Kolawole Emmanuel Oluwaseyi

Excel

01

ExcelPivot TablesPower QueryDashboard

Executive Sales Dashboard

⚡ Business had $1.6M in sales data spread across stores, regions and salespersons with no single view to track performance, identify top performers or spot underperforming areas.

Built a fully dynamic Excel executive dashboard with KPI cards, bubble charts, sparklines and interactive slicers — giving leadership a real-time view of sales performance across every dimension.

Total revenue of $1.6M — up 29% vs prior year with 271 total orders processed

Tablets led all products at $289.9K contributing 17.89% of total revenue

Ryan was top regional manager at $368.5K · North region outperformed all 5 regions

Retail segment accounted for 52.27% of sales · Cash was the #1 payment method at $374.7K

Women Involvement & Peacebuilding — Plateau State, Nigeria

Excel

02

ExcelPivot TablesSurvey AnalysisDashboard

Women Involvement & Peacebuilding — Plateau State, Nigeria

⚡ Local government needed data to understand women's actual participation in peacebuilding, what barriers they face and what strategies would work — based on 400 real survey responses across 6 LGAs.

Analyzed a 400-respondent survey across Bassa, Jos North, Jos South, Wase, Mangu and Shendem LGAs. Used Excel Pivot Tables and dashboard design to surface meaningful patterns in gender, participation and policy needs.

282 of 400 respondents (70.5%) were female — showing strong female representation

184 respondents confirmed women play an active and significant role in peacebuilding efforts

Lack of resources & funding was the #1 barrier cited by 160 respondents — 40% of all responses

Legal reforms to protect women in politics was the top recommended strategy at 140 responses

Power BI

03

Power BIDAXSentiment AnalysisDashboard

Customer Feedback Sentiment Analysis

⚡ Company had 40,000 feedback entries across 10 products and 100 customers but no structured way to track customer satisfaction, which products had problems or how sentiment changed month by month.

Built an interactive Power BI dashboard with dynamic slicers for product and feedback type — allowing management to drill into sentiment patterns, compare products and monitor monthly feedback volumes in real time.

60.12% of all 40,000 feedback entries were positive — indicating overall customer satisfaction

Product G and Product A led all products with 4.1K feedback entries each throughout the year

May recorded the highest monthly feedback at 3,533 entries · February was the lowest at 3,093

Suggestions (13,386), Complaints (13,366) and Praise (13,248) were almost perfectly evenly distributed

$Healthcare Analysis Dashboard & Insights$

Power BI

04

Power BIDAXHealthcare Analytics3-Page Report

Healthcare Analysis Dashboard & Insights

⚡ Hospital network had no unified view of 55,500 patient records — administrators could not track common conditions, billing variations across insurance providers or how admission types affected costs.

Built a 3-page Power BI report (Overview, Medical Insights, Billing Dashboard) giving hospital leadership full visibility into patient demographics, disease patterns, insurance billing and yearly financial trends from 2019–2024.

Total billing reached $1.42bn across 55,500 patients in 40K hospitals over 6 years

Middle-aged adults were the largest patient group at 20.5K · Elderly at 9K · Old-Aged at 8.1K

Germany had significantly higher discharge rates vs Spain (16.67%) and France (16.17%)

Urgent admissions generated the highest billing at $479.49M · 2021 was the peak billing year at $286.55M

Excel

05

ExcelDashboardFinancial ModellingCalculations

Personal Finance Tracker

⚡ Individual earning from 3 different income sources had no clear dashboard to track available balance, monitor expense categories or understand weekly and monthly spending patterns.

Built a dynamic personal finance dashboard in Excel with a credit card balance view, income source breakdown, expense category ranking and both monthly and weekly trend charts — all with interactive month slicers.

Available balance of $35,249 from total income of $65,440 against total expenses of $30,191

Data with Decisions platform contributes 76% of all income at $50,000 — the dominant source

Housing is the biggest expense at $9,000 followed by Groceries at $6,454 and Clothes at $4,304

Monday recorded the highest weekly spending · Max single expense was $900 · Max income $1,500

Excel

06

ExcelMap VisualizationMarket AnalysisDashboard

Nigeria Car Market Dashboard

⚡ No structured data analysis existed of the Nigerian car market — buyers, sellers and analysts had no visibility into pricing, popular brands, transmission preferences or regional hotspots across Nigeria.

Analyzed the Nigerian car market using data from Car45.com. Built an Excel dashboard with integrated Bing map visualization covering vehicle metrics, top models, colour preferences and geographic sales distribution.

Average car price in Nigeria is ₦5.0M with avg mileage 28.4K km, 217hp and 3.1K cc engine

Toyota leads the Nigerian car market significantly over Volkswagen, Nissan and Mercedes-Benz

94% of Nigerian drivers prefer Automatic transmission — only 6% opt for Manual gearbox

Lagos and Abuja are the primary car sales hotspots accounting for the majority of market volume

Business Performance & Customer Insights

Excel

07

ExcelKPI DashboardMap VisualizationCustomer Analysis

Business Performance & Customer Insights

⚡ Business was tracking sales, profit and customer satisfaction in scattered spreadsheets with no unified view of performance vs target, regional breakdown, customer growth or experience quality scores.

Built a professional Excel KPI dashboard with quarterly filters, monthly sales vs target trend lines, regional donut chart, customer growth bar chart, customer experience ratings and a live geographical map.

Sales of $157,361 at 86% completion rate vs $166,999 target — 14% gap flagged for action

Profit of $113,301 at 85% CR · Customer acquisition at 84% with 63 total customers onboarded

East region leads all regions in sales distribution at 32% · followed by North, South and West

Speed is the highest-rated customer experience factor at 96 · Availability is the weakest at only 41

Excel

08

ExcelChurn AnalysisStatistical AnalysisMap Visualization

Bank Customer Churn Analysis Dashboard

⚡ Bank was losing 20% of customers with no data-driven understanding of who was churning, why they were leaving, which demographics were highest risk or what the financial profile of churned customers looked like.

Built a multi-page Excel dashboard (Dashboard, Demographics, Analysis, Dataset) with scatter plots, correlation analysis, geographic churn map and a full written insights panel — a complete churn risk picture.

2,038 customers churned (20.38%) out of 10,000 total · 7,962 retained · Avg satisfaction 3.01/5

Germany had highest churn at 32.44% — significantly above Spain (16.67%) and France (16.17%)

Old age group churns most at 49.76% · Age-churn correlation confirmed at 0.50 — strongest predictor

Female churn (25.07%) significantly higher than male (16.47%) · Churned avg balance $91,109 vs $72,743

Excel + SQL

09

ExcelSQL ServerPivot TablesDashboard

Pizza Sales Analysis — Excel & SQL

⚡ Pizza restaurant had 21,350 orders but no structured analysis of which products drove revenue, when customers ordered most, which pizza sizes sold best or which items were consistently underperforming.

Wrote SQL queries in SQL Server to extract all KPIs, daily/hourly trends, category breakdowns and best/worst sellers. Then built a complete Excel dashboard visualizing all results with charts, slicers and business narrative.

Total revenue $817,860 · 21,350 orders · 49,574 pizzas sold · Avg order value $38.31

Friday is the busiest day at 3,538 orders · Peak hours are 12–1pm and 4–8pm every day

Classic category leads at 26.91% revenue share · Large size dominates at 45.89% of all orders

Classic Deluxe Pizza is the #1 best seller at 2,453 units · Brie Carre Pizza worst at only 490

SQL

10

SQL ServerData CleaningCTEsWindow Functions

World Layoffs — SQL Data Cleaning

⚡ A raw dataset of global tech company layoffs contained duplicates, inconsistent formatting, NULL values stored as strings, mismatched country names and wrong data types — making it completely unreliable for analysis.

Performed a full professional SQL data cleaning pipeline — working on a staging table to protect raw data, removing duplicates using CTEs, standardizing text values, handling NULLs and correcting data types throughout.

Created staging table using CREATE TABLE + INSERT SELECT to safely clean without touching raw data

Removed all duplicates using ROW_NUMBER() OVER(PARTITION BY 9 columns) inside a CTE — deleted row_num > 1

Standardized Crypto variants → "Crypto", trimmed company names, fixed "United States." trailing period

Converted total_laid_off to INT and funds_raised_millions to DECIMAL(10,2) after clearing NULL strings

World Layoffs — Exploratory Data Analysis (SQL)

SQL

11

SQL ServerEDACTEsLAG & RANK Functions

World Layoffs — Exploratory Data Analysis (SQL)

⚡ After cleaning the World Layoffs dataset, the business needed deep exploratory analysis to understand which companies, industries and countries were most affected and how the trend evolved month by month.

Conducted advanced SQL EDA using CTEs, window functions, rolling totals and LAG comparisons — uncovering layoff patterns by company, industry, country, year and funding stage across the entire cleaned dataset.

Built rolling monthly layoff totals using CTE + SUM() OVER(ORDER BY yr, mth ROWS UNBOUNDED PRECEDING)

Used DENSE_RANK() OVER(PARTITION BY yr) to find top 5 companies by layoffs for each year

LAG() function calculated month-over-month differences and tracked acceleration of layoffs over time

Companies with percentage_laid_off = 1 sorted by funds_raised_millions revealed the biggest collapses

Molecular Solubility Prediction — Machine Learning

Python

12

PythonJupyter Notebookscikit-learnLinear Regression

Molecular Solubility Prediction — Machine Learning

⚡ Predicting how well a drug compound dissolves in water traditionally requires expensive lab experiments. The goal was to build an ML model that predicts solubility computationally from molecular descriptors.

Built a Linear Regression ML model using the Delaney solubility dataset. Used MolWt, MolLogP, NumRotatableBonds and AromaticProportion as features to predict logS (aqueous solubility) with scikit-learn.

Features: MolWt, MolLogP, NumRotatableBonds, AromaticProportion — all key molecular descriptors

Applied train_test_split (80/20) then trained LinearRegression model from scikit-learn on training set

Achieved R² score of 0.7374 — model explains 73.74% of variance in molecular solubility

Directly applies a First-Class BSc Chemistry background to real machine learning drug prediction

Wine Chemical Composition Classification — ML

Python

13

PythonJupyter Notebookscikit-learnpandas · matplotlib

Wine Chemical Composition Classification — ML

⚡ Wine quality classification is typically done by expert sommeliers — a slow subjective process. The goal was to build an ML model that classifies wine samples objectively from chemical composition data.

Loaded the sklearn wine dataset (178 samples, 14 chemical features). Performed EDA then built a classification model — demonstrating how chemistry domain knowledge drives better feature interpretation.

Dataset: 178 wine samples · 14 features including alcohol, malic_acid, ash, flavanoids, magnesium

Used sklearn load_wine() · df.head() showed clear chemical variation across wine classification groups

Applied full ML pipeline: data loading → EDA → feature selection → model training → evaluation

Chemistry domain knowledge drives better ML feature interpretation — a rare cross-disciplinary skill

Kolawole
EmmanuelOluwaseyi

My Skill Set

My Projects

Turning Data Into
Decisions That Matter

Open to Opportunities

Send Me a Message

KolawoleEmmanuelOluwaseyi

My Skill Set

My Projects

Turning Data IntoDecisions That Matter

Open to Opportunities

Send Me a Message

Kolawole
EmmanuelOluwaseyi

Turning Data Into
Decisions That Matter