customer_support · ecommerce · workflow
Cotopaxi achieves 28% deflection lift and $76K savings while maintaining 4.5/5 CSAT with Forethought AI
Cotopaxi's high-touch customer experience team was wary of AI, fearing it would lower their care standards, but needed a way to scale support to meet customer demand for speed and personalization without adding headcount.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Customer contacts AI agent
Customers submit questions to Forethought's automated agents.
Tools used
ForethoughtForethought SolveForethought DiscoverKnowledge Gap tool
Outcome
Over their first six months with Forethought, deflection rates rose 28% with savings of $76,000, CSAT was maintained at 4.5 out of 5, and headcount stayed the same as the company grew.
What failed first
Cotopaxi's previous chatbot lacked the ability to automatically route customer questions through specific intents, which was the primary limitation that prompted their switch.
Results
Volume28%
Cost replaced$76,000
Grounding & classification
Source type: vendor customer story
35 fields verified against source quotes.
content generationconversational aiknowledge searchsupport agentknowledge basesupport ticketfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedsource backedtools describedvendor confirmedworkflow describedecommerceretailcost reductioncustomer satisfactiondeflection rateemployee productivityvendor customer storycustomer supportticket triageai draft human approvalautonomous resolutionextract classify route