Market & Category Demand Analysis: Unlocking Insights from Jumia Nigeria Data

As part of the MyOnlineShop Internship Program, I had the privilege of leading a comprehensive market and category demand analysis during Week 3. This project, completed by Tuesday, 29th July 2025, involved scraping and analyzing data from Jumia Nigeria to uncover trends in product popularity and customer engagement. Below, I walk through the step-by-step process, highlight the importance of each stage, and provide actionable insights and recommendations for future growth.

Project Objectives

The primary goal was to identify product categories with the highest demand potential on Jumia Nigeria, using metrics like price points, review volumes, and customer ratings. This analysis aimed to empower MyOnlineShop with data-driven insights to prioritize categories, optimize pricing strategies, and enhance customer satisfaction over the next 2-10 years.

Step-by-Step Process

Data Collection

The first step involved scraping data from jumia.com.ng using the Jumia Product Scraper Chrome extension. I targeted 100 products per category (excluding "Others"), capturing details such as product name, current price, original price, discount percentage, brand, rating, and review count. This process, guided by a YouTube tutorial, ensured a robust dataset. Importance: Accurate data collection is the foundation of any analysis. A diverse and representative dataset allowed us to reflect real customer behavior and market trends on Jumia Nigeria.

Data Cleaning and Structuring

After collecting data into individual CSV files (e.g., appliances.csv, electronics.csv), I used Python and the pandas library to combine them into a single dataframe. This involved handling missing values (e.g., unbranded products) and standardizing formats. Importance: Cleaning ensures data reliability, removing noise that could skew results. Structuring the data into a unified format enabled seamless analysis across categories.

Categorization and Grouping

I grouped products into categories like Appliances, Electronics, and Health & Beauty based on product names and keywords. This step required inferring categories where labels were ambiguous, documented with transparent assumptions. Importance: Proper categorization is critical for meaningful aggregation. It allowed us to compare performance across diverse product types and identify high-potential areas.

Metric Calculation

I calculated the average number of reviews and average rating per category. Using MinMaxScaler from scikit-learn, I normalized these metrics to create a weighted engagement score, combining review volume (popularity) and rating (satisfaction). Importance: Normalization levels the playing field across categories with different scales, while the engagement score provided a holistic view of customer interaction, guiding prioritization decisions.

Visualization and Dashboard Creation

The final output was a market demand dashboard featuring bar charts and heatmaps, built using Python. Key stats included an average price of 15 billion NGN (29% discount), 1100 products (100% coverage), an average rating of 4.18 (66% engagement), and 421.54 average reviews (19% engagement). Visuals highlighted categories like Electronics (1136 reviews) and Health & Beauty (1113 reviews) as leaders. Importance: Visualizations make complex data accessible, enabling stakeholders to quickly grasp insights and make informed decisions.

Key Insights

• Top Performers: Electronics and Health & Beauty led with high review volumes (1136 and 1113) and engagement scores (0.71 and 0.67), indicating strong customer interest and satisfaction.
• Underserved Categories: Phones & Tablets (156 products, 0.24 engagement) and Fashion (131 products, 0.24 engagement) showed lower popularity despite decent review counts, suggesting visibility or marketing gaps.
• Rating Distribution: The "Good" rating level dominated at 46.45%, with "Poor" at a minimal 0.91%, reflecting overall customer satisfaction but room for improvement in lower-rated products.

Recommendations for MyOnlineShop (2025-2035)

• Capitalize on Strengths: Leverage Electronics and Health & Beauty’s high engagement for targeted marketing campaigns and promotions, potentially increasing revenue by 15-20% over five years.
• Boost Underdogs: Enhance visibility for Phones & Tablets and Fashion through influencer partnerships and improved product listings, aiming to raise engagement scores by 0.2-0.3 within two years.
• Address Weaknesses: Focus on improving "Poor"-rated products with quality enhancements or better customer support, potentially reducing poor ratings by 50% in three years.
• Long-Term Innovation: Invest in AI-driven personalization and dynamic pricing over the next decade, using this dataset as a baseline to predict demand trends, potentially growing market share by 10-15% by 2035.

Conclusion

This project honed my skills in data cleaning, aggregation, and visualization while delivering actionable insights for MyOnlineShop. The experience of translating raw data into a compelling story has prepared me to contribute to data-driven decision-making in future roles. I’m excited to bring this expertise to a team where I can continue to uncover value from complex datasets.

Note: Data references include jumia.com.ng and the Jumia Product Scraper extension.