Available for New Opportunities · Nigeria

Gogo Harrison Data Scientist & ML Engineer

I build predictive systems, analytical pipelines, and decision intelligence frameworks that transform raw business data into measurable competitive advantage. 3+ years shipping models that drive revenue, reduce cost, and accelerate growth.

97% Model Accuracy
28% Conversion Lift
25% Efficiency Gain
9+ Case Studies
View Projects Download Résumé
Gogo Harrison
Gogo Harrison
Data Scientist · ML Engineer · Analytics Specialist
Open to Remote Contract / Full-time 3+ Years Exp
About Me

I engineer intelligence that moves the needle on business outcomes.

I'm a Data Scientist and Machine Learning Engineer with 3+ years of hands-on experience building end-to-end analytical systems — from raw data ingestion to production-deployed predictive models. My work spans customer churn prediction, conversion funnel optimization, NLP pipelines, and executive-level BI dashboards.

What sets me apart is my obsession with business impact. Every model I build, every dashboard I design, has a clear downstream metric: revenue gained, cost reduced, or decision speed improved. I don't deliver notebooks — I deliver outcomes.

Currently building data intelligence systems at 10alytics (Canada), where I've achieved 97% model accuracy, driven a 25% improvement in operational efficiency, and partnered with cross-functional teams to translate complex model outputs into executive strategy.

"Data is only valuable when it changes a decision. I build systems where it always does."
Technical Stack

Core Competencies

A full-spectrum toolkit built for every stage of the data lifecycle — from ingestion to insight to deployment.

Machine Learning

Scikit-learn XGBoost Random Forest Classification Regression Clustering Feature Engineering Cross-Validation

Programming & Data

Python Pandas NumPy SciPy SQL CTEs PostgreSQL MySQL

Data Visualization & BI

Power BI Tableau Plotly Dash Matplotlib Seaborn Excel Dashboards

NLP & Text Analytics

NLTK TF-IDF SVM Sentiment Analysis Streamlit Flask

Analytics Engineering

ETL Pipelines Data Cleaning Data Modeling Statistical Inference Hypothesis Testing A/B Testing

Soft Skills & Methods

Stakeholder Management Agile Cross-Team Collab Executive Storytelling Problem Framing
<
Work

Selected Case Studies

8 curated projects across ML, analytics, NLP, and BI — each built for measurable business impact.

Tier 1 · Flagship Projects
01 Brazilian E-Commerce Predictive Intelligence dashboard preview
ML · Geospatial · BI

Brazilian E-Commerce Predictive Intelligence

Built 4 production-grade predictive models (XGBoost, Random Forest) over 1.5M+ records — delivery delay prediction, review score forecasting, churn classification (ROC-AUC: 0.73), and lead conversion scoring. Delivered a 4-tab interactive Dash BI dashboard with 12+ visualizations and identified R$16–19M in annual revenue opportunity.

ROC-AUC: 0.73 1.5M+ Records R$16–19M Potential
Python XGBoost Scikit-learn GeoPandas Plotly Dash SciPy
02 Sales funnel analysis Sankey diagram preview
Analytics · ML · Dash

E-Commerce Sales Funnel Analysis

Uncovered an expansion funnel model generating 5.4× orders per seller across 7,940 MQLs. Identified a statistically significant 6× conversion multiplier tied to delivery speed (p < 0.0001) and a 201% variability bottleneck in the approval stage — directly informing a targeted process automation strategy.

5.4× Expansion Factor 6× Conversion Lift 20–25% CAC Reduction
Python Pandas Plotly Dash SciPy Hypothesis Testing
03
NLP · ML · Web App

Product Review Sentiment Analysis System

Engineered an end-to-end NLP pipeline classifying e-commerce product reviews with 93% accuracy using TF-IDF vectorization and SVM. Deployed a production Streamlit web app for real-time sentiment prediction — cutting manual review analysis time by 20% and providing marketing teams with automated product improvement signals.

93% Accuracy 20% Time Saved Deployed Web App
Python NLTK TF-IDF SVM Streamlit Flask
Tier 2 · Breadth & Domain Range
04 Marketing campaign fraud heatmap and RFM segmentation preview
Analytics · RFM · Risk

Marketing Campaign Performance Analysis

Delivered end-to-end marketing intelligence across 55,910 transactions for a plant e-commerce brand. Identified Google Ads as an 80% revenue driver, applied RFM segmentation to surface 6,090 high-potential customers, and flagged a critical 6.5% fraud rate — 5–7× above industry benchmark — projecting $450K in annual savings with targeted fraud mitigation.

80% Revenue from 1 Channel 6.5% Fraud Rate Flagged $450K Savings Potential
Python Pandas RFM Analysis Plotly Seaborn Risk Assessment
05 Discount impact Power BI dashboard preview
SQL · Power BI · Analytics

Discount Impact & Profitability Study

Designed complex MySQL CTEs and conditional aggregation to evaluate a 15% discount campaign across 5,000+ transactions. Proved the campaign drove cannibalization among existing customers rather than new acquisition — saving the business from a costly reallocation decision and surfacing that Electronics drove 60%+ of overall margin.

+15% Profit Increase Cannibalization Proven Electronics = 60% Margin
SQL MySQL CTEs Power BI Python
06 Superstore sales choropleth map and treemap preview
Analytics · Visualization

Superstore Sales Intelligence — Multi-Dimensional Analysis

Analyzed 9,994 transactions across 4 years to surface revenue drivers, customer value patterns, and geographic performance gaps. Identified a 51% revenue growth trend, Q4 seasonality at 38% of annual sales, and California as a $457K market leader — delivering a full 7-dimension strategic roadmap with choropleth maps, treemaps, and sunburst charts.

51% Revenue Growth $457K Top State Q4 = 38% of Sales
Python Pandas Plotly Seaborn Choropleth Time Series
Tier 3 · Supporting Depth
07 Africa child mortality choropleth map preview
ML · Public Health · Geospatial

Child & Infant Mortality in Africa — Predictive Modeling

Applied Random Forest classification across multi-source WHO health datasets to identify high-risk regions across 54 African countries. Feature importance analysis confirmed vaccination coverage and maternal health access as the dominant mortality predictors — providing SDG 3-aligned, data-driven intervention priorities for policymakers.

SDG-3 Aligned Random Forest 54 Countries
Python Scikit-learn GeoPandas Seaborn Random Forest
08 Customer retention cohort heatmap preview
Analytics · Python

Time-Based Cohort Analysis — Customer Retention

Conducted end-to-end cohort retention analysis for E-Shop Pro across 12 monthly cohorts. Identified critical early churn hotspots in the first 3–4 months post-acquisition, surfaced seasonal retention patterns, and built dual heatmap visualizations tracking retention and churn rates — enabling cohort-specific intervention strategies.

12 Cohorts Tracked Early Churn Mapped Seasonal Patterns
Python Pandas Seaborn Matplotlib Cohort Analysis
09 Sterling E-Commerce Tableau dashboard preview
Tableau · Excel · BI

Sterling E-Commerce Sales Analysis — Tableau Dashboard

Cleaned and analyzed 283,078 orders across 19 variables in Excel, then built a comprehensive Tableau dashboard spanning customer segmentation, market basket analysis, and time series trends. Discovered Mobiles & Tablets drive 56% of revenue despite 29% of orders, and an 800%+ seasonal revenue surge from February to April.

283K+ Orders 56% Revenue Category 800% Seasonal Surge
Tableau Microsoft Excel Customer Segmentation Market Basket Time Series
View All Projects on GitHub
Background

Experience & Credentials

Work History

Nov 2023 – Present
Data Scientist (Remote)
10alytics · Canada
  • Built classification & regression models (Scikit-learn) for churn prediction and delivery delay forecasting — achieving 97% model accuracy
  • Engineered automated SQL/Excel data pipelines delivering a 25% improvement in operational efficiency and reporting speed
  • Applied advanced regression and hypothesis testing — improving forecasting accuracy by 15%
  • Partnered with cross-functional teams to translate model outputs into executive business strategies
Jul 2022 – Aug 2023
Data Analyst
Legend Internet PLC · Nigeria
  • Analyzed end-to-end sales funnels using SQL & Python — increasing lead-to-conversion rates by 28%
  • Led a data validation initiative reducing entry errors by 25% and improving executive report reliability by 20%
  • Overhauled legacy reporting by implementing automated workflows enabling real-time stakeholder access
  • Managed data collection lifecycles across multiple business units, ensuring alignment with corporate KPIs

Certifications & Education

Master Data Scientist

10alytics · Canada  |  April 2024

Statistical Modeling Microsoft Excel Machine Learning Predictive Analytics Data Visualization Computer Vision SQL EDA Python Tableau Power BI

Bachelor of Technology — Petroleum Engineering

Rivers State University, Nigeria  |  Second Class Upper · Sep 2018

Strong quantitative foundation in applied mathematics, thermodynamics, and systems modeling — directly transferable to statistical computing and ML problem-solving.

Let's Connect

Ready to turn your data into a competitive edge?

I'm actively seeking Data Science, Machine Learning, and Analytics Engineering roles. If you're looking for someone who ships models that actually move business metrics, let's have a conversation.

Email gogoharrison66@gmail.com
Phone +234 706 076 4650
LinkedIn linkedin.com/in/gogo-harrison
GitHub github.com/gogoharrison

Send a Message