Customer Service · News · 🇧🇷 Brazil
Banco BV uses AI agents in debt collection and marketing
The Brazilian bank put AI on 85% of its early-stage collections and now uses an agent to organize the requests that reach its marketing team.
- Profession
- Collections Manager, Marketing Professional
- Tools
- Agentes de IA, IA no WhatsApp
- Source
- tiinside.com.br ↗
- Country
- 🇧🇷 Brazil
- Report date
- October 6, 2026
- Verified on
- October 29, 2026
See more: AI for marketing professionals
What they did
After opening a controlled environment for teams to try generative AI, BV focused on areas with measurable returns. For customers 5 to 15 days late on payments, AI now handles 85% of collections, and all WhatsApp customer service is run by AI agents. In marketing, an agent organizes requests from other departments and turns each one into a structured brief with suggested approaches. For lower-risk pieces, AI also helps produce copy and materials.
How it went
The collections resolution rate rose from about 44% to 48%, with lower costs for the operation and outside partners. According to executive Priscila Ferreira, the same marketing team now produces 50 times more. She says some customers tell the collections agent about losing a job or family problems, even though they know they are talking to an AI.
Good to know
The figures were presented by the executive herself at an event, with no detail on how they were measured. She warns that adding an assistant to a task does not help much unless the process is redesigned. In collections, make it clear the customer is talking to an AI and offer a way to reach a person, and check consumer protection rules on how debts can be collected.
The prompt
I work in marketing at [company] and get requests from several departments through [channel, e.g. email or a form]. Turn this request [paste the request] into a brief with: goal, audience, key message, format and channel, deadline, who approves it, and what is missing that I should ask the requester. At the end, suggest two possible creative approaches.
Replace what is in [brackets]. Prompt written by us from the account, not a literal copy of the source. Do not paste other people’s personal data (privacy laws such as LGPD and GDPR apply).
Summary in our own words. Results as stated in the original account; we have not tested or audited them. Read the source ↗ Found a mistake?
Similar cases
01Toy store switches off the support agent that started getting things wrong
Topped Toys used Gorgias's AI agent for months, until it recommended products the store doesn't sell and made promises it couldn't keep.
02Pharmacy uses a voice agent to handle prescription refill calls
At Hells Canyon Pharmacy, in the US, a voice agent answers the call, identifies the patient and submits the refill request.
03Clinic automates appeals of health insurance claim denials
An n8n workflow pulls the patient record when a claim is denied and AI drafts the appeal using the clinical data.
04Requesting an exchange for defective clothing from a photo
A photo of the torn pants, a receipt found in Gmail and an email to the store, ready to review.
05Agent navigates the insurer's phone menu until it finds a human
To change a policy, the user had the agent make the call and transfer only once a representative was available.
06Support triage with draft replies for review
A five-node workflow classifies each email, rates its urgency, summarizes it and writes a reply for a human to approve.
07British car dealer group automates its customer service SMS
Stellantis &You UK closes out messages that need no action and flags unhappy customers.
08TELUS analyzes every call to fix problems before customers complain
The Canadian carrier anonymized conversations and used AI to find sentiment and root causes.