What challenges did LähiTapiola Finance face before AI implementation?

Customer service staff spent significant time answering repetitive questions and manually reviewing contract terms and compliant terminology. Feedback and formal complaints also required separate manual handling and archiving in a regulated industry.

What was the impact of the AI solution?

Freed up customer service capacity. Accelerated ticket processing for simple messages. Improved response quality and standardized complaint management and archiving.

Source

MicrosoftProductionEvidence: Low40/100

Scaling financial customer service with reliable AI agent

Use case typeCustomer service automationUpdated Jul 18, 2026

LähiTapiola Finance partnered with Siili to build Bobotin, an AI agent in production that automates preparation of customer service responses. The agent retrieves unanswered customer messages, generates response suggestions based on contract terms, previous responses, and industry terminology, and automatically categorizes feedback and complaints. The solution runs on Microsoft Azure and is designed for a regulated financial environment with GDPR and financial-sector compliance requirements.

Organization: LähiTapiola Finance
Industry: Finance
Location: Finland
Published: July 2026

Reported outcomes

Strategic outcomes

Customer experience & trustFreed up customer service capacityRisk & complianceAccelerated and standardized complaint management

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: LähiTapiola Finance
Provider: Microsoft
Maturity: Production
Linked source: Siili

LähiTapiola Finance partnered with Siili to build Bobotin, an AI agent in production that automates preparation of customer service responses

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Customer service automation
2Compliance automation
3Case management

Customer service staff spent significant time answering repetitive questions and manually reviewing contract terms and compliant terminology.
Feedback and formal complaints also required separate manual handling and archiving in a regulated industry.

Built Bobotin, a RAG-based AI agent on Microsoft Azure.
Used Azure OpenAI Service for response generation, Azure AI Foundry for prompt and model lifecycle management, and Azure AI Search for RAG vector search.
Implemented compliance and data protection using Azure services, with human review before sending replies.

Technologies

Azure OpenAI Microsoft Foundry Azure AI Search

Freed up customer service capacity.
Accelerated ticket processing for simple messages.
Improved response quality and standardized complaint management and archiving.

Architecture

Bobotin is a Retrieval-Augmented Generation (RAG) based AI agent running on Microsoft Azure. It retrieves unanswered customer messages, generates response suggestions using contract terms, response templates and historical interactions, and automatically categorizes feedback and complaints. Azure OpenAI Service provides response generation, Azure AI Foundry manages prompt/model lifecycle, and Azure AI Search supports vector search for contract terms and response history. Compliance and data protection are built into the architecture using Azure services, with human review before sending responses.

Implementation partners1

Siili

Sources & evidence1

Evidence: Low40/100Evidence strength