Adyen

US regulatory changes effective July 1, 2023 mandated that all US-issued debit cards carry branding from at least two unaffiliated payment networks. For enterprise merchants, this created a routing dilemma at scale: choosing between the lowest-cost network and the highest-authorization-rate network with every transaction. The Federal Reserve's 2024 consumer payment study found debit accounted for nearly one-third of all US payments, with transaction volume up 27% from 2022 to 2023. On Adyen's platform alone, debit reached 58% of all US transactions by June 2024. Without automated optimization, merchants absorbed unnecessary processing fees or accepted lower authorization rates — a cost burden that compounded directly with rising debit volume.

Adyen developed Intelligent Payment Routing, an AI-based feature that applies machine learning and predictive analytics to evaluate every debit transaction in real time and select the optimal network path based on both transaction fees and authorization probability simultaneously. Rather than requiring merchants to configure manual routing rules or maintain static network preferences, the model dynamically weighs cost and performance signals on a per-transaction basis. Adyen's vertically integrated, in-house-built platform enabled a zero-code deployment model: merchants activate the feature with a single toggle, with no additional integration work. The solution operates across all transaction channels — online, in-store, and in-app — drawing on cross-channel data to inform decisions. Because routing logic runs natively within Adyen's infrastructure, it adds no latency to transaction processing. The feature launched as a market-first offering in September 2024 following a pilot with over 20 enterprise merchants.

The pilot cohort — which included eBay, 24 Hour Fitness, and Microsoft — delivered improvements across both cost and authorization dimensions:

• 26% average cost savings on US debit transactions across the pilot group
• 0.22% average authorization rate uplift against prior debit performance
• Up to 55% cost reduction for the top-performing merchant; a second merchant achieved 52%
• Authorization rate improvements up to 1.15% for individual merchants
• $600,000 saved by a single customer within the first 30 days of activation

Critically, cost savings and acceptance rate gains moved in the same direction simultaneously across the cohort, disproving the assumption that these objectives inherently trade off against each other.

• Regulatory mandates that introduce routing complexity — like multi-network debit requirements — can create structural optimization opportunities; where compliance has added inefficiency, AI can systematically resolve it.
• A vertically integrated platform eliminates integration friction: zero-code feature adoption compresses time-to-value from a multi-month engineering project to a configuration toggle.
• Machine learning can optimize across competing objectives simultaneously; the conventional trade-off between authorization rates and processing fees is not technically inevitable.
• Debit transaction volume growth (up 27% in the US from 2022–2023) means routing optimization compounds in value — the savings potential scales directly with volume.
• Pilot results across a diverse enterprise cohort establish credible benchmarks and surface variance in outcomes before broad rollout.