Smart Market Research - Khaireddine Arbouch

Revamp Edge – Smart Market Research & Business Expansion System

Client

Revamp Edge

Timeframe

May 23 - June 5

Services

Python, BeautifulSoup, OCR, Neo4j, NetworkX, Airflow, AWS Glue, Tableau, MongoDB

Source Code

Project Overview

This project delivered a fully automated market research and client acquisition engine for Revamp Edge, solving the inefficiencies of manual outreach and enabling scalable, data-driven business development. By building an intelligent system that scrapes, digitizes, maps, and segments the Moroccan B2B market, I created a reusable and expandable platform that now drives both strategic targeting and long-term growth.

Implemented Components

Multi-Layered Web Scraping Engine: Developed a modular scraping system to collect company data (name, location, size, industry, contact info) from diverse sources—news portals, company directories, business listings—using BeautifulSoup and custom rules per source.
OCR Pipelines for Offline Data: Implemented OCR-based ingestion pipelines to extract company names and contact information from scanned PDFs and legacy documents (especially effective with directories of architects and industrial firms).
Cross-Source Data Fusion: Kept scrapers and data sources separate by design to enable cross-database validation, improve accuracy, and build a clean, centralized customer intelligence warehouse.
Real-Time Data Warehousing & ETL: Created a streaming ETL pipeline using AWS Glue, Airflow, and MongoDB, ensuring the client database continuously updates with new records and changes in company information.
Graph-Based Relationship Mapping: Modeled the Moroccan market as a business graph using Neo4j and NetworkX, enabling strategic targeting by identifying network paths, key intermediaries, and referral leverage.
Segmentation & Roadmap Automation: Segmented the entire database by industry, company size, region, and network proximity. Then created personalized client roadmaps for high-potential accounts—starting small and scaling relationships based on proof of execution.
Market Intelligence Dashboards: Visualized B2B opportunities and market activity using Tableau, helping the agency track movement and prioritize outreach.

Tech Stack & Infrastructure

Scraping & Ingestion: Python, BeautifulSoup, Tesseract OCR
Data Warehousing: AWS Glue, MongoDB, Apache Airflow
Data Modeling: Neo4j (graph database), NetworkX (graph analytics)
Visualization: Tableau
Architecture: Modular microservices architecture with update-friendly ETL flows

Impact & Results

30+ High-Quality Clients Identified: Used the system to uncover and target over 30 previously untapped, high-potential B2B leads—especially in architecture, construction, and industrial sectors.
40% Reduction in Client Acquisition Time: Automated lead generation, data validation, and segmentation dramatically sped up the outreach process.
Strategic Scaling with Roadmaps: Deployed a start-small, grow-big approach to client onboarding, showing clear paths to expand the agency’s involvement as trust and complexity increase.
Long-Term Competitive Advantage: Delivered a scalable, intelligent platform that Revamp Edge can continuously feed and refine, ensuring compounding value over time.

Key Innovations & Learnings

Built a clean, extensible data lake for B2B market data in Morocco, merging unstructured, semi-structured, and scraped datasets.
Designed for truth-over-speed: Data accuracy was prioritized via cross-source conflict resolution before loading into the master database.
Proved that graph modeling is a superpower in B2B targeting—allowing personalized, strategic outreach based on network closeness.
Demonstrated that a data-first GTM strategy (Go-to-Market) can significantly outperform manual, unstructured client acquisition.

This project validated that scalable B2B growth is possible with the right architecture, intelligent data pipelines, and network-aware targeting—transforming how Revamp Edge approaches expansion, partnerships, and positioning in competitive markets.