AI-Powered Discount Approval Enforcement
Sales teams need to approve discounts while maintaining margin guardrails and preventing unauthorized offers that violate pricing policies.
Le problème
When a sales rep or AI agent proposes a discount, Corules evaluates it against customer tier limits, margin floors, and product category rules before the CRM record is written. Non-compliant discounts are blocked immediately. Edge cases escalate to the deal desk with full context.
Comment Corules le résout
BLOCK: Discount of 35% exceeds max_discount_by_tier['standard'] = 25%.
Exemple de politique
// Discount approval policy (CEL)
discount_pct <= params.max_discount_by_tier[context.customer_tier]
&& (context.deal_value * (1 - discount_pct)) >= params.margin_floor
&& context.product_category in params.discountable_categoriesOptions d'intégration
Frequently Asked Questions
Can the policy distinguish between customer tiers?
Yes. CEL expressions reference `params.*` for tenant-specific values like tier-based discount caps, so each customer segment can have different limits without changing the policy logic.
What happens when a rep requests a discount above the limit?
The decision is BLOCKED before it reaches the CRM. The system returns the specific rule violated and the allowed discount range for that customer tier.
Is every discount decision logged?
Yes. Each evaluation produces an audit entry with policy version, input hash, actor identity, and outcome. Audit entries are replayable.
How does this integrate with Salesforce?
A Salesforce Flow calls the Corules Policy API via an HTTP callout before any Opportunity update. The flow only proceeds on ALLOW; it routes to an approval process on ESCALATE.