Browse concise cards that link to richer detail sections, with proof visuals and clear outcomes for each build.
Scan the highlights, then jump to the detailed sections below for proofs and outcomes.
End-to-end Python pipeline transforming messy retail data into executive-grade financial insights with automated SQL validation.
Data cleanup, SQL-driven revenue mapping, and executive-friendly dashboard output.
Engineered a unified revenue recovery system for retail markets. Merged 3 regional datasets into a single USD-standardized master dataset with automated deduplication and reporting.
A high-performance Streamlit dashboard visualizing regional sales trends, product performance, and data quality metrics derived from the cleaned retail dataset.
Successfully restored 20,500+ records of corrupted sales data using custom Python automation engines. Delivered a board-ready BI dashboard with regional performance and seasonal growth insights.
End-to-end data transformation pipeline leveraging Power Query and DAX to automate regional revenue aggregation and interactive performance reporting.
Workbook aggregation pipeline that reads multiple files, preserves source context, and removes duplicates.
Product ranking, laptop revenue comparison, and month-over-month growth analysis in one workbook.
High-performance portfolio risk dashboard managing 2,240 global project records with multi-currency USD normalization.
This build cleaned malformed sales data, removed invalid rows, normalized country labels, validated dates, and generated pivot-ready workbook output for stakeholder review.
Every frame in this sequence captures the analytical process: moving from messy raw data, mapping the ETL architecture, writing advanced PostgreSQL queries, and producing the final Month-over-Month trajectory.
The following interactive briefing is powered by the automated ETL pipeline, serving real-time financial insights and regional performance metrics directly from the processed dataset.
This system starts with intentionally messy regional Excel files and turns them into a clean USD-standardized master dataset, an analysis workbook, and a professional dashboard that a client can open with one click.
Automated Build Status: I have prepared the 30-second transition sequence using your clean captures (5 seconds per frame). This video highlights exactly how the raw inputs (Market IDs) become the final Dashboard.
Every frame in this gallery is sourced directly from the project's clean capture state.
This interactive dashboard provides real-time insights into the cleaned retail sales data. It features executive-level KPIs, regional breakdown charts, and a side-by-side comparison of the dataset's state before and after the cleaning pipeline.
This flagship project involved the mass restoration of over 20,500 records of multi-regional sales data. The challenge was to convert raw, corrupted "Digital Chaos" into a high-fidelity environment with 100% currency normalization (NGN) and professional world identities.
These visualizations were generated directly from the recovered 20,500-record dataset, providing immediate board-level insights into regional performance and seasonal peaks.
Below is the actual Executive Handoff Report delivered to the client, summarizing the 100% quality score reached after the Absolute Perfection cycle.
# EXECUTIVE CLIENT HANDOFF: Extreme Data Recovery v2
**Project Status:** 100% Complete | **Data Integrity:** 100% Perfect | **Senior Analyst:** Chinua Analytics
---
## 🌍 1. GLOBAL REVENUE SEGMENTATION
The following analysis identifies the primary revenue drivers across our multi-regional business model. Nigeria currently stands as the high-growth leader, contributing significantly more to the global total than the USA or UK.
> [!TIP]
> **Strategic Insight:** High-growth investment should be prioritized for the Nigeria region, while the UK and USA require targeted "Re-Engagement" marketing campaigns.
---
## 📈 2. SEASONAL SALES TRENDS
Our time-series analysis reveals a 100% predictable seasonal surge. Peak revenue is concentrated in the Q4 window, specifically late Q4.
---
## 🛒 3. PRODUCT PORTFOLIO AUDIT
Electronics remain the core engine of our business, driving nearly 20% of all global revenue. This is followed closely by the Fashion and Health categories.
---
## 🛡️ 4. DATA INTEGRITY AUDIT (PROOF OF WORK)
Every single record has been processed through the **Absolute Perfection (v2)** cycle. All duplicates have been purged, and all phone numbers/emails have been normalized to world-class standards.
| Metric | Cleaned Rows | Quality Score |
| :--- | :--- | :--- |
| **Total Processed** | 20,500 | 100% |
| **Fuzzy Duplicates** | 2,518 Purged | 100% |
| **Integrity Checks** | Passed | 100% |
This solution replaces manual Excel consolidation with a robust PowerBI ETL pipeline. It uses Power Query to ingest multi-region source data, applies DAX for complex revenue logic, and delivers interactive reporting for stakeholder review.
This pipeline scans workbook outputs, ingests non-empty sheets, tags file and sheet origin, removes duplicate rows, and creates a single master workbook for downstream analysis.
These frames capture the automated script actively scanning source files, standardizing structures, and consolidating outputs into a master file.
These business intelligence charts demonstrate the direct outcomes of the Merge Engine, providing immediate boardroom-level visibility into revenue trends and product performance.
This workbook focuses on comparative analysis: top products in the USA, top laptop revenue contributor, and month-over-month revenue growth in a concise sheet designed for review speed.
Every frame in this sequence captures the real-time bot operations, transitioning instantly from raw data input logic to the finalized Comparative Growth Workbook output.
This premium dark-mode visualization was dynamically rendered against the generated sales pipeline dataset, highlighting the precise revenue peaks without requiring an active Excel GUI.
Click any image to enlarge and inspect the transformation details.
From 2,240 rows of multi-region currency chaos to a unified, executive-grade intelligence platform. This project demonstrates full-stack data engineering: from raw ETL pipelines to premium glassmorphism visualization.
To ensure 100% accuracy in financial reporting across NGN, ZAR, and USD, I implemented a comprehensive pytest suite covering 11 critical data transformation rules.
rootdir: /Global-Construction-Risk-Analytics-Platform
collected 11 items
tests/test_calculations.py .... [ 36%]
tests/test_cleaning.py ... [ 63%]
tests/test_currency.py .... [100%]
================= 11 passed in 0.12s =================
This flagship engineering project solves the "Messy Data" problem for retail enterprises. It automates the ingestion of corrupted CSV records, purges duplicate entries, normalizes financial values, and serves the results to a PostgreSQL data warehouse for real-time reporting.
$10.7M
Peak: $1.07M (Mar '25)< 10s
Reduced from 4 Hours100%
Duplicates Resolved"By implementing an automated ETL pipeline, we transitioned from fragmented Excel silos to a unified PostgreSQL intelligence layer. This system currently identifies that 37.3% of revenue is driven by a single product line (Rice), enabling targeted regional strategy adjustments."
The following interactive dashboard is the final output of the automated pipeline. It delivers real-time regional performance, product concentration, and seasonal trends directly from the cleaned dataset.
This sequence demonstrates the automated pipeline's ability to ingest corrupted sales records, standardize multi-format datasets, and derive high-level business metrics for decision-makers.
