Challenge

Rhone initially partnered with Stylitics to bring their curated in-store merchandising experience to their e-commerce platform. While AI-powered outfitting delivered exceptional results—including a 39% increase in AOV—Rhone recognized a further opportunity to optimize the foundation of their e-commerce engine: product metadata.

The core challenge was ensuring premium products remained visible in a competitive SEM landscape. Existing metadata lacked the granular, shopper-centric language required to match evolving search intent, potentially limiting the reach of their high-performance catalog.

Key Objectives:

1. Bridge the Intent Gap: Align product metadata with the actual natural language used by shoppers in search engines.
   
2. Scale Without Overhead: Find an enterprise-grade solution to enrich thousands of SKUs without the manual burden of internal tagging or experimental R&D projects.
   
3. Prove Incremental Lift: Quantify the direct performance impact of enrichment on Google Shopping metrics through a rigorous, isolated test.

Solution

Stylitics implemented a controlled A/B test to evaluate the performance lift generated by enriched metadata delivered via Rhone’s Google Merchant Center (GMC) instance. By choosing a proven, production-ready platform over an internal “build” project, Rhone bypassed the 12–18 month learning curve typical of in-house AI experiments.

Test Methodology:

• Randomized Product Split: 96% of Rhone’s product catalog was split into test (50%) and control (50%) groups.
  
• Daily Enrichment: The test group received Stylitics’ enriched metadata while the Control group remained unchanged.
  
• Balanced Distribution: Groups were balanced by attribute distribution to eliminate bias and ensure a fair comparison.
  
• Unified Campaign Management: Both groups ran within a single GMC campaign using identical bidding strategies and budgets to isolate the impact of enrichment.

“Stylitics provides the production infrastructure we need to bridge the gap between our premium product catalog and actual shopper intent. Moving to AI Catalog Enrichment delivered a +16.3% lift in CTR and a +4.5% increase in ROAS on Google Shopping by aligning our metadata with how shoppers actually search. This is proven, scalable intelligence that ensures our products are discoverable at every touchpoint.” — Sally Pullman, Vice President of Ecomm & Digital, Rhone

Results

The test confirmed that enriched metadata results in a better match to shopper intent, driving higher-quality traffic and a significant lift in ROAS.

Performance & Efficiency Impact:

• Significant CTR Lift: Enriched listings produced higher quality clicks, driving a +16.3% increase in CTR despite lower overall impressions.
  
• Conversion Growth: High-intent traffic led to a +3.5% lift in CVR as well as down funnel volume growth with a +3.1% increase in total conversions and a +4.7% increase in Conversion Value.
  
• Improved ROAS: Efficiency gains led to a +4.5% increase in ROAS, proving that enrichment more than pays for itself through higher-quality shopper matching.
  
• AOV Lift: Downstream metrics showed a +1.5% increase in AOV, reflecting increased basket value.

Looking Ahead

Rhone’s expansion into catalog enrichment demonstrates the power of treating AI as a strategic, enterprise-grade investment rather than a one-off experiment. By transforming raw data into high-intent signals, Rhone has not only boosted immediate performance but has future-proofed its catalog for the next generation of AI-driven discovery.

Download the case study.