Target Improves Add to Cart Interactions by 11 Percent with Generative AI Recommendations

Target Improves Add to Cart Interactions with GRAM

Target deployed GRAM, a generative AI-powered accessory recommendation system for its Home category, utilizing large language models to prioritize product attributes and aesthetic cohesion. The system increased add-to-cart interactions by 11% through improved product pairing suggestions.

Why This Matters

Traditional rule-based recommendation systems struggle with the scale and complexity of large retail catalogs, often requiring extensive manual curation and failing to capture nuanced relationships between products. This leads to irrelevant recommendations, decreased user engagement, and lost revenue – estimated to cost retailers billions annually in missed sales opportunities.

Key Insights

• 11% increase in add-to-cart interactions: Demonstrated by A/B testing of the GRAM system.
• LLMs for attribute weighting: GRAM uses large language models to dynamically assess which product attributes are most important for accessory recommendations.
• Human-in-the-loop curation: Target combines AI with merchant input to ensure business relevance and maintain merchandising goals.

Practical Applications

• Use Case: Target uses GRAM to suggest visually harmonious pairings, like pillowcases that complement sheets, increasing conversion rates.
• Pitfall: Over-reliance on automated recommendations without human oversight can lead to irrelevant or inappropriate suggestions, damaging customer trust.

References:

• https://www.infoq.com/news/2025/12/target-genai-model-recommend/