How AI agents with cross-session memory transformed banking customer service on WhatsApp
25 de fevereiro de 2026
7 min read
Source: tiinside.com.br
-73%
Repeat Calls
24/7
Availability
Zero
Hallucination
Banco BV took a decisive step by migrating from traditional chatbots to autonomous AI agents on WhatsApp. The difference isn't semantic — it's functional. While chatbots follow scripts and decision trees, BV's agents understand intentions, adapt to context, execute tasks end-to-end, and maintain conversation memory across sessions.
The most striking result: a 73% reduction in repeat contacts — cases where customers call back about the same issue within 30 days. This indicates the agents didn't just respond — they resolved.
-73%
Repeat Calls
24/7
Availability
Zero
Hallucination
Banco BV faced a common banking problem: customers contacting multiple times to resolve the same issue. This repeat-call pattern is a direct indicator that the previous service was not resolution-oriented — merely informational.
The bank needed a solution that went beyond answering FAQs. They needed agents capable of executing actions, maintaining context across interactions, and ensuring each conversation led to a concrete resolution.
Banco BV implemented autonomous AI agents on WhatsApp using a specialized Brazilian platform. The key technical differentiators include:
Cross-session memory: Agents maintain context from previous conversations, eliminating the need for customers to repeat information with each new contact.
End-to-end execution: Agents don't just inform — they execute tasks such as queries, profile updates, and process routing.
Zero hallucination: Agents were trained exclusively with official bank information, eliminating the risk of incorrect or fabricated responses.
24/7 availability: Service operates continuously, covering off-hours that previously generated frustration and demand accumulation.
The headline result — -73% in repeat calls — is particularly significant because it attacks the root of the problem: non-resolution service. When customers don't need to call back, it means their issue was genuinely resolved on the first interaction.
24/7 availability expanded the service window, capturing demands that previously could only be addressed on the next business day. The zero-hallucination guarantee is critical in the financial sector, where incorrect information can generate legal and financial consequences.
Banco BV's case demonstrates that the transition from chatbot to autonomous agent is not incremental — it's transformational. The difference between answering and resolving is the difference between 73% more or fewer repeat calls.
Catalisa enables creating autonomous agents with persistent memory and transactional execution, like Banco BV's. The platform's AI Agents maintain context across sessions, execute complex workflows, and integrate with banking systems via Building Blocks.
The zero-hallucination guardrail is ensured by training with a controlled knowledge base, while visual Workflows allow defining end-to-end resolution flows without code.
TI INSIDE. “Banco BV escala uso de agentes de IA para transformar atendimento via WhatsApp”. tiinside.com.br. Disponível em: https://tiinside.com.br/09/12/2025/banco-bv-escala-uso-de-agentes-de-ia-para-transformar-atendimento-via-whatsapp/. Acesso em: 27 fev. 2026.
BAGUETE. “Banco BV atende no WhatsApp com agentes de IA”. baguete.com.br. Disponível em: https://www.baguete.com.br/noticias/banco-bv-atende-no-whatsapp-com-agentes-de-ia. Acesso em: 27 fev. 2026.
Schedule a demo and see how Catalisa can transform your business
Analysis of how AI agents on WhatsApp evolved from reactive bots to proactive business consultants, with cases from iFood (Gerente, Cris) and Magazine Luiza.
11 de jan. de 2026
·
9 min read
How Synthesia, Lotte, and Carrefour absorbed exponential growth in service demands without proportional hires, using AI as a force multiplier.
04 de fev. de 2026
·
9 min read
Side-by-side comparison of how Brazil's three largest banks use artificial intelligence on WhatsApp to serve 74 million customers
26 de fev. de 2026
·
10 min read
·
Source: microsoft.com
© 2026 Catalisa. Todos os direitos reservados.
Conforme LGPD
Meta WhatsApp Business
Dados no Brasil
We use cookies for analytics and load external resources (images) to improve your experience. By clicking "Accept", you consent to the use of cookies and third-party resources in accordance with our Privacy Policy.