Microsoft Travel and Hospitality Global

Holland America Line

Customer Service Automation

Holland America Line wanted to provide even faster, more personalized support for customers—without driving up contact center volumes or running a traditional chatbot.

Multi-intent detection: Ability to understand queries such as “Help me find a cruise to Alaska for 6 to 8 days from Seattle.” Custom entity extraction: Ability to expand on predefined entities such as dates and add custom entities (i.e., “I want to go next summer” will generate a JSON object for the June–August summer season {"departurefromdate": "2025-06-01","departuretodate": "2025-08-01"}) Single-page reasoning: Ability to reason over a particular cruise itinerary page to answer questions like “Which day are we getting to the Bahamas and what are we seeing before that?” “With Microsoft Copilot Studio, we have an effective platform for delivering the benefits of generative AI to our customers, providing them with faster service and an even better overall cruise experience.” Scot Pettit, Sr.

Reported impact

Business outcome

Telemetry is also indicating that Anna will reduce the number of basic informational queries to contact center agents. Significant increase in bookings Reduced contact center volume “With so many stateroom options and packages, booking a cruise can be a complicated process, and once you’re booked, it’s not always easy to understand how to prepare for a cruise and add additional services,” says Scot Pettit, Sr.

  • Customer Experience
  • Revenue Growth
  • Speed to Market

Solution

Holland America Line used Microsoft Copilot Studio, Microsoft Dynamics 365, Dataverse to support customer and agent interactions with AI-assisted automation. Holland America Line wanted to provide even faster, more personalized support for customers—without driving up contact center volumes or running a traditional chatbot. Key implementation elements included Microsoft Copilot Studio builds the digital concierge or voice agent that handles customer requests across channels; Microsoft Dynamics 365 provides the common business process layer that standardizes operations; Dataverse stores structured app data, conversation logs, and feedback for the service; Azure Monitor tracks platform health and operational telemetry. Customer questions enter the assistant from web, voice, or messaging channels; retrieval services ground responses with website and product content; the ai layer interprets intent and generates responses or actions; interaction logs and feedback are stored for tuning;...

Data flow

Customer questions enter the assistant from web, voice, or messaging channels; retrieval services ground responses with website and product content; the ai layer interprets intent and generates responses or actions; interaction logs and feedback are stored for tuning; answers or next steps are returned to customers, agents, or clinicians through the front-end experience.

Solution architecture

6 components · 4 layers
  1. Storage
    • Dataverse Stores structured app data, conversation logs, and feedback for the service.
  2. Serving
    • Microsoft Copilot Studio Builds the digital concierge or voice agent that handles customer requests across channels.
    • Bing Search APIs Supplies website content used to ground responses.
  3. Orchestration
    • Microsoft Dynamics 365 Provides the common business process layer that standardizes operations.
  4. Uncategorized
    • Azure Monitor Tracks platform health and operational telemetry.
    • Application Insights Captures application diagnostics and usage signals for tuning.

Architecture clues

  • Operates with real-time or near-real-time telemetry and analytics.

Evidence from the source

Holland America Line wanted to provide even faster, more personalized support for customers—without driving up contact center volumes or running a traditional chatbot.
Multi-intent detection: Ability to understand queries such as “Help me find a cruise to Alaska for 6 to 8 days from Seattle.” Custom entity extraction: Ability to expand on predefined entities such as dates and add custom entities (i.e., “I want to go next summer” will generate a JSON object for the June–August summer season {"departurefromdate": "2025-06-01","departuretodate": "2025-08-01"}) Single-page reasoning: Ability to reason over a particular cruise itinerary page to answer questions like “Which day are we getting to the Bahamas and what are we seeing before that?” “With Microsoft Copilot Studio, we have an effective platform for delivering the benefits of generative AI to our customers, providing them with faster service and an even better overall cruise experience.” Scot Pettit, Sr.
Telemetry is also indicating that Anna will reduce the number of basic informational queries to contact center agents. Significant increase in bookings Reduced contact center volume “With so many stateroom options and packages, booking a cruise can be a complicated process, and once you’re booked, it’s not always easy to understand how to prepare for a cruise and add additional services,” says Scot Pettit, Sr.
The company developed a virtual agent using Microsoft Copilot Studio that acts as a digital concierge on its website to support new and existing customers and travel advisors.
This is what led the company to Microsoft Copilot Studio and the development of a virtual “agent”—an advanced, natural language conversational experience that uses AI to support the guest’s digital journey.
Transparency Evidence & methodology 90/100 reference value · 92% confidence · 2 sources

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value90/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact15/25
    1 metric describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence92%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability23/25
    11 of 12 key fields link to a named source.
  • Direct evidence20/20
    5 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency9/10
    2 source references document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Microsoft

    Holland America Line customer story

    First-party
    https://www.microsoft.com/en/customers/story/19787-holland-america-dataverse
    Source type
    Vendor-published customer story
    Published
    Apr 25, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date
  2. LinkedIn

    Holland America Line | LinkedIn

    Independent secondary
    https://www.linkedin.com/company/holland-america-line/

    The public company page's Organization JSON-LD includes numberOfEmployees value 9064; that maps to Enterprise.

    Source type
    Structured company profile
    Supports
    • Company size

Date evidence

  • Published on source page Apr 25, 2025 · Microsoft · First-party
    Open date source
Not separately sourced Industry, Region. Treat these normalized attributes as lower-confidence until a dedicated source is added.