top of page

Zayon Develops Hybrid Banking Recommendation Model with Explainability Built In

Writer: Zayon
Zayon
Nov 10, 2025
2 min read

Updated: 1 day ago

Tela preta com logo Zayon e texto Trustworthy AI Decision Infrastructure; círculos vermelhos à direita.

São Paulo, Brazil, November, 10, 2025: Zayon has developed a new hybrid recommendation model for banks and credit cooperatives, designed to help institutions recommend the right financial products to each customer while keeping every recommendation explainable and within the institution's business rules.


Zayon's engineering team tested several architectures before arriving at a hybrid approach that combines complementary machine learning techniques. In internal testing, the model achieved above 90% accuracy in its recommendations and outperformed the previous baseline.


Performance was only one of the design criteria. Every recommendation the model produces is explainable in terms of the institution's own business rules, so relationship managers can understand why a product is suggested for a given customer, and the institution can verify that recommendations follow its eligibility and suitability criteria.


The model separates what is predicted from what is decided. Machine learning identifies which products are relevant to each customer's situation. The institution's rules determine what can be recommended, to whom and under which conditions. Relationship managers decide what to offer.


"A recommendation is only useful if the institution can trust it, explain it and govern it," said Deborah Nogueira, Founder and CEO of Zayon. "Better models matter, but what institutions need is recommendations that follow their rules and that people can act on with confidence."


The model is being incorporated into Zayon's recommendation solutions for banks and credit cooperatives seeking to improve profitability without increasing risk.


About ZayonZayon builds Trustworthy AI Decision Infrastructure that turns AI capabilities into decisions institutions can trust, govern and audit. Its XAID architecture binds intelligence, evidence and policy into decisions that are explainable, governed and auditable, starting with financial institutions in Brazil. Zayon was founded in Brazil by specialists in economic engineering, AI and large-scale public-sector projects. Learn more at zayon.tech.


Media contact: secom@zayon.tech

Related Posts

See All

Comments


The Zayon Briefing

Topics you're interested in:

A periodic briefing on AI decision infrastructure, governance and what it means for financial institutions. No noise.

STAY INFORMED 

bottom of page