Skip to content

Analyse

Carrefour Voyages Agency Network Optimization

Built a Python dashboard combining clustering and interactive mapping on agency revenue, foot traffic, demographics and local competition to identify the Top 10 and Flop 10 agencies and guide opening and closing decisions.

PythonClusteringGeospatial AnalysisNetwork OptimizationDashboard

Interactive demo

The agency network dashboard

A replica of the decision tool: every agency on the map, coloured by score or cluster. Filter, click an agency to see how it compares with its peers, and switch to the Top 10 / Flop 10 or the opening opportunities.

Simulated in your browser · illustrative data

Loading the demo…

The challenge

During a three-week Business Deep Dive with Carrefour Voyages, the goal was to evaluate the performance of the physical travel agency network across France and build an objective decision-support tool to guide future agency openings, relocations and closures.

The question

How can internal sales data and external territorial indicators be combined to rank agency performance and decide where Carrefour Voyages should open or close agencies?

What I did

Cleaned and merged internal commercial metrics with local socio-demographic and foot-traffic data in Python. Applied clustering algorithms to group agencies with comparable market profiles and built an interactive Python dashboard featuring a geospatial map to benchmark every location.

Backstage · Data sources

Internal agency datasets (revenue, sales volume, customer base) crossed with territorial and external indicators by city: population, local foot traffic and competitor presence.

The result

Delivered an interactive Python mapping dashboard and a clear ranking highlighting the Top 10 high-performing / high-potential locations and the Flop 10 underperforming agencies, providing actionable recommendations for network optimization.