At Payments MAGnified, I had the pleasure of sharing insights on how merchants can make smarter, more strategic payment decisions. The following article outlines the top takeaways that I believe merchants must know.
With today’s increasingly complex digital payment landscape, many businesses feel overwhelmed with the choices of how and when to apply certain tools, such as 3DS, network tokens, debit rails, or processor routing. However, each of these choices is important as they directly impact revenue by influencing transaction costs, customer experience, and authorization rates.
Luckily, the growth of AI/ML is unlocking merchants’ ability to optimize the use of each of these technologies and route transactions in the best manner to increase authorization rates and reduce costs.
Two Types of Routing
Merchants should consider two main routing types when optimizing payment performance: payload routing and processor routing:
1. Payload Routing
Payload routing adjusts the transaction payload to improve authorization rates and reduce costs. It usually factors in one or more of the following components:
- Pinless debit vs. credit rails: For some transactions, merchants can determine whether to use the pinless debit rails or major card network rails to send transactions to the issuer
- International vs. domestic rails: If operating internationally, merchants can sometimes choose to process on international rails (such as Carte Bancaires in France) or major card network rails
- Network token vs. primary account number (PAN): Network tokens can improve authorization rates and reduce payment costs, but not all issuers are accepting them yet
- 3DS authentication vs. standard: Optimizing the use of 3DS can help reduce customer friction and increase authorization rates
- Data sharing vs. standard: Sending more data to issuers can help issuers make better risk decisions
- AVS (Address Verification Service) capture vs. none: Adding AVS capture may improve some issuer authorization rates
- CVV capture vs. none: Implementing CVV capture may enhance some issuer authorization rates
Each of these payload decisions, however, comes with tradeoffs. When optimizing each decision, it’s important to consider three key areas when making a choice. First, evaluating authorization rate is essential to understanding how a specific issuer responds to different technologies. For example, does your authorization rate improve when using network tokens vs. PAN? And does 3DS authentication help or hurt you?
The second key factor to consider is payment costs. There are differing costs to different decisions. For example, network tokens and pinless debit rails are usually cheaper to process than primary account numbers (PAN) or major credit network rails.
The third and final key factor to consider is transaction risk and how easily chargebacks can be disputed. Some considerations can help minimize fraud, while others might offer more flexibility in challenging disputes.
2. Processor Routing
The second main area of routing is processor routing. For merchants using multiple processors, choosing the right one for each transaction can be challenging. When deciding which processor to use, important considerations include:
- Cost: Which processor offers the lowest fees for a particular transaction
- Authorization rate: Some processors perform better than others depending on the issuer, transaction type, or geography
- Failover/Redundancy: If one processor goes down, having an alternative ensures transactions can still go through. Additionally, merchants can decide to retry a transaction with another processor or retry with the same, but with a different payload (for example, retrying the transaction with a PAN if a network token fails)
- Processor-specific value-added services: Some processors offer additional benefits that could enhance performance
Strategically selecting a processor can optimize both costs and performance, ensuring higher approval rates while keeping fees low.
Predictive Routing: The Next Level
If all of this analysis sounds like a lot to track — you’re not alone. Managing authorization rates, costs, and processor performance manually is overwhelming. That’s where predictive routing powered by AI/ML comes in. While AI/ML is already widely used for fraud prevention, payment routing is still heavily reliant on manual rules — which means missed opportunities for optimization.
Predictive routing automates payment decisions by optimizing transactions based on real-time data. When leveraging AI-driven routing, merchants gain the ability to:
- Automatically optimize payment routing without constant manual oversight
- Ensure business priorities are considered in routing decisions
- Improve revenue impact by dynamically adjusting to changes in costs and authorization rates
Key Takeaways
The bottom line? Smarter routing means better revenue outcomes. As payments continue to evolve, merchants who embrace data-driven, AI-enhanced decision-making will be in the strongest position to reduce costs, improve authorization rates, and enhance customer experience.
Here are a few parting reminders I’d like to leave you with:
- If you want to gain more control over outcomes, consider a multiprocessor strategy
- Take the time to fully understand the payment decisions within your control and how to optimize them
- If you leverage AI/ML for predictive routing, you will be able to ensure flexibility and long-term scalability
Galit Shani-Michel is the VP of Payments at Forter, where she merges operational acumen with a deep understanding of compliance to drive strategic payment solutions. With years of experience leading teams through complex fraud and regulatory challenges, Galit is dedicated to helping businesses navigate risk, optimize revenue, and thrive in the ever-evolving payments landscape.
