A national flag-carrier airline
Refund turnaround cut from about 14 days to about four hours.
A refund at this airline used to involve multiple validations, policy checks, backend interactions and cross-team hand-offs, and took about two weeks. Agentforce went live on a class of refund cases first and, eighteen months on, has been extended to email resolution, name changes and a knowledge agent for service teams.
At a glance
- Client
- A national flag-carrier airline
- Engagement
- Operational agents on an existing Service Cloud estate
- Based in
- India
The stack
01
The situation
What they were working against.
Service processes such as refunds ran through multiple validations, policy checks, backend systems and teams, and took about fourteen days end to end. As the airline scaled it needed those workflows resolved faster and more consistently, with governance and human oversight intact.
02
What was built
The build, in the order it happened.
Refunds first
Agentforce live on a defined class of refund cases, on top of an existing Service Cloud, Sales Cloud, Data 360 and Einstein estate.
Automated email resolution
One email with several requests in it: the agent identifies each, validates data across systems, takes the action and drafts a consolidated reply from approved templates.
Name-change automation
Eligibility checks, validations, ticket reissue and sending the updated ticket, taking three days to thirty minutes.
A knowledge agent for staff
Conversational access to policies, procedures, fare rules and operational guidance for service teams.
03
The results
From assisting the team to acting within guardrails.
Refund turnaround fell from approximately fourteen days to about four hours and passenger name updates from three days to thirty minutes: faster resolutions, greater consistency and lower customer effort, with more than thirty agentic initiatives under way.
- Refund turnaround time, approximately
- 14 d → 4 h
- Passenger name updates
- 3 d → 30 min
- Agentic initiatives under way across a 140-plus-system estate
- 30+
Client name withheld. Figures are as reported for this implementation; where one was a stated target rather than a measurement, the label says so.
04
Our read
How we would build this at your scale.
What we take from this build, and what the same outcome looks like for a team of fifty to five hundred people rather than a global enterprise.
One case class, then the next
The airline did not automate refunds; it automated a defined class of refund cases, then widened. That sequencing is how we scope every operational agent, with the exclusions written down.
Agentforce & AI 02The validations are the integration work
Checking eligibility across booking, payment and policy systems is where the four hours actually come from. For a regional carrier or a hotel group, that is a MuleSoft or platform-integration design with idempotent actions.
MuleSoft Integration 03Governance is an operating rhythm
Guardrails, human oversight and approved templates drift unless someone owns them monthly. That is what a managed retainer is for once the agent is live.
Managed ServicesMore case studies
A North American paperboard and folding-carton manufacturer
A customer portal where packaging buyers track orders, shipments and build requests themselves.
ReadA regulated consumer-products manufacturer selling wholesale to retailers
Salesforce and QuickBooks Desktop kept in step every hour, with nothing re-keyed between them.
ReadThe U.S. wealth-management arm of a global bank
Client onboarding cut from weeks to 24 minutes, then meeting prep handed to an agent.
ReadWant the same result, without the enterprise bill?
Thirty minutes with an architect. Bring the part of this story that looks like your problem and we will tell you what it takes to get there: which products, how many weeks, and what we would leave out.