertia. Either way, the operators in these categories cannot use the dominant gateways.
Closed vs open payment networks
The legacy gateway model is closed. Stripe membership opens you to Stripe's bank partners, Stripe's risk policies, Stripe's settlement timing, and Stripe's vertical risk appetite. If your business doesn't fit, your only option is to find a different closed network.
Open payment infrastructure is composable. Pay-in, pay-out, FX, settlement, treasury, compliance — each is a layer with multiple providers, swappable independently. A merchant can use UPI for inbound, stablecoins for cross-border float, and SEPA for European settlement, all orchestrated through one API but none locked to a single vendor.
| Property | Legacy gateway | Open infrastructure |
|---|---|---|
| Onboarding | 4–12 weeks, manual KYB | Days, async API-driven KYB |
| Reserve policy | 5–25% rolling, 90+ days | Risk-priced; often 0% with stablecoin settlement |
| Settlement timing | T+2 / T+3 | T+0 via stablecoin float |
| Geographic reach | Strong core, weak edges | Composable per-corridor |
| High-risk vertical fit | Restricted | Underwritten on operational risk, not category |
What modern internet businesses actually need
- Self-serve onboarding with async KYB and a published risk appetite.
- Stablecoin settlement as a first-class option, not a workaround.
- Native integration of local rails (UPI, Pix, AANI, NIBSS, GCash, M-Pesa).
- Per-corridor pricing and risk policy, not a single global rate.
- Programmatic treasury — sweeps, hedging, balance management exposed through APIs.
- Audit-grade compliance: sanctions screening, transaction monitoring, partner-bank KYB, exportable reports.
Where Credible fits
We built Credible because every customer we talked to was running three or four gateways stitched together. We composed pay-in, pay-out, treasury, FX, and compliance into one API, with stablecoin settlement as the underlying float mechanism and 40+ markets accessible through local rails. The result is a single integration that handles the verticals the legacy gateways won't touch — with onboarding measured in days, not quarters.
We are not trying to replace Stripe for a US-only SaaS company. We are trying to be the gateway for everyone else.