1. /
  2. POC
  3. /

The project aims to develop software that assists European manufacturers and retailers in detecting counterfeits, making informed assortment and pricing decisions, and identifying product innovations and trends early in the e-commerce market, addressing the growing global competition.

Start date: 01/06/2021

Duration in months: 18

Problem Description

The growing B2B and B2C e-commerce market presents both opportunities and challenges for European manufacturers and retailers. Combining image and textual data in a single search approach could create a reliable model for product identification and comparison but would be highly compute-intensive.


New services


The complexity of product assortment, pricing, and identifying innovations, trends, and counterfeits has led to the need for a new product matching method, as traditional identifiers like the European Article Number (EAN) and Manufacturer Part Number (MPN) have been disabled by many platforms.

Innovation results

Pre-processing is crucial for market analysis product matching algorithms. It transforms product data into vectors, enabling machine learning. Price Intelligence used pre-trained networks for image and text data, but the developed solution can combine both data sources for better results.

Business impact

Price Intelligence has developed a matching service for optimizing pricing and portfolios of over 100 European manufacturers and retailers, aiming to increase sales by up to 30%. The service is already being used by Parsionate for product suggestion and assortment analysis.

Project page

Follow the external link