Every project here shows the problem, the data used, the process, and most importantly the measurable result. Dashboards and analysis samples are shown live below.
A dataset covering buyer behavior across three regions had no visual layer. Stakeholders could not quickly identify which regions, age groups, or income brackets were most likely to purchase bikes. Cross-referencing manually took hours and produced inconsistent answers every time.
Cleaned and restructured the dataset, built pivot tables segmenting buyers by age, gender, region, income, and car ownership, then designed an interactive Excel dashboard with dynamic slicers. Every chart updates automatically based on filter selection.
Bike buyers dataset with variables including region (Europe, North America, Pacific), age category, gender, income level, number of cars owned, commute distance, and purchase decision. Full data cleaning performed before analysis.
Stakeholders could instantly filter by any demographic combination and identify buyer profiles that previously took hours to assemble manually. The dashboard surfaced an income-to-purchase correlation and regional differences that were completely invisible in raw tabular form.
Buyers with zero cars are 3.7x more likely to purchase a bike than those with 3 or more. North America had the highest purchase rate at 92%, with adults aged 25 to 45 making up 61% of all purchases.
Sales data across five Canadian provinces was spread across multiple disconnected sheets with no consolidated view. Identifying revenue by segment, category, or manufacturer required manually navigating separate tabs, taking hours and producing inconsistent results.
Built a consolidated multi-sheet model using INDEX MATCH for dynamic data retrieval. Created pivot charts covering six revenue dimensions and added interactive province and month filters across the entire model.
Product sales dataset across Alberta, British Columbia, Manitoba, Ontario, and Quebec. Variables: revenue by month, product category (Mix, Rural, Urban, Youth), market segment, and manufacturer.
What previously required navigating multiple disconnected sheets was consolidated into one interactive dashboard with live filtering. Ontario emerged as the top province at $3.8M. The All Seasons segment dominated across all categories. These findings were completely invisible in the raw multi-sheet format.
An e-commerce business needed to extract monthly revenue performance, top-selling categories, customer repeat rates, and regional distribution from their transactional database. Manually pulling from spreadsheets was producing errors and taking hours each month.
Wrote structured SQL queries to extract, aggregate, and join data across four tables. Delivered a clean monthly report with five KPIs, a top-10 product ranking, and customer segmentation by order frequency.
| Product Name | Category | Total Revenue | Units Sold | Revenue Rank |
|---|---|---|---|---|
| Pro Running Shoes X2 | Footwear | $142,800 | 1,428 | 1 |
| ErgoPro Office Chair | Furniture | $118,500 | 395 | 2 |
| WirelessPro Headset | Electronics | $97,200 | 810 | 3 |
| CarbonTrack Bicycle | Sports | $88,400 | 221 | 4 |
| Organic Cotton Tees | Apparel | $76,100 | 3,805 | 5 |
Replaced a manual, error-prone monthly process with structured SQL queries that run in seconds. The five KPIs (total revenue, average order value, unique customers, repeat rate, and top products) were delivered as a clean report that required zero manual calculation.
Public agricultural data across Nigerian states exists but is fragmented, inconsistently formatted, and entirely unvisualized. Seasonal performance gaps and regional production differences are not surfaced in any form that supports farming decisions or policy work.
A Power BI dashboard mapping crop yield trends across Nigerian states using public agricultural datasets. Designed to surface seasonal gaps, regional production patterns, and year-on-year changes in a format decision-makers can act on directly.
This project is currently in development. Full case study and dashboard sample will be published on completion.