customer_support · saas · workflow
Otter Assistant: LLM-powered in-house support agent autonomously handles ~50% of customer requests
Otter's broad restaurant software feature set created high customer support demand requiring deep system integration, but existing vendors offered only hard-coded decision trees without the integration flexibility Otter needed.
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 request triggers workflow
A customer request triggers the bot to look up the corresponding predefined procedure.
Tools used
LLMvector dbStreamlitRAG
Outcome
Otter Assistant autonomously handles approximately 50% of inbound customer requests without human intervention and without compromising customer satisfaction, while also exposing previously undetected product and platform issues.
What failed first
Established vendors relied on hard-coded decision trees without a clear LLM strategy, and LLM-native startups lacked the ability to handle the complexity of Otter's resolution workflows.
Results
Volume~50%
Running sinceQ2 2024
Grounding & classification
Source type: technical build writeup
36 fields verified against source quotes, 1 dropped as unverifiable.
agentic workflowchatbotconversational aiknowledge searchragsummarizationsupport agentknowledge basesupport ticketbuilder submittedfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedhospitalitysoftwareautomation ratecustomer satisfactiondeflection ratetechnical build writeupcustomer supportticket triageautonomous resolutionescalation workflowrag answering