Ecommerce ops · Production

Target's GRAM model uses LLMs to boost Home accessory recommendation conversions by 12%

The problem

Providing high-quality accessory recommendations across Target's vast catalog was infeasible manually because of the sheer volume of items and the many product attributes that must be considered for each pairing.

Workflow diagram · grounded in source
1
LLM attribute importance weighting
ai_action
“using LLMs to automatically analyze the product data, identify the most important attributes, and assign them importance weights for various pairs of core (or seed) and accessory item types”
2
LLM aesthetic matching
ai_action
“the LLM uses attributes such as color, material, and style to create harmonious sets of attribute values that enable more diverse and creative recommendations”
3
Accessory item scoring and ranking
ai_action
“The model formulates scoring rules using LLMs and uses them to compute scores for all accessory items. When a core item's attribute value matches an accessory item's attribute value, the attribute weight is added to the accessory item re…”
4
Merchant HITL collaboration
human_review
“we collaborated with site merchants to create a list of the most commonly co-purchased accessory items, enabling support for cross-category recommendations”
5
Add-to-cart recommendations output
output
“we added the Home Accessory model to the add-to-cart flyout”
Reported outcome

A/B testing of the Home Accessory model in the add-to-cart flyout in February 2025 showed around an 11% increase in interaction rate, roughly a 12% increase in display-to-conversion rates, and more than 9% growth in attributable demand; the model was fully rolled out to production in April 2025.

Reported metrics
Interaction ratearound 11% increase
Display-to-conversion rateroughly 12% increase
Attributable demandmore than 9% growth
Reported stack
LLMsGRAM
Source
https://tech.target.com/blog/accessory-recommendations-with-llms
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

A/B testing of the Home Accessory model in the add-to-cart flyout in February 2025 showed around an 11% increase in interaction rate, roughly a 12% increase in display-to-conversion rates, and more than 9% growth in a…

What tools did this team use?

LLMs, GRAM.

What results were reported?

Interaction rate: around 11% increase; Display-to-conversion rate: roughly 12% increase; Attributable demand: more than 9% growth (source-reported, not independently verified).

How is this ecommerce ops AI workflow structured?

LLM attribute importance weighting → LLM aesthetic matching → Accessory item scoring and ranking → Merchant HITL collaboration → Add-to-cart recommendations output.

Related ecommerce ops cases