What challenges did Hedvig face before AI implementation?

Rapid growth led to siloed, inconsistent data causing reporting delays and analytics inefficiencies. Needed a unified data definition and scalable self-service analytics for multiple departments.

What was the impact of the AI solution?

Enabled teams to independently create dashboards and explore data. Reduced ad hoc reporting time and increased data team productivity. Improved predictive pricing models through focused AI data preparation. Set foundation for AI-driven conversational analytics to enhance decision making.

Source

GCPExploringEvidence: Medium55/100

Hedvig uses Google Cloud Looker and Vertex AI to scale insurance analytics with trusted semantic layer

Use case typeRisk assessmentUpdated Jun 13, 2026

Hedvig replaced siloed data infrastructure with a unified semantic layer using Looker and BigQuery to enable self-service analytics. Implemented advanced data workflows with dbt and Vertex AI for predictive pricing models and AI-ready data structure. Democratized data access across claims, underwriting, and sales teams, reducing manual report generation. Exploring conversational AI to allow natural language queries on self-explainable data models. Achieved rapid claim processing and scalable analytics without increasing data team headcount.

Organization: Hedvig
Industry: Insurance
Location: Sweden
Published: May 2026

Reported outcomes

Strategic outcomes

Better decisions & insightUnified data definition for analyticsCustomer experience & trustEnabled self-service dashboard creationNew product / capabilityAI-ready conversational analytics foundationScale & capacityScaled analytics without added headcount

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

Customer: Hedvig
Provider: GCP
Maturity: Exploring
Linked source: Google Cloud Customer Stories

Exploring conversational AI to allow natural language queries on self-explainable data models

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 2 of 2

1Data Democratization
2AI-Enhanced Insurance Analytics

Rapid growth led to siloed, inconsistent data causing reporting delays and analytics inefficiencies.
Needed a unified data definition and scalable self-service analytics for multiple departments.

Adopted Looker as a trusted semantic layer with a central business logic definition.
Used BigQuery and dbt for heavy data transformation and machine-learning readiness.
Refactored data models for AI agent compatibility and conversational analytics.
Empowered business users to build dashboards independently, reducing data team load.

Technologies

Looker BigQuery dbt Vertex AI

Enabled teams to independently create dashboards and explore data.
Reduced ad hoc reporting time and increased data team productivity.
Improved predictive pricing models through focused AI data preparation.
Set foundation for AI-driven conversational analytics to enhance decision making.

Architecture

Unified Looker semantic layer combined with BigQuery, dbt transformations, and Vertex AI for AI-enhanced insurance analytics and agent-ready data models.

Sources & evidence1

Evidence: Medium55/100Evidence strength