I'm a Data Analyst based in Nairobi, Kenya — I turn messy, real-world data into clear findings and actionable business decisions.
My work spans the full analytics pipeline: cleaning raw datasets, writing SQL queries, building star-schema databases, running exploratory analysis in Python, and publishing interactive dashboards in Power BI. I care about making data understandable — not just to analysts, but to the business people who act on it.
| Category | Tools |
|---|---|
| Languages | Python (pandas, NumPy, matplotlib, seaborn), SQL (T-SQL) |
| Visualization | Power BI (DAX, interactive dashboards), Tableau, Excel charts |
| Databases | SQL Server (star schema design, CTEs, joins, aggregations) |
| Analysis | EDA, hypothesis testing, customer segmentation, KPI tracking |
| Workflow | Git & GitHub, Jupyter Notebooks, Google Workspace, MS Office 365 |
Analysed 10 years of Central Bank of Kenya mobile money data + 7 waves of FinAccess
Household Survey data (2006–2024) across Kenya's 47 counties.
Key findings: M-Pesa transaction value grew ~7× over the decade · Tana River County had
the highest financial exclusion rate (37.6%) · People with no formal education were 4.6×
more likely to be excluded than the national average.
Python SQL Server Power BI CBK Open Data FinAccess Survey
Built a complete pipeline across Python, SQL Server, and Power BI covering 5 datasets.
Designed a star-schema database, wrote exploratory SQL queries, and published 3 interactive
dashboards on store performance, product profitability, and customer behaviour.
Key finding: A small segment of stores and loyalty members drove the majority of revenue.
Python pandas SQL Server Power BI Star Schema
Analysed ~7,000 customer records across 21 features to identify the key drivers of churn.
Built a 5-task Python pipeline for data cleaning, EDA, and visualisation using pandas,
matplotlib, and seaborn.
Python pandas matplotlib seaborn EDA
Merged and cleaned two international match datasets, compared scoring patterns and win rates,
then applied statistical hypothesis testing to determine whether goal-average differences
were statistically significant.
Python pandas scipy hypothesis testing matplotlib
- 🏅 Deloitte Australia — Data Analytics Job Simulation (Forage, April 2026)
Forensic technology simulation · Tableau dashboard · Excel data classification - 📚 DataCamp — Data Analyst Track (In Progress)
Python · SQL · Power BI · Statistical Analysis
- 🌐 Portfolio: peter-ngamau.github.io
- 💼 LinkedIn: linkedin.com/in/peter-ngamau
- 📧 Email: ptahmwangi@gmail.com