Who is Lagardère Travel Retail?
The challenge
Forecasting in travel retail is special. There’s a constant ebb and flow of customers, with a vast range of customer profiles, encompassing people of countless nationalities, cultures, languages, socioeconomic status, age, occupations, … And it can change dramatically from one day to the next. Traffic flow can be intense or sporadic, seasonal factors such as holidays, weather and events can be strong predictors of sales … or be irrelevant.
![](https://cdn.statically.io/img/i0.wp.com/jetpack.ai/wp-content/uploads/2024/06/Lagardere_retail.jpg?fit=1920%2C1285&ssl=1)
Our approach
Jetpack.AI developed a solution tailor-made to Lagardère requirements that provides optimal data analysis to deliver real-time actionable insights.
-
Fine-grade forecasting and real-time performance analysis and alerting.
-
Resource planning with real-time updates depending on traffic and staff presence.
-
Optimal product assortment and merchandising, lists of key traffic builders and basket generators.
-
Optimal shop locations across the airport or train station.
-
…and many more custom-built answers to analytically empower the people in the shops.
The results
+5
revenue per passenger in airports
Jetpack.AI is delighted to have contributed to Lagardère’s success. The Client has reported that thanks to the solution provided they have achieved:
-
+ 5% SPP (Sales Per Passenger) in just a few months
an improved entrepreneurial spirit among store managers
-
increased inventory accuracy
-
increased presence of store managers on the shop floor
better product tracking, with reduced losses due to theft prevention.
![](https://cdn.statically.io/img/i0.wp.com/jetpack.ai/wp-content/uploads/2024/06/Nicolas_Vanbrands.png?fit=647%2C573&ssl=1)
”Jetpack helps our company to discover underlying facts, trends and figures that we translate into major strategic turns and immediate value creation. And not the least to say that everything is done within short and interactive processes
Nicolas Van BrandtCEO - LTR Benelux
Tech stack
The whole solution is hosted on Amazon Web Services (AWS). The backend engine consists of Scala/Spark aggregation pipelines running on Amazon EKS. Those engines are orchestrated via Airflow on Managed Workflow for Apache Airflow (MWAA). The Kubernetes clusters are managed via Argo CD. A series of forecasting engines are running with Python.
-
The resulting data are written in a PostgreSQL database hosted on Amazon RDS (Aurora).
-
The API is built using nodeJSand hosted on AWS Elastic Beanstalk controlling EC2 instances.
-
The front-end progressive web application (PWA) is built using ReactJS. Charts are made using a combination of d3.js and Plot by Observable libraries. The application is hosted on AWS S3 and served using Cloudfront.