AI customer support case study · Azure OpenAI · Travel and vacations

An AI customer support assistant that answers from approved content

A travel and vacations company in North America issues travel certificates, and its support team kept answering the same questions about registration, bookings, deposits and travel information. We built a web chat assistant on ASP.NET Core and Azure OpenAI that answers certificate holders from the company's own approved content and its certificate data. It's live in production. When a question needs a person, the assistant creates an agent request for the support team.

Agent run · Customer supportRunning
Ask
Understand
Check
Answer
Handoff
Certificate holder asks about a booking and deposit
Topic
Booking and deposit
Certificate
Checked through the client's API
Source
Approved FAQ and Terms & Conditions
Next step
Answer, or agent request if a person is needed
Sa
Agent request for the support teamSupport agent
ApproveHold
Nothing posts until a person approvesIllustrative
Before and after

What changed for the support team.

BeforeWith the assistant
Repetitive questionsCertificate, registration, booking and deposit questions all reached the support teamThe assistant answers them, so fewer repetitive questions reach agents
Certificate detailsAgents checked certificate information by handThe assistant checks certificate and business context through the client's API
Finding answersStaff searched for content manuallyStaff search and review conversations in an admin portal
Content changesUpdating answers leaned on developersThe support team manages FAQs and supporting content without developers
Hard casesAgents were spread across simple and complex questionsAgents focus on cases that need a human
How it works

Questions answered from approved sources only.

The assistant understands what the customer means, checks their certificate, and answers only from content the client has approved. If it can't, a person takes over.

1 · What comes in
Customer chat question
Approved FAQs and Terms & Conditions
Certificate and inventory data
2 · Web chat assistant
Azure OpenAI intent understanding
Semantic search over approved content
Certificate-aware answers: status, registration, booking, next steps
PDF and document processing for document-based answers
3 · Agent requests
Question that needs human judgment
Request the assistant can't resolve from approved content
Support team
4 · Posted in
ASP.NET Core and SignalR
SQL Server and Entity Framework Core
Client's certificate API
Every conversation, feedback and answer source is logged to the admin dashboard.
Results

Real production counts, no inflated claims.

1,526 conversations

During the measured production period, the assistant handled 1,526 customer conversations from 1,225 distinct certificate holders.

2,474 of 2,556 answered from approved content

That's 96.8% of questions answered using the client's FAQs, Terms & Conditions, inventory, certificate data and knowledge content.

1,366 questions after hours

53% of questions arrived outside the 9-5 Pacific, Monday-Friday support window, including 394 weekend questions.

Why it worked

Grounded answers and a person behind them.

Approved content only

The assistant answers from FAQs, Terms & Conditions, knowledge documents, inventory and certificate data the client approved.

Clear handoff

When a question needs a person, the assistant creates an agent request, so the customer always has a next step.

The team stays in control

The support team curates the FAQs and content, reviews conversations, and spots recurring issues from dashboards on activity, satisfaction and answer sources.

FAQ

About this project.

Is this AI customer support assistant live?

Yes. It's in production, and the results above are measured production counts.

Where do the assistant's answers come from?

From approved content only: FAQs, Terms & Conditions, knowledge documents, inventory information, certificate data and approved documents. During the measured period, 2,474 of 2,556 questions were answered this way.

What happens when the assistant can't answer?

It creates an agent request so the support team picks the question up and the customer isn't left without a next step. The support team also curates the content the assistant answers from.

Did response speed, satisfaction or ticket volume improve?

We don't claim that.

Your support team

Want an assistant that answers from your content?

Tell us which questions keep landing on your support team. We'll show you how an assistant grounded in your approved content, with people handling the hard cases, could work.

Book Appointment
Sahil kataria (1)
Sahil Kataria

Founder and CEO

amit Kumar
Amit Kumar

Chief Sales Officer

Talk To Sales

USA

+1 270-550-1166

flag

+1 270-550-1166

Phil J.
Phil J.Head of Engineering & Technology​
QServices Inc. undertakes every project with a high degree of professionalism. Their communication style is unmatched and they are always available to resolve issues or just discuss the project.​
Assured

Thank You

Thanks, we’ve got your details. A named member of our team will reply within one business day.