customer_support · ecommerce · workflow

IPSY's AI agent delivers 943% ROI in four months with Ada's ACX platform

IPSY's scripted chatbot was insufficient as customer expectations rose and efficiency became a priority. Common issues like missing and damaged items required human intervention, and members who asked to cancel were simply allowed to leave with no retention effort.

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 Glam Bot
IPSY members contact Glam Bot, which was promoted from a scripted chatbot to an AI agent powered by Ada's ACX platform.
Tools used
AdaPlaybooksGlam Bot
Outcome

Within four months of deploying Ada's AI agent, IPSY achieved a 41% improvement in CSAT, a 63% improvement in Automated Resolution rate, and a 943% return on AI investment through containing ~160K conversations.

Results
Volume943%
Running sinceMay 2024
Source

https://www.ada.cx/case-study/ipsy

How we source this →

Grounding & classification
Source type: vendor customer story
37 fields verified against source quotes.
agentic workflowai agentconversational aisupport agentchat transcriptsupport ticketfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedtools describedvendor confirmedworkflow describedecommerceretailautomation ratecost reductioncustomer satisfactiondeflection rateemployee productivityvendor customer storycustomer supportticket triageautonomous resolutionescalation workflow