Customer Service Automation
Engine needed a voice-support experience that could answer traveler questions, manage bookings, and handle hotel cancellations at scale without sounding robotic or forcing callers to repeat themselves to human agents.
Reported impact
Engine scaled voice support with a more human-sounding agent, rolled the system out to most inbound calls, and simplified rep work by handing off completed voice transcripts and summaries instead of forcing customers to restart the conversation.
Engine launched a customer-facing voice-support workflow on Salesforce with Agentforce Voice at the front end and Agentforce for Service behind the scenes so high-volume travel calls can be handled conversationally. Because Engine's stack was already described as 'wall-to-wall Salesforce,' the project could use native CRM, customer-record, and service context instead of bolting a separate voice layer onto fragmented systems. NeuraFlash, Engine, and Salesforce iterated rapidly before and after general availability, testing live conversations, refining prompts and voice settings, and rolling traffic out in phases from 10% of calls to 50% and then 75%. When the AI cannot finish the task, the transfer includes a completed transcript and summary inside Agentforce for Service, which shortens handoffs, preserves traveler context, and lets reps focus on exceptions instead of re-collecting information.
Incoming traveler calls are answered by Agentforce Voice, which uses CRM and Data 360 context to answer questions or complete tasks; if the workflow needs a human, the call is transferred into Agentforce for Service with a full transcript and AI summary already attached.
It's natively integrated with Salesforce CRM, Data 360, and customer records.
They were looking for a conversational voice agent that could answer traveler questions, manage bookings, and handle hotel cancellations end to end.
We started by routing about 10% of calls to Agentforce Voice ... Then we increased usage ... Now we're at 75% of incoming calls.
When a case gets transferred, reps see a fully filled out voice transcript and summary in Agentforce for Service.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Headquarters Denver, Colorado
The Organization JSON-LD lists addressLocality 'Denver', and addressCountry 'US', and numberOfEmployees value 1194.
Related
Salesforce Sample Oct 9, 2025 Heathrow Airport Providing smooth, connected experiences requires innovation and access to real-time insights . As one of the world’s busiest and most iconic airports, Heathrow Airport welcomes over 83 million passengers annually. 40 % improvement in digital contact efficiency Agentforce · Data 360 · Service Cloud Salesforce Sample Apr 14, 2025 Boat Bike Tours Boat Bike Tours relied on fully manual quote building and service follow-up even as guests expected immediate answers, itinerary flexibility, and personalized travel planning across a growing operation. 15,000 quotes annually Agentforce · Data Cloud · Experience Cloud Salesforce Sample Apr 9, 2024 Air India Air India was handling 550,000 incoming service cases a month across a patchwork of airline systems created by mergers, leaving agents without a complete customer view and making it hard to deliver consistent support. 550,000 incoming service cases each month Data Cloud · Einstein 1 Service · Salesforce Platform 