Introduction
For this project, we are using Microsoft Excel to transform raw Jumia product data into meaningful, actionable insights.
The goal is to build an interactive dashboard that makes it easier to understand product pricing, discounts, promotions, sales performance, and customer engagement.
Through data cleaning, analysis, and visualization, the dashboard helps turn large amounts of raw data into clear information that can support better pricing decisions, more effective promotions, and a deeper understanding of customer behavior.
Dataset
Our dataset is defined as below:
The Product
The current Price
The Old Price
The discount
The review
The ratings
With any dataset given , you need to understand what the organization expects you to achieve even before proceeding with data cleaning.
Here the objectives stated , will help us achieve an end goal of understanding how the products perform and some recommendations that we will give the business to ensure the low performing products are will improve, or how to make their business grow.
Below are our objectives:
To understand:
- whether larger discounts are associated with more reviews.
- whether highly rated products attract stronger engagement.
- whether price and rating move together.
- which products perform best based on ratings and reviews; and
- which products may need a different pricing or marketing strategy.
Data Cleaning
From our data we see alot of inconsistencies as stated below:
Blanks
Price columns, Both Current and old price are defined as texts instead of numerical/ currencies.
Some Duplicate Products, with same values all through to the last column.
Misspelled Header:
Ratingdintsead ofRatings.Negative Reviews.
Data Cleaning Steps:
Renaming the misspelled Header
After Indentifying the issue Change it fromRatingdtoRatingsChanging the Prices from Texts to numerical then updating the currency.
-Select the whole column, then pressCTRL+Hrename theKshtoBlank. Then format the cell by choosing the correct currencies available , in our case we will convert it toKES
3.Changing the negative Reviews. CTRL+Hthen replace it with a Blank
Replacing blank cells with Values like:
NullorUnknown
CTRL+H, Replace Balnk withNullChanging the ratings, now that we know 5 is the highest out of all the rating. We remove the
out of 5so that we just remain with numerical ratings and for it to be aa numerical column.We have an item with prices not defined but left as shown.
With this we will have to calculate the median/Average because we dont know the exact price but we are given the range.
=AVERAGE(1620,1980)1800=AVERAGE( 2200, 3200)2700
- Removing only duplicates that are similar in all columns. After identifying the duplicates, Select the whole data>>data tab>>remove duplicates>> select all columns.
Creating Additional Columns.
Creating Additional columns easens the work when creating pivot tables which we will use for our analysis, As we are comparing the Price, Rating,Discounts and Reviews.
- Pricing Category:
- Discount Category:
Pivot Tables
From the cleaned data set, click any cell>>insert>>pivot tables>>New page.
Whether larger discounts are associated with more reviews.?
We add the row for discount Category and average Reviews to understand if its possible.
From our Analysis, Discount isn't associated with more Reviews.
Some items despite having discounts, don't even have high reveiws.
For Jumia , they shouldn't really depend on the discounts to give those products high/more reviews.
Whether highly rated products attract stronger engagement.
From Our Analysis , yes High rated products do attract high review, meaning these products are good and rated well.
whether price and rating move together.

From our Analysis Price and rating go hand in hand. It helps though but that shouldn't be a dependant. For Items with High ratings they do have hig prices. Which is a good strategy because for thos hight rated items, they make more revenue.
However Jumia should take note on poor rated products , either imporve the items/change suppliers for that specific item, inorder to get high ratings and maybe also increase price for those items.
which products perform best based on ratings and reviews
By Reviews:120W Cordless Vacuum Cleaners Handheld Electric Vacuum Cleaner
By Ratings:
Anti-skid Absorbent Insulation Foaster for Home Office
Bedroom Simple Floor Hanging clothes rack single pole Hat Rack White.
Classic Black Cat cottonHemp Pillow Case for Home Car.
which products may need a different pricing or marketing strategy.
5-PCS Stainless Steel Cooking Pot Set With Steamed Slices
Since this item has a poor /the least review, Jumia should atleast reduce the price , since currently it is a meduim category price to abit lower to attract customers or change startegy for marketing it to get more clients to use the products and gauge the ratings. If the ratings will not improve then change the product or supplier for that specific product.
Recommendations:
Look beyond discounts as an engagement strategy.
Prioritize quality control for products with many reviews but low ratings.
Review the products with both high discounts and low ratings. Instead of applying further discounts, sellers should investigate whether product quality, customer expectations, or listing accuracy is contributing to the poor ratings.
Encourage customers to leave reviews, particularly for high-selling products with few or no ratings.
Conclusion
This project reinforced an important lesson in data analysis: assumptions do not always hold up when tested against real data. The expectation that larger discounts automatically drive customer engagement, or that price is directly linked to product quality, was not necessarily supported by the data.
The data-cleaning process was equally important. Handling missing ratings appropriately rather than simply excluding or hiding them helped ensure that the analysis and correlation results remained as reliable and transparent as possible.
Ultimately, the dashboard transforms raw Jumia product data into insights that can help sellers make more informed decisions around pricing, promotions, product quality, and customer engagement.

















Top comments (0)