Building an Interactive Excel Dashboard For E-commerce for Product Analysis: A Case Study for Jumia Products

Introduction

E-commerce platforms generate large amounts of product information that can help explain pricing strategies, customer engagement, and product performance.

For this project, I analyzed a dataset of products from Jumia, an e-commerce marketplace, using Microsoft Excel. The goal was to transform raw product data into a cleaned dataset, perform descriptive and relationship analysis, and develop an interactive dashboard that communicates the most important findings.
The analysis focused on four major areas:

• Product pricing

• Discounts

• Customer reviews

• Product rating
  The final result was an Excel-based analytical dashboard designed to help sellers identify high-performing products, products requiring attention, and potential pricing and marketing strategies.

  Project Objectives

  The goal of this project is to create an interactive Excel dashboard that provides insights into the
  performance of products listed on Jumia and is capable of answering the following questions:

• What is the average product price?

• What is the average discount?

• Which products have the highest ratings?

• Which products have the highest number of reviews?

• Do higher discounts result in higher customer engagement?

• Are highly rated products more expensive?

• Which products are performing well?

• Which products may require improved pricing or product strategies?

• What recommendations can be made to Jumia sellers?

  Dataset Description

  The dataset contains information about Jumia products
  using the following fields:
  Product:product name
  Current Price:Current selling price
  Old Price:Original/listed price
  Discount:Percentage discount
  Review:Number of customer reviews
  Rating:Customer rating out of 5
  The original dataset contained 115 records.
  After data cleaning and duplicate removal, the final dataset contained 112 products

  Raw Data

  The first step was to inspect the raw dataset before making any changes.

The initial audit looked for:

• Missing values
• Duplicate records
• Incorrect data types
• Negative review counts
• Invalid ratings
• Invalid discounts
• Price inconsistencies

Data Cleaning

Data cleaning was important because raw e-commerce data is not always immediately suitable for analysis.
For my data cleaning, I used Power Query, which made the cleaning process much easier.

Removing Duplicate Records

Duplicate records can cause products to be counted more than once and can distort totals and averages.
The dataset contained three duplicate records. These duplicates were removed, leaving 112 unique records.

Cleaning Review Counts

Some review values were negative.
A review count cannot logically be negative, so these values were converted to their absolute values.
Blank review values were retained as missing values rather than automatically being treated as zero.
This distinction is important because a missing review count does not necessarily mean that the product received zero reviews.

Cleaning Ratings

Ratings were originally stored in text formats such as:
4.5 out of 5
These values were converted into numerical ratings so that averages, sorting, and correlations could be calculated.

Cleaning Discounts

Discount values were originally represented as percentages, such as:
38%
These were converted into numerical percentage values for analysis.

Cleaning Prices

Currency text was removed from the price fields so that prices could be treated as numerical values.
One product contained a price range rather than a single price. A midpoint was used for the numerical analysis so that the record could be included in price calculations.
Also, we removed the 'KSh' and ',' before changing prices to whole numbers.

Data Enrichment

After cleaning the original fields, additional analytical columns were created.

Discount Amount

The discount amount was calculated using:

Discount Amount = Old Price - Current Price

This provides the actual monetary reduction rather than only the percentage reduction.

Rating Category

Products were categorized according to their rating performance.
The categories were used to make it easier to analyze the overall distribution of customer ratings.

Discount Category

Products were also grouped into discount categories:

• Low
• Medium
• High These categories made it possible to compare product performance across different discount levels.

Price Category

Products were grouped into price ranges to make price analysis easier and to support interactive dashboard filtering.

Excel Techniques Used

Several Excel features were used throughout the project.

Excel Formulas

Formulas were used to create calculated fields and perform summary calculations.
For example, the discount amount was calculated by subtracting the current price from the old price.
Functions such as AVERAGE, SUM, COUNT, MAX, MIN, IF, and other conditional functions were useful for transforming and analyzing the dataset.

Sorting and Filtering

Sorting was used to identify:

• Highest-rated products
• Lowest-rated products
• Products with the most reviews
• Products with the highest discounts
• Products requiring attention

Pivot Tables

Pivot Tables were used to summarize product performance by categories such as rating and discount levels.
They provided a flexible way to aggregate the data without manually calculating every summary.

Pivot Charts

Pivot Charts converted the Pivot Table results into visualizations suitable for the dashboard.

Slicers

Slicers were added to make the dashboard interactive.
Users can filter the dashboard based on variables such as:

• Rating Category
• Discount Category
• Price Category

Data Analysis

1 Descriptive Analysis

After cleaning the data, the following statistics were obtained:

Trend Analysis

Relationship Between Discounts and Reviews

One of the main questions was whether larger discounts result in greater customer engagement.
Customer engagement was represented using the number of reviews.
The correlation between discount percentage and review count was approximately:
-0.14
This represents a very weak negative relationship.
Therefore, the analysis does not provide evidence that increasing the discount percentage automatically results in more customer reviews.
Some products with substantial discounts had very few reviews, while some products with moderate discounts had significantly more reviews.

Business Interpretation

Sellers should therefore avoid treating discounts as the only method of increasing engagement.
Product quality, value, visibility, presentation, and customer experience may also influence customer response.

Relationship Between Price and Rating

The correlation between current price and rating was approximately:
+0.11
This is a very weak positive relationship.
The result suggests that expensive products were not necessarily rated much higher than cheaper products.
For example, the Electronic Digital Display Vernier Caliper had a relatively low current price but achieved a rating of 4.6/5 alongside 49 reviews.

Business Interpretation

Price alone does not appear to determine customer satisfaction.
Sellers should focus on providing strong value for money rather than assuming that a higher selling price will result in higher ratings.

Relationship Between Ratings and Reviews

The correlation between rating and review count was approximately:
+0.06
This is effectively a negligible linear relationship.
A product can therefore have many reviews without necessarily having a high rating.
This is particularly important when evaluating product performance.
A high number of reviews indicates strong customer engagement, but it does not automatically indicate customer satisfaction.

Sales Performance Analysis

The following table summarizes the product performance analysis by:

• Strong demand proxy: Which products have review counts above the engagement threshold?
• Pricing/marketing attention: Which products combine weak engagement or weak ratings with high price/high discount?
• Many reviews, average experience: Which products have high engagement but fall in the Average rating category? -Promotion inefficiency: Which high-discount products remain below the engagement threshold?

Building the Dashboard

The final stage of the project was creating an interactive Excel dashboard.
The dashboard was designed to provide a quick overview of product performance without requiring the user to inspect the raw dataset.

KPI Cards

The dashboard includes key performance indicators such as:

• Total Products
• Average Current Price
• Average Discount
• Average Rating
• Total Reviews ##Product Performance Charts Charts were created to display:
• Top 10 products by rating
• Top 10 products by reviews
• Top 10 products by discount ###Category Analysis The dashboard also presents the distribution of:
• Rating categories -Discount categories -Price categories ###Interactive Slicers Slicers allow the user to filter the dashboard by selected product characteristics.

Screenshot: Pivot Tables

Insert screenshot of your Pivot Tables here.

Screenshot: Charts

Insert screenshot of your charts here.

Screenshot: Final Dashboard

Insert screenshot of your completed Dashboard here.

Key Business Insights

The analysis produced five major conclusions.

1. Large discounts do not guarantee engagement

The discount-review correlation was approximately -0.14, showing a very weak negative relationship.

2. Higher prices do not guarantee better ratings

The price-rating correlation was approximately +0.11.

3. Review volume and rating measure different aspects of

performance
Review count reflects customer engagement, while rating provides an indication of customer satisfaction.

4. High-performing products combine engagement and satisfaction

Products such as the Cake Decorating Tool Set and Vernier Caliper demonstrated strong performance through both review volume and rating.

5. Some heavily discounted products still perform poorly

High discounts cannot compensate for poor product quality or customer dissatisfaction.

Recommendations for Jumia Sellers

Based on the findings, sellers should consider the following strategies.

1. Do not rely entirely on discounts

Discounting should be part of a broader marketing strategy rather than the only method used to generate engagement.

2. Monitor customer ratings carefully

A product receiving many reviews but maintaining a low rating should be investigated immediately.

3. Prioritize high-rated and highly reviewed products

Products demonstrating both strong engagement and customer satisfaction should receive greater promotional attention.

4. Improve product quality

Products with consistently poor ratings should be reviewed for quality and functionality issues.

5. Improve product descriptions

Accurate product descriptions and specifications can help reduce differences between customer expectations and the actual product.

6. Use competitive pricing

The analysis indicates that higher prices do not automatically result in higher ratings. Sellers should therefore balance profitability with customer-perceived value.

7. Use data for decision-making

Instead of deciding on pricing and promotions based solely on intuition, sellers can use review counts, ratings, discounts, and prices together to identify products requiring action.

Conclusion

This project demonstrated how Microsoft Excel can be used to transform raw e-commerce data into an interactive business intelligence dashboard.
The analysis showed that product performance cannot be explained by discounts or prices alone.
The weak relationship between discounts and reviews indicates that larger discounts do not automatically produce higher engagement. Similarly, the weak relationship between price and ratings indicates that expensive products are not necessarily better rated.
The most useful approach is to evaluate multiple indicators together.
Products with high review counts and strong ratings represent stronger performers, while products with high engagement but poor ratings may require quality or customer-experience improvements.
Ultimately, the dashboard provides sellers with a practical way to identify opportunities, monitor product performance, and make more evidence-based pricing and marketing decisions.

Final Takeaway

The key lesson from this project is simple:
A large discount can attract attention, but customer satisfaction and product value are essential for sustainable product performance.