Mumbai: Criteo, the global platform connecting the commerce ecosystem, has introduced its Agentic Commerce Recommendation Service, a new solution designed to power AI shopping assistants with highly accurate, commerce-grade product recommendations built on Criteo’s proprietary commerce intelligence.
As large language model (LLM) platforms evolve into AI-powered shopping assistants and retailers develop their own AI chatbots, the way consumers discover, compare, and purchase products online is undergoing a fundamental shift. With these AI-led shopping journeys scaling rapidly, the need for recommendation systems that go beyond static product descriptions and tap into real shopping behavior has become critical.
Criteo’s new service addresses this gap by enabling AI assistants to access real-world shopping and transaction signals, helping deliver personalized and trustworthy recommendations. The company said the solution builds on its previously outlined agentic commerce vision.
According to Criteo, internal testing showed that the Agentic Commerce Recommendation Service delivered up to a 60% improvement in recommendation relevancy compared to third-party approaches that rely solely on product descriptions. This uplift is powered by Criteo’s large-scale commerce data, spanning 720 million daily shoppers, $1 trillion in annual transactions, and 4.5 billion product SKUs.
The service is available through Criteo’s Model Context Protocol (MCP) and connects AI-powered shopping assistants directly with merchant inventory. It translates consumer queries into curated, transaction-ready product recommendations, applying real purchase and browsing signals that are not accessible through traditional web crawling.
How the Service Works
When a consumer asks an AI shopping assistant for a product aligned with their needs, preferences, and budget, the assistant queries Criteo’s Agentic Commerce Recommendation Service. Criteo then filters and ranks products using real-world commerce data, factoring in popularity, availability, and user intent. Instead of returning raw catalog data, the system delivers a curated shortlist of relevant products.
The AI assistant can then present these results, compare options, and even support add-to-cart or checkout within the same agentic experience. The service is designed to understand both broad and specific shopping intent and can suggest complementary products where relevant.
“The real competitive advantage in agentic commerce will come from access to high-quality commerce data at scale,” said Michael Komasinski, Chief Executive Officer of Criteo. “This service brings that intelligence into AI-driven shopping experiences in a way that works for the entire ecosystem, delivering relevancy for consumers while respecting retailer data, brand integrity, and platform trust.”
Criteo said it has been testing the service with a major LLM platform since 2025 and is now expanding trials to additional LLM platforms, retailers, and brands. The company added that its definition of recommendation relevancy is “the degree to which a product matches a shopper’s current intent, needs, and preferences, and therefore helps them progress toward a purchase.
