Introduction
For my first data analytics project, I have had the chance to work with a Jumia products dataset and used Excel to turn the raw data into something that could actually be used to understand product performance.
The main goal of the project was not just to clean it and make a pretty dashboard. I wanted to use the data to answer some simple questions about product pricing, discounts, ratings and reviews.
some of the questions answered within the project were:
- Do products with bigger discounts get more reviews?
- Do highly rated products get more engagements?
- Is there a relationship between price and rating?
- Which products have the highest rating and reviews?
- which products might need better pricing or marketing strategies? This project also gave me a chance to practice some of the basic data analytics skill I have been learning, especially data cleaning, Excel formulas, PivotTables, Charts and Dashboard creation
THE DATASET
The dataset contains products information from Jumia. The main columns I worked with were:
- Product
- Current price
- Old price
- Discount
- Review
- Rating The data was not completely ready for analysis, so I first had to inspect it and make some changes before building the dashboard
One important thing I kept in mind is that reviews are being used as a measure of engagement, not as actual sales data. A product having more reviews does not automatically mean its more products.
The Raw Data
I started by creating a Raw_Data sheet.
I kept this sheet as the original version of the dataset so that I could always go back to the starting point if something went wrong during cleaning.
While checking the data, I noticed some issues such as duplicate records, price stored with currency text, price range and some missing reviews and rating values.
I also found some products that appeared more than once with exactly the same information.
Cleaning the Data
After checkinh the raw dat, I created a Cleaned_Data sheet.
The main cleaning steps include:
- Removing exact duplicate records. -Working with the price fields so the be used for calculations
- Making sure discount values could be used in calculation
- Checking rating and review values
- Keeping missing values blank instead of automatically treating them as zero
I also added some calculated fields that would help with the analysis.
One of them was Discount Amount,which shows the difference between the old price and current price.
For example:
=old price-Current Price
I also used categories to make some of the analysis easier to understand, such as rating categories, discount categories and price categories.
Using Excel for the Analysis
once the data was prepared, i moved on to the analysis section.
I calculated some of the basics statistics such as:
- Total number of products
- Average current price
- Average old price
- Average discount
- Average rating
- Total reviews
- Highest product price
- Lowest product price I also used correlation to look at the relationship between some of the variables. which I looked at: Discount vs Reviews Rating vs Reviews Current Price vs Rating
The purpose was to see whether there was a noticeable relationship between these variables.
However, correlation doesn't mean that one variable caused another. It only helps show whether two variables tend to move together.
PivotTables and Charts
After the calculation, I created PivoTables to make it easier to summarize the data.
Some of the PivotTables looked at:
- Rating categories
- Discount categories
- Price categories
- Reviews by discount category
- Top products by rating
- Top products by reviews
I then used these results to create charts.
Building the Dashboard
The final step was putting the important information into or dashboard.
I included KPI sections for things like:
- Total Products
- Average Price
- Average Discount
- Average Rating
- Total Reviews
I also included charts for the different relationships I investigated and tables showing some of top-performing products.
I added slicers so that the dashboard could be filtered by categories such as Rating category, Discount Category and price category.
Conclusion
Overall, this project was a good way for me to practice the basic data analytics process using Excel.
I started with raw Jumia product data, Cleaned and prepared it, analyzed the information, created PivotTables and finally build an interactive dashboard.
This project also showed me that even a relatively simple dataset can provide useful insights when it is properly when it is properly organized and analyzed.
This is one of my first steps in building my data analytics portfolio, and I plan to continue improving my skills by working on more projects using Excel, SQL, Python and other data analytics tools.
My GitHub project
The complete project, including the dataset, Excel dashboard and project documentation, is available in my GitHub repository. [https://github.com/gitosh-star/JUMIA_PRODUCTS_PERFOMANCE_DASHBOARD]
Top comments (0)