Reduced ad-hoc data reporting request queues inside engineering teams
Inherited native, model-level data security controls for conversational AI
Expanded analytics to departments previously data isolated
Increased business intelligence tool utilization by 50 percent year over year
Replaced costly, legacy self-hosted data replication architecture
HiPay migrated legacy BI environment to Looker and BigQuery, increasing user adoption by 50 percent through governed conversational analytics.
HiPay migrated legacy BI environment to Looker and BigQuery, increasing user adoption by 50 percent through governed conversational analytics.
Graduating past fragmented legacy BI silos for trusted data
HiPay, a leading independent payment service provider headquartered in Europe, processes complex omnichannel transactions for regional merchants expanding across international borders. Operating in an intricate retail environment, the company supports over 50 hyper-localized payment methods, navigating diverse European fiat currencies like Euros, Dollars, and Pounds seamlessly. Managing this specialized transactional variety generates an immense volume of financial data.
Historically, HiPay utilized a self-hosted deployment of Tableau to deliver operational reporting. This legacy infrastructure forced the company to continuously replicate, pre-compute, and store massive data fragments across disconnected servers. The resulting data duplication bloated infrastructure costs, while slow query processing and outdated user interfaces throttled strategic corporate decision-making.
To eliminate this brittle technical debt, HiPay embarked on a comprehensive modernization strategy, consolidating its entire data ecosystem onto Google Cloud. They established a serverless, fully managed data platform centered around BigQuery, orchestrating robust data transformations with dbt and Airflow. To anchor their visualization tier, HiPay selected Looker to exploit the unique power of the LookML semantic layer.
Before rolling out the platform company-wide, the engineering team conducted a spring cleaning inventory of their legacy reporting estate, throwing away the 90 percent of inactive dashboards that constituted operational garbage.
Migrating off our outdated, self-hosted Tableau infrastructure allowed us to entirely eliminate the costly and slow data replication cycles that choked our processing performance. Centralizing our data architecture within BigQuery and Looker's LookML semantic layer acted as an essential spring cleaning that cleared out ninety percent of legacy dashboard clutter, dramatically accelerating query speeds while providing absolute trust in our core business metrics.
Anas El Khaloui
Head of Data and AI, HiPay
Centralizing their business definitions inside version-controlled LookML code allowed HiPay to eliminate data replication entirely, executing complex analytical calculations directly on the fly within BigQuery by leveraging native Looker scratch schemas and Persistent Derived Tables (PDTs).
Curing dashboard fatigue with governed conversational intelligence
Deploying centralized analytics to diverse non-technical business teams often introduces adoption friction, as standard business intelligence explorer interfaces can overwhelm mainstream users. To bridge this gap, HiPay’s BI team maintains a centralized governance model, controlling LookML explorer schemas while enabling power users across product and finance teams to safely build ad-hoc dashboards. To democratize data across all corporate levels, HiPay integrated conversational analytics directly with their semantic layer. They deployed a natural language chatbot named Transaction Expert, allowing employees to query live transaction volumes across multiple countries using plain text questions.
The combination of Looker’s semantic layer and natural language processing has completely cured dashboard fatigue across our organization. By mapping our 'Transaction Expert' chatbot directly to governed data models, we have unlocked absolute metric consistency while inheriting native security permissions. Teams who previously avoided complex data tools are now driving self-service insights effortlessly.
Anas El Khaloui
Head of Data and AI, HiPay
Because Looker’s conversational framework queries the semantic model using governed API endpoints, the interface strictly enforces row- and column-level security through Looker User Attributes and LookML access filters. This natural language interface has completely eliminated dashboard fatigue by serving answers instantly without requiring users to configure complex visualization filters.
Previously data-isolated departments, such as the sales and account management teams, are now leveraging the chatbot to instantly extract real-time figures for client presentations. To protect data integrity across these diverse user bases, HiPay implemented a rigorous dashboard certification process. Data analysts audit calculations for highly technical payment indicators, applying a digital seal of quality to authorized dashboards to guarantee absolute calculation accuracy and build cross-functional enterprise trust.
Driving business intelligence adoption and scaling advanced AI workflows
Unifying the data architecture under Looker and BigQuery has driven a profound cultural transformation across HiPay's international operations. Overall platform utilization skyrocketed 50 percent year over year, converting passive staff into data-driven decision makers. This widespread self-service adoption drastically reduced reactive data requests, resulting in a significant drop in incoming Jira support tickets.
Reclaimed from manual database administration and server maintenance tasks, the core engineering group has shifted its velocity toward building advanced machine learning capabilities on Google Cloud.
The team utilizes Google’s Model Garden on Gemini Enterprise Agent Platform to deploy and manage predictive models, alongside hosting automated intelligent agents on Google Cloud Agent Builder. For example, HiPay leverages Gemini Flash models using APIs to handle complex optical character recognition (OCR) workflows with optimal price-performance efficiency.
Looking forward, HiPay’s technical roadmap includes deploying Looker's dashboard conversational agents, utilizing the built-in Gemini pane to provide instant context and business definitions to onboarding employees. By keeping compute interactions tied directly to Looker's semantic layer, HiPay ensures that future agentic workflows can safely execute advanced operations, such as threshold alerts and cost-governed query quotas, while maintaining absolute platform stability, security, and financial control across their entire European payment footprint.
Deploying Looker has triggered a massive fifty percent year-over-year surge in business intelligence utilization, turning formerly isolated teams into completely data-driven decision makers. By deflecting ad-hoc reporting requests through natural language intelligence, we have freed our engineers to build high-value operational AI pipelines on Gemini Enterprise Agent Platform and Google Cloud Agent Builder.
Anas El Khaloui
Head of Data and AI, HiPay
HiPay is an independent European payment service provider that offers specialized, localized omnichannel payment processing and transaction analytics for international retailers.
Industry: Finance
Location: France
Products: BigQuery, Looker, Gemini Enterprise Agent Platform
