my_data_analysis_portfolio

Portfolio Banner

📧 Connect with me

LinkedIn GitHub Email

A curated portfolio of data analysis, business intelligence, and SQL projects demonstrating real-world problem solving using Excel, Power BI, DAX, Power Query, and SQL.
Each project reflects an end-to-end analytics workflow: raw data → transformation → modeling → insight generation.

🎓 Professional Experience

Data Analysis Intern — LearnWithWida

Internship

Completed a hands-on data analysis internship focused on solving real-world business problems using structured datasets. Gained practical experience across the full analytics lifecycle including data cleaning, exploratory analysis, modeling, and insight communication.

Key Contributions

Tools & Technologies

🔗 Organization: https://learnwithwida.com/

Table of Contents

  1. Professional Experience
  2. DAX_PowerBI_Analysis
  3. Netflix_PowerQuery
  4. Data_Wrangling_PowerQuery
  5. StudentManagementSystem
  6. Advanced_SQL_Forensic_Analysis
  7. Core Skills & Tools

DAX_PowerBI_Analysis

HR Analytics Portfolio: Employee Satisfaction & Work-Life Balance

Repository:https://github.com/tybolalekan/DAX_PowerBI_Analysis

Project Overview

This project involves a comprehensive analysis of employee survey data to uncover insights into workplace satisfaction, productivity, and work-life balance. By leveraging Power BI, I transformed raw survey data into an interactive dashboard that helps HR leadership identify key drivers of employee engagement and potential areas for improvement.

Technical Implementation

1. Data Transformation (Power Query)

The raw dataset underwent several cleaning and shaping steps in Power Query to ensure data integrity and usability:

2. Data Modeling & DAX

Advanced DAX (Data Analysis Expressions) was the core of this project, enabling complex calculations and dynamic filtering.

Data Shaping & Tables

Advanced Measures

  1.     Work-Life Balance scores by Department and Job Level.

  2.     Correlation between Stress levels and Sleep hours.

3. Visualizations & Insights

The final Power BI Dashboard features:

Key DAX Functions Leveraged

Table Manipulation SELECTCOLUMNS, ADDCOLUMNS, CALENDAR

Iterators SUMX, AGGREGATEX

Logical & Conditional SWITCH, IF, AND, OR

Time Intelligence TOTALYTD, SAMEPERIODLASTYEAR, DATEADD

Date & Time DATE, YEAR, MONTH, QUARTER

This project was completed as part of my professional data analysis training and internship experience. —

Netflix_PowerQuery

Netflix

Repository:https://github.com/tybolalekan/NetFlix_PowerQuery

Project Overview

An exploratory and analytical project on the Netflix Titles Dataset (8,807 records) aimed at transforming inconsistent metadata into insights on global content strategy and growth trends.

Technical Implementation

1. Data Transformation with Power Query

The raw dataset was “dirty,” containing significant gaps and inconsistent formatting. I utilized Power Query to implement a robust ETL (Extract, Transform, Load) process:

2. Insight Generation with Power BI

The cleaned data was modeled and visualized in Power BI to uncover key business metrics:

Key Deliverables

Conclusion

The project demonstrates the ability to manage complex data cleaning tasks using Power Query and deliver professional-grade business intelligence through Power BI, turning unstructured metadata into strategic insights. This project was completed as part of my professional data analysis training and internship experience.

Data_Wrangling_PowerQuery

Repository:https://github.com/tybolalekan/Data_Wrangling_PowerQuery

Overview

This project showcases the development of an interactive Business Intelligence (BI) solution using Power BI and Power Query. The goal was to transform raw, fragmented sales data into a high-performance dashboard that provides real-time strategic insights for an international apparel brand.

Technical Implementation: The ETL Pipeline

1. Extract & Transform (Power Query)

Leveraging the M language engine in Power Query, I implemented a robust ETL process to ensure data reliability:

    - Forced Date types on transactional timestamps to enable time-intelligence functions.

    - Standardized Fixed Decimal types for RATE and GROSS AMT to prevent rounding errors in high-value calculations.

    - Cleansed SKU and Style identifiers to ensure uniform grouping across all visuals.

2. Advanced Analytics & DAX Modeling

To go beyond simple aggregations, I developed custom DAX (Data Analysis Expressions) measures:

Key Dashboard Features

Financial Performance Tracking

Market & Customer Profiling

BI-Driven Business Value

  1. Dynamic Decision Making: Replaced static reports with an interactive dashboard, reducing analytical lead time by 90%.

  2. Standardized Reporting: Created a single source of truth for Sales, Finance, and Inventory teams.

  3. Proactive Inventory Management: Dashboard highlighting “Dead Stock” candidates allows for immediate liquidation planning.

—  This project was completed as part of my professional data analysis training and internship experience. —

StudentManagementSystem

Repository:https://github.com/tybolalekan/StudentsManagementSystem

The primary objective of this project is to design and create a relational database for a Student Management System (SMS) using SQL. The database should efficiently store and manage data on students, courses, instructors, and enrollments, enabling users to perform key administrative functions such as adding students, assigning courses, and generating dynamic reports through SQL queries.

Welcome to the Student Management System (SMS) project. This repository showcases a complete end-to-end data solution, from relational database design and automated ETL pipelines to advanced analytical reporting.

Project Overview

In many academic institutions, managing student records, course enrollments, and faculty assignments manually is error-prone and inefficient. This project provides a robust, scalable SQL-based system that automates these processes, ensuring data integrity and providing actionable business insights.

Key Highlights:

Technology Stack

Database Architecture

The system is built around five core entities: Departments, Students, Instructors, Courses, and Enrollments.

Entity Relationship Diagram (ERD)

ER Diagram

Analytical Insights (SQL)

The system isn’t just for storage; it’s built for analysis. Below are examples of key business questions solved using T-SQL.

1. Enrollment Distribution

Querying the most popular courses to inform resource allocation.

SELECT c.CourseName, COUNT(e.StudentID) AS EnrolledStudents

FROM Courses c

LEFT JOIN Enrollments e ON c.CourseID = e.CourseID

GROUP BY c.CourseName

ORDER BY EnrolledStudents DESC;

2. Multi-disciplinary Students

Identifying students taking multiple courses for scholarship eligibility tracking.

SELECT s.Name, COUNT(e.CourseID) AS CourseCount

FROM Students s

JOIN Enrollments e ON s.StudentID = e.StudentID

GROUP BY s.Name

HAVING COUNT(e.CourseID) > 1;

[!NOTE]

View all analytical queries in QUERIES.sql.

Repository Structure

Quick Start

  1. Run Database_Creation.sql.

  2. Execute the Create_*.sql scripts in order (Departments -> Students -> etc.).

  3. Run the Populate_*.sql scripts to load the sample data.

  4. Open QUERIES.sql to view insights.

This project was completed as part of my professional data analysis training and internship experience. —

Advanced_SQL_Forensic_Analysis

Repository:https://github.com/tybolalekan/advanced_sql_forensic_analysis

Welcome to the Roland Greene Murder Investigation. This project is a comprehensive technical case study demonstrating the power of Relational Database Management Systems (RDBMS) and T-SQL in solving complex, multi-layered problems.

Project Overview

This investigation simulates a real-world data analysis scenario where digital evidence from disparate sources must be reconciled to uncover the truth.

Tech Stack & Skills

    - Advanced Joins & Subqueries

    - Common Table Expressions (CTEs)

    - Window Functions (ROW_NUMBER)

    - Mathematical Evidence Scoring

    - Forensic Data Modeling

Technical Architecture

Data Models

I designed a relational schema to handle four critical streams of evidence:

  1. Suspects: Personal profiles, roles, and initial alibis.

  2. Access Logs: Security door interactions across the estate.

  3. Call Records: Communication logs including timestamped interactions with the victim.

  4. Forensic Events: Key investigative milestones that ground the timeline.

The Scoring Algorithm

To avoid manual guesswork, I implemented an Evidence-Based Scoring System. Each suspect was ranked based on:

Key Findings

[!IMPORTANT]

The investigation successfully isolated one suspect who matched 100% of the forensic evidence, including a provably false alibi and a verified presence at the crime scene during the 10-minute murder window.

Automated Workflow

The project features a modular architecture of 13 SQL scripts, taking the investigator from database creation to a final executive summary.

  1. Setup: Database creation and optimized data ingestion.

  2. Analysis: Individual scripts for vault access, call logs, and alibi validation.

  3. Resolution: A final “Identify Killer” SQL script that synthesizes all scores into a definitive conclusion.

This project was completed as part of my professional data analysis training and internship experience.

Core Skills & Tools

Analytics & BI

Technical Skills

Professional Strengths

📧 Connect With Me

LinkedIn GitHub Email