Hotel Personalization - Pick'N Stays

An executive intelligence platform for multi-property hotel portfolios built on Snowflake's AI Data Cloud.

Built By: Sri Subramanian, Senior Solution Engineer
GitHub | LinkedIn

Overview

This solution demonstrates how hospitality organizations can leverage Snowflake to build an end-to-end guest intelligence system combining:

• Medallion Architecture — Bronze/Silver/Gold data layers for 100 global properties
• Guest 360 Profiles — Comprehensive guest behavior, preferences, and lifetime value
• ML-Powered Personalization — Upsell propensity scoring for spa, dining, technology, and pool services
• Cortex Agents — 5 specialized AI agents for natural language querying via Snowflake Intelligence
• Real-time Dashboard — Streamlit-powered executive intelligence

Full Solution Guide: For detailed architecture, business value, use cases, and technical capabilities, see the Developer Guide.

What You Will Learn

• How to implement a Medallion Architecture (Bronze/Silver/Gold) for hospitality data
• How to build Guest 360 profiles with lifetime value and behavior analytics
• How to create Semantic Views for natural language SQL queries via Cortex Analyst
• How to build Cortex Agents that enable conversational analytics
• How to access your agents through Snowflake Intelligence
• How to build executive dashboards with Streamlit in Snowflake

What You Will Build

• A complete data pipeline transforming raw guest, booking, and service data through Bronze → Silver → Gold layers
• Guest 360 profiles with upsell propensity scores for spa, dining, technology, and pool services
• Semantic Views enabling natural language queries across portfolio, loyalty, and guest experience domains
• Five specialized Cortex Agents for guest analytics, personalization, amenities, experience optimization, and master intelligence
• A multi-page Streamlit dashboard for portfolio oversight, loyalty intelligence, and service quality monitoring

Prerequisites

• Snowflake Account with ACCOUNTADMIN role (sign up for free trial)
• Snowflake CLI (install guide)
• OS: macOS, Linux, or Windows (with WSL or Git Bash)

Verify installation:

snow --version

Setup Guide

Step 1: Install Snowflake CLI

pip install snowflake-cli

Or download installer from Snowflake CLI releases

Verify: snow --version

Step 2: Configure Snowflake Connection

snow connection add

Find your account identifier in the URL: https://<account-identifier>.snowflakecomputing.com

Step 3: Test Connection

snow connection test -c demo

Step 4: Download Repository

Option A: Clone with Git

git clone https://github.com/Snowflake-Labs/sfguide-hotel-personalization-picknstays.git

Navigate into the cloned folder:

cd sfguide-hotel-personalization-picknstays

Option B: Download ZIP

1. Go to GitHub Repository
2. Click Code → Download ZIP
3. Extract the ZIP file

Navigate to the extracted folder:

cd /path/to/sfguide-hotel-personalization-picknstays

Note: Replace /path/to/ with the actual location where you cloned or extracted the repository (e.g., ~/Downloads/ or ~/Desktop/). All subsequent commands should be run from inside this folder.

Step 5: Deploy

chmod +x deploy.sh
./deploy.sh -c demo

Deployment takes approximately 10-15 minutes to complete.

Step 6: Validate Deployment

./run.sh validate -c demo

Step 7: Access the Solution

Streamlit App:

• Snowflake UI → Projects → Streamlit → Hotel Personalization - PickNStays
• Role: HOTEL_PERSONALIZATION_ROLE

Cortex Agents (Snowflake Intelligence):

• Snowflake UI → Projects → Snowflake Intelligence → Select an agent
• Role: HOTEL_PERSONALIZATION_ROLE

What Gets Deployed

Database Schemas

• BRONZE — Raw data: guests, bookings, stays, amenity transactions, service cases
• SILVER — Cleaned & enriched: guest profiles, stay details, service metrics
• GOLD — Analytics-ready: Guest 360, personalization scores, portfolio KPIs
• SEMANTIC_VIEWS — Semantic views for Cortex Analyst
• STREAMLIT — Streamlit app deployments

Cortex Agents

• Hotel Guest Analytics Agent — GUEST_ANALYTICS_VIEW
• Hotel Personalization Specialist — PERSONALIZATION_INSIGHTS_VIEW
• Hotel Amenities Intelligence Agent — AMENITY_ANALYTICS_VIEW
• Guest Experience Optimizer — GUEST_ANALYTICS_VIEW, AMENITY_ANALYTICS_VIEW
• Hotel Intelligence Master Agent — All semantic views

Explore the Solution

Streamlit Dashboard

Cortex Agent Capabilities

All agents use Cortex Analyst (text-to-SQL via semantic views). Try these sample queries:

• "Show me our top 10 most valuable guests by lifetime revenue"
• "What are the booking patterns of our Diamond tier members?"
• "Which guests have the highest spa upsell propensity this week?"
• "Show me guests at medium or high churn risk who have high lifetime value"
• "Which loyalty tier has the highest average stay length?"

Tip: Use the Hotel Intelligence Master Agent for cross-domain queries spanning guests, personalization, amenities, portfolio, loyalty, and service data.

Cleanup

./clean.sh -c demo

Conclusion

You've successfully deployed an AI-powered hotel personalization platform that demonstrates how hospitality organizations can transform guest data into actionable intelligence. This solution showcases the power of combining Snowflake's Medallion Architecture with Cortex AI capabilities to deliver:

• Unified Guest Intelligence — A single source of truth for guest profiles, preferences, and behavior across all properties
• Conversational Analytics — Natural language access to complex data through Cortex Agents and Snowflake Intelligence
• Executive Visibility — Real-time dashboards for portfolio performance, loyalty optimization, and service quality monitoring

To extend this solution, consider integrating real-time data feeds from your property management systems, adding custom propensity models for your specific amenity offerings, or building additional Cortex Agents for specialized use cases like revenue management or marketing campaign optimization.

Additional Resources

• Snowflake CLI
• Cortex Analyst
• Cortex Agents
• Semantic Views
• Snowflake Intelligence
• Streamlit in Snowflake

License

Copyright (c) Snowflake Inc. All rights reserved.

Licensed under the Apache 2.0 license.