Data CloudIntermediateinsights-builderaffinityretail
Configure Insights Builder for product affinity scoring
Real World Scenario
Merchandising wants product affinity scores on profiles for email recommendations.
Expected Answer
• Define affinity from co-purchase or browse-to-buy patterns
• Set rolling window avoiding outdated seasonal skew
• Normalize scores 0-1 for cross-category comparison
• Exclude returned purchases from affinity calculation
• Refresh weekly balancing freshness vs compute cost
• Validate top affinities manually for brand safety
Follow-Up Questions & Answers
Click to expand — each follow-up includes a direct, interview-ready answer
Direct answer: Define affinity from co-purchase or browse-to-buy patterns Also consider: Set rolling window avoiding outdated seasonal skew In practice: Normalize scores 0-1 for cross-category comparison Balance speed of delivery with maintainability.
Architect Perspective
Affinity without returns adjustment recommends products customers already sent back.