Table of Contents

• Introduction
• Project Overview
• Data Processing
  • Data Collection and Cleaning
  • Data Transformation
• Exploratory Data Analysis (EDA)
• Strategic Recommendations
• Tools and Technologies
• Challenges and Lessons Learned
• Conclusion

Onyx Data 2025 Challenge: Lee Chatmen Merchandise Sales Analysis

Introduction

This project showcases my data analysis and visualization skills using Power BI to analyze merchandise sales data for TikTok influencer Lee Chatmen (7M+ followers). The goal was to provide actionable insights to inform strategic business decisions for his merchandise line launched in 2023.

Project Overview

I obtained raw sales data from Onyx Data and performed a comprehensive analysis, from data cleaning and transformation to exploratory data analysis (EDA) and strategic recommendations. The project highlights my ability to translate complex data into clear, concise, and actionable insights for both technical and non-technical stakeholders.

Data Processing

Data Collection and Cleaning

The raw sales data required extensive cleaning to ensure accuracy. This involved:

• Identifying and correcting inconsistencies.
• Handling missing values.
• Removing duplicate entries.

Data Transformation

Following data cleaning, I transformed the data for optimal analysis and visualization. This included:

• Encoding categorical variables.
• Scaling numerical features.
• Engineering new variables to uncover deeper insights.

Exploratory Data Analysis (EDA)

Using Power BI, I conducted a thorough EDA to identify key trends and patterns. Key findings include:

• Sales Trends: May was the highest-performing month (80,704 units), a 24.25% increase compared to August (the lowest-performing month). (Line chart showing sales over time here)
• Top-Selling Category: Clothing dominated sales (637,201 units). (Bar chart showing sales by category)
• Geographical Distribution: Domestic sales comprised 69.5% of orders, while international sales accounted for 30.5%. (Pie chart visualizing this distribution)
• Customer Demographics: Male buyers represented 70.2% of the customer base, and female buyers 29.8%. (Bar chart or donut chart showing gender distribution)

Strategic Recommendations

Based on the EDA findings, I developed the following strategic recommendations:

• Seasonal Promotions: Leverage high-sales months like May by aligning marketing campaigns to further boost sales.
• Inventory Management: Optimize stock levels and pricing strategies for clothing, the top-performing product category.
• International Expansion: Explore strategies to capitalize on the existing international customer base and expand global reach.
• Targeted Marketing: Develop targeted campaigns tailored to the predominantly male demographic to improve engagement and conversion rates.

Tools and Technologies

• Power BI: Used for data cleaning, transformation, analysis, and visualization.

Challenges and Lessons Learned

• Challenge: Initially, the sales data contained numerous inconsistencies and missing values, requiring significant effort in the data cleaning phase.
• Lesson Learned: This experience reinforced the importance of thorough data cleaning and the impact it has on the reliability of subsequent analysis. I developed a more systematic approach to data cleaning, including documenting all cleaning steps and implementing data validation checks.
• Challenge: Visualizing the geographical distribution of sales proved tricky due to inconsistencies in address data.
• Lesson Learned: I learned to leverage Power BI's mapping capabilities and data transformation tools to standardize location data and create accurate geographic visualizations.

Conclusion

This project demonstrates my proficiency in data analysis and visualization using Power BI. I successfully transformed raw data into actionable insights, providing a clear path for strategic decision-making. My ability to present data effectively caters to both technical and non-technical audiences, ensuring that insights are understood and implemented.