Infrastructure selection follows architecture definition.

The baseline B2B payment rail. NACHA return rate thresholds apply — programs with high return rates face compliance risk. STP architecture requires remittance design that fits within addenda constraints. Speed tier monetization opportunity: charge premium for same-day delivery over standard ACH baseline.

The 9,000-character ISO 20022 remittance data capacity is the key differentiator. Full invoice-level detail, line-item data, and structured remittance can travel with the payment. Request for Payment (RfP) capability enables pull-payment workflows. Coverage gaps relative to ACH require cascade positioning.

The Federal Reserve's instant payment rail, complementary to RTP. Combined RTP + FedNow coverage significantly improves instant payment reach versus either rail alone. Routing intelligence — detecting which instant rail a receiving bank participates in — is a program design decision. As adoption grows, combined instant coverage approaches ACH coverage levels.

The broadest coverage instant-adjacent rail. Reaches consumer and SMB payees who may not have DDA accounts on RTP or FedNow-participating banks. Critical cascade position for programs with consumer-adjacent payee populations. Higher per-transaction cost than ACH or RTP warrants premium pricing tier.

The highest interchange-potential rail for B2B programs. Level 1 interchange is the baseline; Level 2 (tax amount, customer reference) and Level 3 (full line-item detail) data transmission unlock significantly higher rates. Most programs leave 30–60bps on the table by not transmitting enhanced data. STP rate architecture — how the VCC number is delivered and processed — determines whether commercial interchange is captured or lost to manual processing. Provider evaluation dimensions: Level 2/3 capability, supplier network size, electronic delivery formats, email-based VCC acceptance.

Current use cases: cross-border B2B payments, B2B2C disbursements where 24/7 settlement matters, and programmable payment conditions. Honest constraints: limited mainstream domestic supplier acceptance, regulatory framework still developing in most jurisdictions. Position in architecture: component of multi-rail design for specific use cases — not a wholesale replacement for ACH or card rails in domestic B2B programs.

MICR quality, security features, positive pay integration, remittance insertion, return mail handling. Check printing in a well-designed program is a managed cost center — not a default fallback. Every check is an enrollment opportunity. Programs with high check volume should be running active supplier conversion campaigns. Conversion architecture: check → electronic is a separate program design effort, not a one-time campaign.

Three tiers of lockbox architecture: basic lockbox (40–60% STP), eLockbox with OCR processing (70–85% STP), and intelligent lockbox with AI and exception handling (85–95% STP). The difference between tiers is an architecture decision, not an operations retrofit — it needs to be designed into the AR workflow, not bolted on later. Programs treating eLockbox as an ops improvement are leaving significant STP gains unrealized.

Supplier enrollment automation, email-based VCC acceptance, ACH + remittance bundling. Well-designed check displacement programs achieve 60–80% electronic conversion in 90 days when the remittance design solves the receiver's reconciliation problem. The failure mode: electronic payment programs that don't convert suppliers because the remittance data doesn't eliminate manual reconciliation work. The design question isn't just "can we send VCC?" — it's "does the remittance eliminate the supplier's reason for preferring check?"