Expand description
Recommendation models personalized upsells with inventory, preference, and care-safety gates before any customer copy is approved. Retail recommendation models for personalized upsell candidates, review gates, and safe customer copy.
Modules§
- customer_
copy - Customer-copy policy keeps retail recommendation text in draft/review state until approval and rejects unsafe claims.
- rationale
- Rationale text keeps the staff-readable evidence for why a retail recommendation exists.
Structs§
- Candidate
- Retail recommendation candidate containing customer/pet context, product, rationale, inventory, preference, and care-safety signals.
- Candidate
Builder - Use builder syntax to set the inputs and finish with
build(). - Policy
- Evaluates retail recommendation safety so staff see useful candidates without bypassing care, inventory, or preference gates.
Enums§
- Care
Sensitivity - Care-sensitivity status that decides whether supplement/diet/product recommendations need care or manager review.
- Decision
- Recommendation decision that either drafts an internal candidate, requires review, or suppresses the upsell.
- Preference
- Customer preference state used to suppress opted-out recommendations and review unknown consent.
- Reason
- Business reason for recommending a product, kept separate from customer-facing copy.
- Review
Reason - Review reasons that explain whether care facts, care-plan conflicts, or consent uncertainty paused the candidate.
- Rule
- Recommendation rule that names the operational event that can produce an upsell candidate.
- Suppression
Reason - Reason a recommendation is suppressed before staff/customer use.