Aerofit is a prominent brand in the fitness equipment industry, offering a diverse range of products including treadmills, exercise bikes, gym equipment, and fitness accessories. The brand caters to all fitness levels, enabling individuals to achieve their fitness goals through well-designed products.
Aerofit’s market research team seeks to enhance their understanding of the customer base for their treadmill products. The goal is to analyze customer characteristics to improve product recommendations for potential buyers.
- Develop customer profiles for each treadmill model.
- Use descriptive analytics, including visualizations (tables and charts), to uncover key insights.
- Build two-way contingency tables to calculate conditional and marginal probabilities, assessing the impact of customer demographics on treadmill purchases.
The dataset comprises information on customers who purchased a treadmill from Aerofit in the last three months. The key features of the dataset include:
- Product Purchased: KP281 (Entry-level), KP481 (Mid-level), KP781 (Advanced)
- Age: Customer’s age in years
- Gender: Male/Female
- Education: Number of years of education
- Marital Status: Single/Partnered
- Usage: Weekly treadmill usage (average number of sessions)
- Income: Annual income in USD
- Fitness: Self-rated fitness level on a scale of 1-5 (1 = poor, 5 = excellent)
- Miles: Expected number of miles the customer plans to walk/run weekly
- KP281: Entry-level treadmill ($1,500)
- KP481: Mid-level treadmill ($1,750)
- KP781: Advanced treadmill ($2,500)
- Data Exploration: Initial data analysis and cleaning.
- Descriptive Analytics: Visualizing data to identify trends and patterns in customer demographics.
- Probability Analysis: Using contingency tables to compute conditional and marginal probabilities, assessing the likelihood of treadmill purchases based on customer characteristics.
- Insights & Recommendations: Based on the analysis, propose strategies for improving product recommendations for new customers.
The full analysis is available on Google Colab. View the Notebook.
A detailed analysis report is available in the following PDF file: View Report.
The Python code and analysis are available in the following Jupyter Notebook: View Notebook.