Use Case Intelligence RSS Search use cases Search
IBM Telecommunications Czech Republic

O2 Czech Republic

Customer Service Automation

O2 needed to manage a broad mix of 24x7 customer-service requests more efficiently while still giving customers fast answers across voice, mobile-app and operator-assisted channels.

Read the original story Explore all use cases
Approximately 30% of O2 customers communicate exclusively by voice with the virtual assistant
more than 250,000 questions a year
nearly 1,000 scenarios

Reported impact

Business outcome

Eva now handles a significant share of support traffic across voice and digital channels, giving customers faster self-service while reducing the load on human representatives and broadening the reach of AI-assisted care.

  • Customer Experience
  • Productivity Improvement
  • Scalability

Solution

O2 worked with IBM and AddAI.Life to create Eva, a virtual assistant built on IBM watsonx Assistant running on IBM Cloud. The assistant was trained on nearly 1,000 scenarios and connected directly to O2's customer-support systems so it could interpret intent, retrieve account-related information and guide users through tasks rather than only answering static FAQs. Within the mobile app, Eva greets users, handles requests and automatically redirects unresolved issues either to the correct page on O2's website or to a specialized operator. The implementation also extended to voice interactions, with Microsoft-supported voice synthesis used to give Eva a recognizable spoken persona for the app, support line and advertising. By treating watsonx Assistant as the decision-making core while surrounding it with customer-service integrations, O2 created a multichannel AI support layer that scales across billing, roaming, order-status and service questions.

Data flow

Customer questions arrive through the mobile app or voice channels, watsonx Assistant running on IBM Cloud interprets intent, exchanges data with O2's support systems for billing, order-status or roaming information, and returns an answer or routes the request to the right operator or web page.

Solution architecture

2 components · 2 layers
  1. Serving
    • IBM watsonx Assistant Acts as Eva's conversational brain, interpreting user intent and orchestrating answers or guided actions.
  2. Orchestration
    • IBM Cloud Hosts the assistant runtime that exchanges data with O2 support systems and powers the virtual-assistant experience.

Architecture clues

  • Eva communicates directly with the O2 customer support system for targeted answers.
  • The deployment spans the mobile app, customer support line and website handoff workflows.
  • The voice layer was synthesized from a named Czech actress and reused across multiple channels.

Evidence from the source

IBM watsonx Assistant to create a virtual assistant, which they named 'Eva'
The virtual assistant has been trained on nearly 1,000 different scenarios
The watsonx Assistant running on IBM Cloud is, so to speak, Eva's brain
about 30% of O2 customers now can communicate exclusively by voice with EVA
Transparency Evidence & methodology 90/100 reference value · 89% 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 confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    4 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. IBM

    O2 Czech Republic customer story

    First-party
    https://www.ibm.com/case-studies/o2-czech-republic
    Source type
    Vendor-published customer story
    Published
    May 6, 2024
    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

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
IBM May 15, 2024 Vodafone Ireland Telecommunications Ireland Vodafone Ireland needed to modernize its TOBi virtual assistant because an older assistant deployment could no longer meet containment goals or take advantage of the newer generative-AI, NLU, and self-learning… 3x faster for creating new conversational journeys Customer Service Automation IBM watsonx Assistant IBM Apr 18, 2024 Vodafone Telecommunications Europe and Africa Vodafone wanted to use generative AI to speed up how its teams design, test, and improve TOBi conversational journeys because manual journey testing and gap analysis slowed releases and limited how much optimization the… 99% improvement in turnaround time of journey testing Customer Service Automation IBM Cloud Code Engine · IBM watsonx Assistant · IBM watsonx.ai IBM Apr 3, 2024 WINDTRE Telecommunications Italy WINDTRE needed to automate and accelerate claims handling for telecom service issues because high report volumes, stateful operational complexity, and repetitive service-desk tasks were slowing response times and… 10,000+ reports handled per month with AI and automation Customer Service AutomationIT Operations Automation IBM Cloud · IBM Consulting · IBM Watson Knowledge Studio