Global Travel Retail Data & AI Platform

Client background & business context

Travel Retail · 5,700+ locations · 180+ countries · Switzerland

/Key Results

10+ Year Partnership (Since: 2015 → Ongoing: 2026)
#3 Retail Lakehouse (Scale: Global → Worldwide rank)
180+ Countries Served (Before: Siloed → After: Unified)
Live AI Workloads (Before: POC → After: Production)

/The Challenge

• Fragmented data landscape: No unified view across 5,700+ locations and 180+ countries
• Siloed integrations: Retail, F&B, loyalty, and commercial systems disconnected
• No AI foundation: Legacy warehouse architecture blocked ML and agentic workloads
• Governance gaps: Inconsistent data quality and no centralised cataloguing
• Scalability ceiling: Platform unable to support post-merger growth

/Project Goals

Legacy Stabilisation → Lakehouse Modernisation → AI Co-creation

/mindit.io’s Solution

Starting in 2015, mindit.io applied a three-phase approach: stabilise critical applications first, then modernise the data architecture, then co-create AI-driven products. The result is the 3rd largest retail data lakehouse worldwide — a unified backbone where ingestion, transformation, warehousing, and machine learning operate as a single, governed platform.

• Unified Data Lakehouse: Databricks on Azure with Delta Lake at the core — ingestion to ML in one platform
• Unity Catalog governance: Centralised data discovery, lineage, and access control across all domains
• Integration layer: Robust connectors keeping retail, loyalty, F&B, and commercial platforms in sync
• Mosaic AI & agentic patterns: ML models and AI agents promoted from POC to live production workloads
• Data observability: Automated quality checks, SLA monitoring, and alerting at every pipeline layer
• Team enablement: Embedded coaching and knowledge transfer throughout each modernisation phase

TECHNOLOGIES
Platform: Databricks · Delta Lake · Unity Catalog · Mosaic AI
Cloud: Microsoft Azure · Azure Data Lake Gen2 · Azure Data Factory
BI & ML: Power BI · MLflow · Databricks AI/BI
Integration: TIBCO · REST APIs · Event streaming
DevOps: Azure DevOps · Databricks Asset Bundles · CI/CD pipelines

/Results

• Data lakehouse rank: No unified platform → #3 worldwide (retail)
• Countries on one platform: Siloed by region → 180+ countries unified
• AI workloads in production: POC only → Live (Mosaic AI)
• Integration stability: Brittle & manual → Automated & governed
• Data governance: Inconsistent → Unity Catalog, end-to-end compliant
• Partnership duration: 10+ years and ongoing

/Customer Testimonial

“The lakehouse acts as a backbone where ingestion, transformation, warehousing, and machine learning stay in the same conversation. Modernisation is not a big bang — it is a steady march: tighten governance, make observability routine, and measure business outcomes rather than activity.” — Global Head for Commercial Applications & Integration

/Strategic Takeaways

• Stabilise first, then modernise, then co-create: Phased approach delivers value at every stage
• Embedded partnership model: mindit.io engineers work alongside client teams throughout
• Observability by default: Quality gates and SLA tracking built into every pipeline layer
• Business-outcome metrics: Progress measured by commercial impact, not activity
• Open-format architecture: Delta Lake and Unity Catalog prevent vendor lock-in at every tier

/What’s Next?

Results like these don’t happen by accident. They come from a partner who understands your data, your constraints, and what it actually takes to deliver in enterprise environments.

mindit.io is an AI-native transformation partner serving enterprise clients across DACH — with 300+ professionals, offices in Switzerland, Germany & Romania, and a track record of 200+ complex enterprise solutions delivered.

Ready to explore what’s possible? Get in touch with our team →

/ retail

Project Overview

A global travel retail group operated a fragmented technology landscape spread across 180+ countries. Legacy systems powering 5,700+ retail locations lacked a unified data backbone, making enterprise-wide analytics and AI initiatives impossible to scale. Integration between retail, F&B, loyalty, and commercial operations was brittle, slowing decision-making and limiting the ability to personalise the traveller experience.

get in touch /

helping enterprises become AI-native organizations