Building an Interactive Excel Dashboard for E-commerce Product Analysis
A Case Study of Jumia Products When managing an e-commerce retail operation, offering product discounts may seem like the easiest way to attract customer attention. But does a larger discount actually translate into greater customer engagement? To explore this question, I built an end-to-end Excel analytics pipeline using a dataset of 109 Jumia products . The project took me from messy raw data…
A Case Study on Jumia's E-commerce Products A journalist built an end-to-end Excel analytics pipeline using a dataset of 109 Jumia products to investigate the relationship between product discounts and customer engagement. Initially, the data contained numerous issues such as missing prices, unreadable ratings, duplicate records, and negative review values.
The journalist cleaned the data, removing duplicates, fixing formatting problems, and creating new, analytical fields. While creating rating categories, the journalist discovered a logic trap in Excel's IFS function that led to misclassification of products. After correcting the formula, the analyst created calculated fields for discount amounts, discount percentages, and price categories.
The discount-and-engagement question was addressed by examining the correlation between discount percentage and review count, which turned out to be extremely close to zero. This indicated little evidence of a linear relationship between discount size and customer reviews. However, a noticeable pattern emerged: products with medium discounts (20%-40%) had the highest average review count, while high-discount products had the lowest.
Additionally, a single product, a 120W cordless vacuum cleaner, stood out with 69 reviews and a 2.8 rating, demonstrating that review volume doesn't always correlate with product ratings. The journalist then compiled the findings into an interactive Excel dashboard featuring KPI cards, price-category analysis, discount-category analysis, and rating categories.
The dashboard included interactive slicers for dynamic filtering by price, discount, and rating. From this analysis, it was clear that larger discounts didn't necessarily lead to more customer reviews. The medium discount category performed best in terms of reviews, and high-review products could still have relatively low ratings.
Price and discount segmentation proved to be valuable tools for exploring product behavior, while review count should be viewed as an engagement indicator rather than a direct measure of sales.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
