When a payment remains outstanding, the way a company handles the situation can determine whether a digital process provides clarity or creates further uncertainty. In a guest article for Payment & Banking, André Heyden, Managing Director of cAI Technology GmbH, part of the coeo Group, explains how AI can make this moment a better experience for customers. Heyden had previously discussed the topic at KI Exchange 2026.
In Heyden’s view, payment does not end at checkout. Resolving outstanding payments is also part of the relationship between a brand and its customers. Debt collection is therefore a sensitive stage in the customer journey. “Speed matters, but it is no substitute for context,” he writes. For Heyden, AI adds real value when it acts as a bridge between processes and people, making clear what the issue is, what information is still needed and what options are available.
With cAI, coeo has evolved its debt collection operations into an AI-native platform. It brings together interactions across letters, emails, documents, phone calls and other touchpoints, connecting them with the relevant case context and specialist knowledge. To analyse large volumes of incoming emails and documents, coeo also uses open-weight models that run within its own controlled infrastructure. This allows sensitive data flows to be managed more effectively. Legal requirements, data protection and governance provide the framework for both development and day-to-day operations.
Heyden illustrates what this looks like in practice with an example: on Sunday evening, someone realises they are unlikely to be able to make a payment due the following week. Rather than worrying until the next working day, they send coeo an email. If an instalment plan is permitted in the circumstances and represents the best solution, the customer can opt for it, and cAI confirms the arrangement immediately. A request for an individual payment extension, or a case that falls outside the system’s authorised scope, is instead referred for human review.
The decision remains with people, and responsibility rests with the company and its employees. “Responsible AI must reliably support routine tasks while recognising when it should step back,” Heyden explains. Measuring success solely by the number of cases resolved without human involvement captures only part of the picture. Clarity, consistent responses and effective handovers are just as important. In Heyden’s view, AI can build trust only when companies establish clear boundaries and take the quality of outcomes seriously. coeo’s goal is to use AI to transform how debt collection feels for the people on the receiving end.
André Heyden’s full guest article was published by Payment & Banking.
