Hi, my name is Prince Uwagboe
I'm a Data Analyst.

To address the common problem of inaccessible business analytics tools for small enterprises, I developed a Business Dashboard using Python, Streamlit, and Facebook Prophet. My task was to turn a messy Superstore dataset into an intuitive, web-based platform for revenue insight and forecasting. I automated preprocessing steps including real-time currency conversion, ran EDA to identify patterns, and implemented Prophet for time-series forecasting. The dashboard displayed sales anomalies and generated 12-month sales projections. As a result, users gained access to a centralized interface that visualized $2.3M in revenue and a clear Q4 growth trend, enabling informed decisions around marketing focus and category optimization.

Recognizing the difficulty many users face in managing personal and business finances across formats, I created the Expense Tracker Dashboard using Streamlit and Python. The task was to simplify financial data ingestion, cleaning, and visualization. I built automated upload logic supporting multiple file types (CSV, Excel, JSON, PDFs), integrated fuzzy matching for column mapping, and deployed visual dashboards with Plotly. I also engineered features like savings rate and rolling expense averages. The result was an intuitive tool that helped users detect anomalies, track monthly spending patterns, and reduce duplicate entries,providing a solid foundation for better budgeting and financial wellness decisions.

Faced with the challenge of financial institutions lacking visibility into loan performance, I built a Loan Portfolio Dashboard using Power BI to solve this gap. The task was to transform 5,000 records of raw loan data into a meaningful, interactive reporting tool. I cleaned and modeled the dataset using Power Query and DAX, applied feature engineering to derive metrics like Fully Paid Rate and Average DTI, and structured a star schema for efficient analysis. As a result, the dashboard revealed that 13.7% of loans were charged off and uncovered patterns linking defaults to high DTI ratios and long-term loans. This helped simulate actionable strategies like risk-tier filtering and borrower verification, enabling smarter portfolio management decisions.