Customer support · Production

Cotopaxi achieves 28% deflection lift and $76K savings while maintaining 4.5/5 CSAT with Forethought AI

The problem

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.

First attempt

Cotopaxi's previous chatbot lacked the ability to automatically route customer questions through specific intents, which was the primary limitation that prompted their switch.

Workflow diagram · grounded in source
1
Customer contacts AI agent
trigger
“what their customers frequently ask automated agents”
2
AI defines customer intent
ai_action
“Forethought allowed them to clearly define customer intent before the AI kicked in, so they could be sure someone asking for information on a return, for example, was given the correct information”
3
Route by specific intent
routing
“the ability to route customer questions automatically through specific intents”
4
AI deflects high-volume questions
ai_action
“automated agents easily answer frequently repeated questions about sales, promo codes, orders, and returns”
5
Human agents handle complex requests
human_review
“their agents can spend their time on more complex issues that require a human's touch”
6
Knowledge Gap tool generates article drafts
ai_action
“Discover's Knowledge Gap tool finds ticket topics not covered in Cotopaxi's knowledge base and generates ready-to-edit articles to close those gaps”
7
Team edits and publishes drafts
human_review
“The team tweaks these drafts, activates them quickly, and gets them live quickly”
8
Ongoing knowledge base audit
feedback_loop
“regularly reviewing, adding, and cleaning up articles to keep them useful for customers”
Reported 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.

Reported metrics
Deflection rate28%
Cost savings$76,000
CSAT score4.5 out of 5
Headcount stability during growthkeep its headcount the same as the company grew
Reported stack
ForethoughtForethought SolveForethought DiscoverKnowledge Gap toolKustomerShopify
Source
https://forethought.ai/case-studies/cotopaxi-168-percent-roi-forethought
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

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 tools did this team use?

Forethought, Forethought Solve, Forethought Discover, Knowledge Gap tool, Kustomer, Shopify.

What results were reported?

Deflection rate: 28%; Cost savings: $76,000; CSAT score: 4.5 out of 5; Headcount stability during growth: keep its headcount the same as the company grew (source-reported, not independently verified).

What failed first in this deployment?

Cotopaxi's previous chatbot lacked the ability to automatically route customer questions through specific intents, which was the primary limitation that prompted their switch.

How is this customer support AI workflow structured?

Customer contacts AI agent → AI defines customer intent → Route by specific intent → AI deflects high-volume questions → Human agents handle complex requests → Knowledge Gap tool generates article drafts → Team edits and publishes drafts → Ongoing knowledge base audit.

Related customer support cases