Introduction

Data has become one of the most valuable tools for understanding customers, products, and business performance. Platforms such as Jumia generate large volumes of product data, including prices, discounts, customer ratings, and reviews.
This project focuses on analysing Jumia product data using Microsoft Excel to understand product performance from different perspectives. The analysis explores key factors such as product pricing, discount levels, customer ratings, and review volumes to identify trends and relationships within the dataset.
The objective of the analysis was to answer important business questions: Which price ranges dominate the marketplace? Do higher discounts translate into better customer ratings? Which products attract the most customer engagement? And what can customer ratings and reviews tell us about product performance?
Using data cleaning techniques, calculated fields, pivot tables, charts, slicers, and dashboard visualizations, the raw Jumia dataset was transformed into meaningful and actionable insights. The analysis demonstrates how Excel can be used not only for reporting and calculations, but also as a practical business analytics tool for turning raw data into information that can guide pricing, promotional, and product management decisions.

Dataset Description

The dataset contained six fields per product:
Product — the product name
Current Price — the current selling price, in KSh
Old Price — the price before any discount
Discount — the discount percentage applied
Review — the number of customer reviews
Rating — the average customer rating, out of 5
At a glance, the raw data had a few problems typical of scraped or exported e-commerce data: prices stored as text with currency symbols and commas, ratings embedded in strings like "4.5 out of 5" instead of numbers, and some review counts stored as negative values. None of this was usable for analysis until it was cleaned.

Data Cleaning and Preparation

Before i conducted my analysis, the raw dataset was carefully cleaned and prepared to ensure that the results would be accurate, consistent, and reliable. Since the data was collected in a format intended primarily for online display, several fields required transformation before they could be used for analysis.
The cleaning process involved the following key steps:

• Converting prices to numeric values . The Current Price and Old Price fields contained currency symbols, commas, and other non-numeric characters. These were removed to create clean numerical fields that could be used for calculations and comparisons.
• Extracting product ratings .Ratings were originally recorded as text in formats such as "4.5 out of 5". The numerical rating was extracted and converted into a format that could be analysed statistically.
• Correcting review values .Some review counts appeared as negative values in the raw dataset. These sign errors were identified and corrected so that the review figures accurately represented customer engagement.
• Handling missing values. Each column was checked for missing or incomplete information. A significant finding was that 55 out of 112 products, approximately 49%, had no rating provided. Instead of removing these records and losing potentially useful information about their prices, discounts, and reviews, the missing ratings were retained and classified as "Not Provided" in the rating category. This ensured that missing information was treated transparently rather than incorrectly influencing the analysis.
• Checking for duplicates and inconsistencies .The dataset was reviewed for duplicate products, near-duplicate entries, and inconsistent formatting. This helped improve the overall quality and consistency of the dataset before proceeding to the analytical stage. Through these cleaning and preparation steps, the raw Jumia dataset was transformed into a more structured and analysis-ready dataset. This provided a reliable foundation for the subsequent analysis of pricing, discounts, customer reviews, and product ratings.

Data Enrichment: Creating New Calculated Columns

After cleaning and standardizing the raw dataset, the next step was to enrich the data by creating additional calculated columns. These derived fields made it possible to segment the products and perform deeper analysis of pricing, discounts, and customer ratings.

The following calculated columns were created:

1. Discount Amount

The Discount Amount was calculated by finding the difference between the original price and the current selling price:
Discount Amount = Old Price − Current Price
=(D2-B2)
This provided the actual monetary value saved by a customer on each product, complementing the percentage discount already available in the dataset.

2. Rating Category

To make product ratings easier to compare, ratings were grouped into three categories:

• Poor — rating below 3
• Average — rating from 3 to below 4.5
• Excellent — rating of 4.5 and above The Excel formula used was:

=IF(ISBLANK(I2),"Not provided",IF(I2>=4.5,"excellent",IF(I2>=4,"good",IF(I2>=3,"avarage","poor"))))

Products without a rating were left blank so that missing information could be distinguished from products that actually received a low rating.

3. Discount Category

Products were also segmented according to the percentage discount applied:

• Low Discount — below 20%
• Medium Discount — 20% to 40%
• High Discount — above 40%

The following Excel formula was used:
=IF(E2<20%,"low",IF(E2>=40%,"above","medium"))
This categorization made it easier to investigate whether different discount levels were associated with differences in customer ratings and product performance.

4. Price Category

To analyse products based on their selling prices, the products were divided into Low, Medium, and High Price categories. The categories were determined using price quartiles, with approximately KSh 493 and KSh 1,670 serving as the breakpoints.
The Excel formula used was:
=IF(B2<990,"low",IF(B2>=2500,"high","medium"))
This classification enabled comparisons across different price segments and helped identify where most of the products were concentrated.

Excel Techniques and Analysis

With the data cleaned, standardized, and enriched with calculated fields, I moved to the analysis stage. The objective was to transform the prepared dataset into meaningful information that could reveal patterns in product pricing, discounts, customer engagement, and ratings.
To achieve this, I used a combination of Excel formulas, PivotTables, and Pivot Charts. These tools made it possible to calculate key performance indicators, compare product categories, identify trends, and summarize large amounts of information in a simple and interpretable format.

Excel Functions Used
Several Excel functions were applied during the analysis, including:
AVERAGE— to calculate the mean price, discount, and customer rating.
MAX — to identify the highest values, such as the most expensive product or highest discount.
MIN — to identify the lowest values within the dataset.
CORREL — to examine relationships between variables, such as price, discount, reviews, and ratings.
COUNTIFS — to count products that met specific conditions or belonged to particular categories.
These formulas provided the numerical foundation for the analysis, while PivotTables and Pivot Charts helped present the results in a more structured and visual manner

The full workbook which has raw data, cleaned data, all analysis sheets, Pivot Tables, charts, and the final interactive dashboard, is available in my GitHub repository linked below.
https://github.com/ndungujanerose/JANEROSE-JUMIA-PROJECT-ONE