customer_support · travel · workflow
Cebu Pacific replaces scripted chatbot with Ada's generative AI agent, achieving 34%+ higher automated resolution rate and 50%+ higher CSAT
Cebu Pacific's scripted chatbot Charlie lacked personalization, required heavy manual effort to maintain flows and integrations, and could only answer the latest question in a conversation without retaining context.
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 sends chat inquiry
Customers access live chat via Cebu Pacific's website, mobile app, or Facebook page.
Tools used
AdaAda Academy
Outcome
After deploying Ada's generative AI agent, Cebu Pacific achieved a 34%+ higher automated resolution rate and 50%+ higher CSAT scores, with wait times for high-priority disruption cases reduced to under one minute.
What failed first
The declarative scripted chatbot approach required constant manual effort — hardcoding flows, uploading content manually, and line-by-line data reviews — and could not understand conversation context or deliver personalized responses.
Grounding & classification
Source type: vendor customer story
23 fields verified against source quotes, 4 dropped as unverifiable.
ai agentconversational aiknowledge searchragsummarizationchat transcriptknowledge basefailure mode describedhuman review describednamed customerproduction runtime claimedsource backedtools describedworkflow describedtravelautomation ratecustomer satisfactionresolution time reductionvendor customer storycall center aicustomer supportautonomous resolutionescalation workflowintake to triage