How AI Agents are Evolving Shopify's Product Taxonomy at Scale (2025) - Shopify

Summary

Shopify’s product classification system processes over tens of millions of predictions daily, while its taxonomy contains more than 10,000 categories and 2,000 attributes that must adapt as commerce changes. Manual curation created bottlenecks, expertise gaps, and inconsistencies that could reduce discoverability, search quality, filtering, and classification quality. The proposed AI multi-agent system combines structural analysis with product-driven analysis grounded in merchant titles, descriptions, and categories, then uses intelligent synthesis, equivalence detection, and quality assurance to recommend taxonomy improvements. Agents can search and validate hierarchical relationships, while human experts retain oversight; Shopify tested the method in Electronics > Communications > Telephony against its previous manual expansion approach and presents it as a proactive, scalable way to improve taxonomy consistency before gaps affect merchants and customers.

Context

Shopify’s taxonomy must support more than 10,000 categories and 2,000 attributes while commerce evolves. Manual updates could not keep pace with new products, shifting merchant needs, and emerging attribute sets, and the growing taxonomy developed expertise gaps and inconsistencies in naming, relationships, and categorization.

Approach / What changed

The system uses specialized AI agents for structural analysis and product-driven analysis grounded in merchant data. An intelligent synthesis stage combines their findings, while agents search the taxonomy, inspect hierarchical relationships, validate proposed changes, detect complex equivalence relationships, and apply quality assurance. Human experts remain involved in oversight and strategic decisions.

Takeaways

• Structural analysis finds gaps, naming inconsistencies, and organizational opportunities within the taxonomy itself, while product-driven analysis examines merchant titles, descriptions, and categories.
• The system detects equivalences in which a specific category corresponds to a broader category filtered by attribute values, allowing merchants flexibility in catalog organization.
• Shopify applied the method to Electronics > Communications > Telephony and compared it with the previous manual expansion approach as a proof of concept.

Related reading