E EREACE
Back
Cut costsFintech / Neobank - Customer Service and Loyalty

Generative AI Customer Service Copilot at a Neobank

A neobank with 4M customers receives 180,000 chat enquiries a month (fees, limits, blocked cards, transfers). 70% are answered with information that is already in 1,200 pages of terms, FAQs and internal procedures, which change every month. The COO wants to reduce the cost per contact without giving invented answers that lead to complaints to the regulator.

Estimated annual impact on EBITDA
+€5.00M
Neobank with 180,000 enquiries a month. Illustrative example with indicative figures: replace them with your own company figures.

Where the money comes from

Contacts resolved without a human agent
+€5.80M
LLM and search usage cost
-€500k
Content and evaluation team
-€300k
Annual EBITDA impact
+€5.00M
Adds profitReduces profitSystem cost

What to do now

  1. 1Organise the knowledge base and create 500 test questions with compliance.
  2. 2Launch as a copilot for agents (it suggests answers).
  3. 3Move to self-service only the topics with more than 95% accuracy.

For your committee

One page with the decision, the impact, the risks, the steps and the KPI to track.

The print dialog opens: choose "Save as PDF".

Watch out: more volume is not always more profit

Measuring only the savings: if the assistant answers badly about fees, churn and complaints rise. A closed account costs more than many enquiries saved.

KPI for the committee

Cost per contact, % resolved without a human and CSAT.

Detail for your data team

Proposed technique: LLM + RAG

Why: The task is not to classify, but to generate accurate answers about knowledge that changes every month. RAG (Retrieval-Augmented Generation) separates the knowledge (documents, updatable without retraining) from the language (LLM). Retrieval provides traceability: each answer cites its source, which is key for the regulator.

Cost of getting it wrong: An invented answer about fees can end in a formal complaint and a fine; escalating too much to a human keeps the cost up. The balance is managed with a confidence threshold and banned topics.

Data needed: Knowledge base: 1,200 versioned documents (terms, rates, procedures).; Chat history: 2M conversations with the human agent's resolution.; Real-time customer data (via API): account type, card status, latest transactions.; Post-contact surveys (CSAT) and reasons for escalation.

The Ereace system

Test decisions with synthetic people before you test them with real ones.

This tool helps you decide with numbers. Before you invest, understand, test and rehearse the decision with the people involved: your customers, your team or your committee.

Understand
QualiSynth

Before you invest, find out what your customers think, need or reject about the initiative.

Test
StrataSynth

Put the offer, the message or the journey in front of synthetic people who do not always follow the happy path.

Train
ArenaSynth

Rehearse the committee presentation or the vendor negotiation against synthetic decision-makers.

See all Ereace products

About this tool. It is free and made by Ereace, an AI-native product studio and human intelligence ecosystem: it connects real human signals, synthetic people and the systems built around them, so teams can understand, test and act on human behavior before decisions reach the real world. About Ereace

Comments or suggestions? info@ereace.es