Competition Synergy Analysis

A comprehensive analysis framework for understanding competitive dynamics and strategic positioning in the hospitality industry, with a focus on hotels in Riva Del Garda. Hotel names are anonymized.

Overview

This repository contains two complementary analytical approaches:

1. Competition-Synergy Analysis: Understanding competitive relationships between hotels
2. Strategic Positioning Analysis: Mapping optimal positioning strategies for revenue optimization

Project Structure

Competition_Synergy/
├── data/
│   └── Riva_del_garda_hotels.csv     # Hotel dataset
├── output/
│   ├── competition_synergy/           # Competition analysis outputs
│   └── positioning/                   # Strategic positioning outputs
├── competition_synergy.ipynb          # Competition-synergy analysis
├── strategic_positioning.ipynb       # Strategic positioning analysis
└── README.md                         # This file

Analyses

1. Competition-Synergy Analysis (competition_synergy.ipynb)

Purpose: Identify competitive relationships between hotels to understand market dynamics.

Methodology:

• Data preprocessing with StandardScaler normalization
• Spearman correlation analysis between hotels
• Visualization of competition-synergy matrices

Key Outputs:

• Competition-Synergy Matrix: Heatmap showing relationships between all hotels
• Mean Competition-Synergy: Bar chart of mean competitive value
• Hotel Comparison: Example of a Side-by-side analysis of selected hotels

Insights:

• Positive correlations indicate synergy (complementary positioning)
• Negative correlations suggest direct competition
• Identifies key features driving competitive dynamics

2. Strategic Positioning Analysis (strategic_positioning.ipynb)

Purpose: Map optimal strategic positions for revenue maximization using dimensionality reduction and predictive modeling.

Methodology:

• Principal Component Analysis (PCA) for dimensionality reduction (95% variance retained)
• RandomForest regression for revenue prediction
• Gaussian smoothing for strategic positioning heatmaps

Key Outputs:

• PCA Scatter Plot: Hotel positions in reduced strategic space (latent space)
• Revenue Heatmap: Predicted revenue across strategic positioning space
• Component Analysis: Feature contributions to strategic dimensions

Insights:

• Identifies high-revenue strategic positions ("green zones")
• Shows how hotels differentiate in strategic space
• Reveals which features drive strategic positioning

Key Features

• Standardized Data Processing: Ensures fair comparison across different metrics
• Interactive Visualizations: Plotly-based charts for better exploration
• Predictive Modeling: Revenue optimization through strategic positioning
• Component Interpretation: Understanding what drives strategic differences

Requirements

pandas
numpy
plotly
scikit-learn
scipy

Usage

1. Competition Analysis:

   jupyter notebook competition_synergy.ipynb

2. Strategic Positioning:

   jupyter notebook strategic_positioning.ipynb

Data Requirements

The analysis expects a CSV file with hotel features including:

• Various hotel characteristics
• Revenue data (tot_revenue_euro) for strategic positioning
• Hotel identifiers as index

Outputs

All visualizations are saved in the respective output folders:

• PNG images for static analysis
• HTML files for interactive exploration

Applications

• Hotel Management: Understand competitive positioning and identify optimization opportunities
• Market Analysis: Assess competitive landscape and market gaps
• Strategic Planning: Data-driven positioning decisions for revenue growth
• Investment Decisions: Identify high-potential strategic positions

Methodology Notes

• Spearman Correlation: Used for robust relationship detection regardless of data distribution
• PCA: Reduces complexity while preserving 95% of variance
• Random Forest: Captures non-linear relationships for revenue prediction
• Gaussian Smoothing: Creates interpretable strategic positioning zones

Future Enhancements

• Multi-city comparative analysis
• Temporal analysis of positioning changes
• Clustering analysis for market segmentation
• Real-time competitive monitoring