What challenges did HiPay face before AI implementation?

Legacy self-hosted Tableau reporting required repeated data replication, precomputation, and storage across disconnected servers. Data duplication increased infrastructure costs, while slow queries and outdated interfaces slowed decision-making. Non-technical teams faced dashboard fatigue and reactive data requests.

What was the impact of the AI solution?

BI tool utilization increased 50 percent year over year. Reactive data requests dropped, with fewer Jira support tickets. Previously isolated departments such as sales and account management can now access real-time figures for client presentations. The engineering team reclaimed time from manual database administration and server maintenance to focus on advanced machine learning capabilities on Google Cloud.

Source

GCPProductionEvidence: Medium65/100

HiPay: governed conversational analytics on BigQuery + Looker with Gemini Enterprise Agent Platform

Use case typeConversational analyticsUpdated Jun 19, 2026

HiPay, an independent European payments service provider, modernized a legacy BI environment by moving from self-hosted Tableau and replicated data fragments to a serverless Google Cloud analytics stack built on BigQuery and Looker. The company centralized business definitions in LookML, used dbt and Airflow for transformations, and deployed a natural-language chatbot called Transaction Expert on top of the Looker semantic layer so employees could ask plain-text questions about transaction data across countries. HiPay also uses Gemini Enterprise Agent Platform and Google Cloud Agent Builder, including Gemini Flash APIs for OCR workflows, and plans to extend conversational agents and threshold alerts while keeping governance and security tied to the semantic layer.

AI-generated summary. Generated from indexed evidence or computed site data. Underlying records pass applicable deterministic checks; verify important details with source links where available. How content is created.

Organization: HiPay
Industry: Finance
Location: France
Published: June 2026

Reported outcomes

BI tool utilization: More than 50 % year over year

Productivity & throughput

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

BI tool utilization: 50 % year over year increase

Google Cloud Customer StoriesJun 19, 2026Customer storyExplicit claimMedium evidence strength

Increased business intelligence tool utilization by 50 percent year over year

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

Customer: HiPay
Provider: GCP
Maturity: Production
Linked source: Google Cloud Customer Stories

The company centralized business definitions in LookML, used dbt and Airflow for transformations, and deployed a natural-language chatbot called Transaction Expert on top of the Looker semantic layer so employees could ask plain-text questions about transaction data across countries

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Conversational analytics
2Data platform modernization
3Real-time analytics

HiPay consolidated analytics on Google Cloud with BigQuery as the serverless data platform and Looker as the governed semantic layer.
The BI team centralized business definitions in version-controlled LookML, used Looker User Attributes and LookML access filters for row- and column-level security, and eliminated data replication by running calculations directly in BigQuery through scratch schemas and Persistent Derived Tables.
To democratize access, the company built Transaction Expert, a natural-language chatbot integrated with the Looker semantic model so users could query live transaction volumes in plain text.
The company also uses Gemini Enterprise Agent Platform, Google Cloud Agent Builder, and Gemini Flash APIs for OCR workflows.

Technologies

BigQuery Looker dbt Airflow Gemini Enterprise Agent Platform Google Cloud Agent Builder Gemini Flash

BI tool utilization increased 50 percent year over year.
Reactive data requests dropped, with fewer Jira support tickets.
Previously isolated departments such as sales and account management can now access real-time figures for client presentations.
The engineering team reclaimed time from manual database administration and server maintenance to focus on advanced machine learning capabilities on Google Cloud.

Architecture

A serverless Google Cloud analytics platform centered on BigQuery and Looker, using dbt and Airflow for transformations, centralized LookML semantic modeling, Looker User Attributes and LookML filters for security, and conversational analytics via a Transaction Expert chatbot connected to the Looker semantic layer. The roadmap includes Gemini Enterprise Agent Platform, Google Cloud Agent Builder, and Gemini Flash APIs for OCR workflows.

Sources & evidence1

Evidence: Medium65/100Evidence strength