Use Case Intelligence RSS Search use cases Search
In-house Financial Services Global Company newsroom

BNP Paribas Cardif

Customer Service Automation · Real-time Analytics

BNP Paribas Cardif handles more than 20 million customer phone calls each year, but manual review was too limited and slow to extract actionable insight from conversations for service improvement, product fit checks, and customer journey optimization.

Read the original story Explore all use cases
At least 30 minutes of listening and analysis for a 15-minute telephone call
More than 20 million customer phone calls are handled every year
over 10,000 calls in 3 months

Reported impact

Business outcome

The solution enables faster, broader analysis of customer conversations so BNP Paribas Cardif can improve service quality, optimize telephone insurance processes, strengthen customer loyalty actions, fine-tune sales training, and improve products based on policyholder feedback such as cancellation reasons.

  • Customer Experience
  • Productivity Improvement

Solution

BNP Paribas Cardif built an in-house AI solution called Call Analyzer to industrialize analysis of customer phone conversations. The implementation starts with call audio files sent to the application through an API. Call Analyzer uses speech-to-text AI models to transcribe recordings into text, then applies AI analysis to the transcripts to examine vocabulary, speech speed, lexical field, and other conversation signals. The processed outputs are converted into statistics and verbatim comments that feed dashboards in data visualization tools. Those dashboards are consumed by marketing and customer operations teams to monitor KPIs, identify service issues, refine telephone insurance subscription processes, and assess whether products sold over the phone match customer needs. The platform is already operating across several Latin American countries and is being deployed more broadly in Europe, with newer generative AI capabilities now added to deepen analysis.

Data flow

Customer call audio files are sent via an API into Call Analyzer, where speech-to-text AI models transcribe the recordings into text. The system then analyzes transcript content such as vocabulary, speech speed, and lexical field, and turns the results into statistics and verbatim comments that are surfaced in dashboards for marketing and customer operations teams.

Solution architecture

3 components · 3 layers
  1. Compute
    • Speech to Text artificial intelligence models AI models that transcribe customer call audio files into text for downstream analysis.
  2. Serving
    • data visualization tools Dashboarding layer used by business teams to monitor KPIs and act on the analyzed conversation data.
  3. Orchestration
    • Call Analyzer In-house application that receives call recordings, orchestrates transcript analysis, and produces statistics and verbatim outputs for operational review.

Architecture clues

  • Built by the Analytics and IT teams at BNP Paribas Cardif
  • Deployed in several Latin American countries and expanding into Europe
  • Produces statistics and verbatim comments for dashboard consumption
  • Uses API-based ingestion of audio files

Evidence from the source

"BNP Paribas Cardif in Peru successfully analyzed over 10,000 calls in 3 months."
"Before artificial intelligence came along, the ability to listen to these calls was limited and time-consuming: at least 30 minutes of listening and analysis for a 15-minute telephone call."
"Call Analyzer now uses some of the capabilities offered by generative AI"
"Call Analyzer ultimately generates statistics and verbatim comments that can be used to create dashboards"
"More than 20 million customer phone calls are handled by BNP Paribas Cardif ... and its partners every year."
"Then the solution uses AI to analyze the content of these transcripts by looking at vocabulary, speed of speech and lexical field, etc."
"Via an API (Application Programming Interface), Call Analyzer receives audio files and transcribes them into text using so-called 'Speech to Text' artificial intelligence models."
Transparency Evidence & methodology 90/100 reference value · 94% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value90/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact15/25
    1 metric describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence94%
  • Source authority25/25
    The deployment is described by the adopting organization or a government source.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    7 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. BNP Paribas Cardif

    BNP Paribas Cardif customer story

    First-party
    https://group.bnpparibas/en/news/ai-at-bnp-paribas-2-analyzing-telephone-conversations-with-bnp-paribas-cardif-customers-to-improve-service-quality
    Source type
    Customer-published story
    Published
    Feb 6, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Feb 6, 2025 · BNP Paribas · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
In-house Sep 3, 2026 Grupo Falabella Retail Latin America Grupo Falabella needed to turn AI into a cross-functional business capability that improves customer experience, sales conversion, service responsiveness, and logistics efficiency across its five business units rather… US$265 million to technology in 2026 Customer Service AutomationMarketing Personalization Banco Falabella app · Falabella.com · cameras Microsoft Mar 5, 2025 HYPE Financial Services Italy HYPE promises customers a streamlined, simplified, and personalized banking app and needed a cloud-based CRM to optimize the value of its data to enhance its customer journeys. Dynamics 365 Customer Service, Copilot in Customer Service, and Copilot Studio met HYPE's omnichannel needs and… Customer Service AutomationReal-time Analytics Copilot Studio · Dynamics 365 Customer Insights · Dynamics 365 Customer Service Microsoft Dec 27, 2024 Bank of Queensland Financial Services Australia Bank of Queensland wanted to launch a digital modernization strategy to ensure agility and resilience, build and scale for the future, and support its customer-first approach to business. Deploying a solution that included Microsoft Azure, Microsoft 365, and Microsoft Copilot, Bank of Queensland… Customer Service AutomationReal-time Analytics Copilot Studio · Microsoft 365 · Microsoft 365 Copilot