Automated Analysis of Retail Flyers

The project

For a leading retailer in the Pet Care segment, we developed a fully automated system for the analysis of promotional flyers in the retail sector.

The solution integrates web scraping techniques and computer vision models to capture, recognize, and structure in real time the information contained in digital flyers.

• Web scraping scripts enable systematic and continuous acquisition of flyers published online.
• Computer vision models automatically identify and segment graphical elements, extracting prices, products, and discount information.
• The data is then normalized and structured into Excel format and stored in an internal data lake, creating an integrated information base that combines internal and external sources.

Results

The new process reduced manual data collection time by 95%, eliminating low-value repetitive tasks while ensuring a much broader and more up-to-date coverage of the retail promotional landscape.

With standardized, real-time data, the company was able to enhance pricing analysis and competitive benchmarking, improve the speed and accuracy of commercial decision-making, and transform a complex data collection activity into a tangible and scalable competitive advantage.