Business Intelligence | Data Analytics | Python · SQL · Power BI · Excel
I started out in finance and accounting, where I saw first-hand how better information leads to better decisions. That curiosity grew into a career in Business Intelligence: a Master's degree in the field, and today a role on international initiatives at Inter IKEA Group, where I work with data from multiple sources to help teams plan, prioritise, and make informed decisions.
One thing I have learned is that data rarely tells the whole story on its own. Understanding the business context and asking the right questions matter just as much as the analysis. The projects below reflect that: each one starts with a business question and ends with findings a stakeholder can act on.
jepchumbakangogo@gmail.com · Medium
Python · SQL (SQLite) · pandas · scipy · seaborn
End-to-end analysis of 3,231 Stockholm Airbnb listings to find what drives nightly prices.
- Entire homes command 2.6× the price of private rooms (2,570 vs 1,002 SEK)
- Södermalm and central districts top prices; the cheapest outer districts average less than half
- Welch's t-test showed Superhost status has no significant price premium (p = 0.09)
Power BI · DAX · Excel
Interactive dashboard advising a startup on marketing allocation across its 150 stores and 10 new expansion cities, combining store sales with city demographic data.
- Built custom DAX measures for revenue, marketing spend, and ROMI
- Found the 10 new expansion stores outperform the established base (+18% avg revenue, ROMI 15.8 vs 13.4)
- Flagged top markets (Glendale CA, Brownsville TX) and one underperformer needing review
Excel · Pivot Tables · Interactive Slicers
Analysis of 1,000 customer records to profile who buys bikes and why, presented in an interactive Excel dashboard. My first portfolio project, later revamped with improved chart design; the Medium write-up covers the original version and the repository shows both.
- Bike buyers have higher average incomes than non-buyers across both genders
- Purchase likelihood drops sharply beyond a 5-mile commute
- Middle-aged customers (31-54) are the most likely purchasers
| Area | Tools |
|---|---|
| Analysis | Python (pandas, scipy), SQL (SQLite), Excel |
| Visualisation | Power BI (DAX), matplotlib, seaborn, Excel dashboards |
| Statistics | Hypothesis testing, EDA, outlier handling |
| Business | Finance and accounting background, reporting, forecasting, stakeholder collaboration |
Always learning: new projects in SQL and Python are on the way. Feedback and collaboration welcome!