Jumia Product Performance Analysis & Pricing Dashboard

Introduction

As part of understanding how Microsoft Excel is used in E-commerce. I analyzed a dataset of products listed on Jumia to understand how product performance from pricing to discount affect sales and customer engagement. The first step was to clean and prepare the raw dataset. Then perform data analysis to create visualizations, and later developed an interactive dashboard that presented the findings.
These analysis aim to provide useful insights that can support better pricing strategies, promotional decisions, and customer engagement for Jumia.

Cleaning the Data

Before cleaning I copied the worksheet to a new sheet for reference purposes or incase something goes wrong or I accidentally delete data.

• Missing Values- I checked every column for empty cells. Which were plenty in the ratings column. I wroteNot available in the blanks spaces for accurate output during analysis.
• ** Duplicates**- Searched and removed duplicates and fixed other inconsistencies.
• Numeric transformation-Fixed the review column data as it was stored as negative numbers. Also removed the Ksh and commas in the Current price and old price columns to numeric values. Lastly was removing the out of 5in the ratings column. After cleaning this is what was displayed:

The new calculated columns

At this point I added the new columns which are:

• Discount Amount-Discount Amount = Old Price − Current Price.=C2-B2
• Rating Category- =IF(G2="","",IF(G2<3,"Poor",IF(G2<=4,"Average",IF(G2<=4.5,"Good","Excellent"))))

• Discount Category-=IF(D2="","",IF(D2<20%,"Low Discount",IF(D2<=40%,"Medium Discount","High Discount")))
  This means below 20%-Low, 20%-40%=Medium and above 40% is high

• Price Category- =IF(B2<1000,"Low Price",IF(B2<=2000,"Medium Price","High Price"))
  This means that Below Kes 1000=Low price, Kes 1000-2000= Medium Price and above Kes 3000.

Data Analysis

For the KPIs I used the following formulas

Total Products-=COUNTA('Excel_jumia_dataset'!A2:A113)
Average Current Price-=AVERAGE('Excel_jumia_dataset'!B2:B113)
Average Old Price-=AVERAGE('Excel_jumia_dataset'!C2:C113)
Average Discount-=AVERAGE('Excel_jumia_dataset'!E2:E113)
Average Rating-=AVERAGE('Excel_jumia_dataset'!I2:I113)
Total Reviews-=SUM('Excel_jumia_dataset'!H2:H113)
Most Expensive Product-=XLOOKUP(B9,'Excel_jumia_dataset'!B2:B113,'Excel_jumia_dataset'!A2:A113)
Least Expensive Product-=XLOOKUP(B11,'Excel_jumia_dataset'!B2:B113,'Excel_jumia_dataset'!A2:A113)
This results to:

Dashboard creation process

Putting everything together to create a dashboard I used
KPIs-The screenshot above to be specificthat shows total products, average price, average discount, average rating, and total reviews.
Pivot charts- I created multiple pivot charts to showcase the top 10 products by discount, rating and reviews.
Slicers- To connect all PivotTables/charts for Rating Category, Discount Category, and Price Category, so the whole dashboard can be filtered interactively.

Correlation Analysis

Used CORREL() to test three relationships:
Discount vs Reviews--0.1368, This means that higher discounts don't influence higher customer engagement.
Rating vs Reviews-0.0572, So a product having a high rating doesn't necessarily mean it has lots of reviews.
Current Price vs Rating-0.1080, Expensive products are not clearly rated higher than cheaper products.
Key Insights and Business Recommendations

• Prioritize highly reviewed products when evaluating customer engagement because review volume varies substantially across products.
• Monitor products with low ratings and identify opportunities to improve product quality, descriptions, or customer experience.
• Focus promotional efforts on products where discounts can attract attention without compromising product quality.
• Use the dashboard's slicers to compare product performance across different price, discount, and rating categories.
• Investigate products with high discounts but poor ratings, as large discounts may not compensate for poor customer satisfaction.

Conclusion

The analysis showed that discounts, ratings, and customer reviews did not have strong relationships in this dataset. In particular, the weak correlation between discount percentage and reviews suggests that offering larger discounts does not necessarily result in higher customer engagement. The dashboard also made it easier to identify highly reviewed products, highly discounted products, and products that may require attention due to low ratings. The interactive dashboard brings these findings together through KPI cards, charts, PivotTables, and slicers, allowing users to explore product performance across different rating, discount, and price categories.
Overall, this project strengthened my understanding of Excel for data cleaning, analysis, visualization, and dashboard development, while also demonstrating how data-driven insights can support better e-commerce business decisions.
You can find the README.md and workbook that contains the clean dataset, pivot charts, pivot tables, Excel formulas, slicers and the dashboard on my GitHub link below:

https://github.com/naomionderi-codes/Jumia-Product-Performance-Dashboard