SAS Ride-Halling 

<h1>RFM Driver Segmentation&nbsp;</h1><h2>Context</h2>As part of the Growth &amp; Retention strategy, I led an end-to-end project to design and implement a scalable RFM (Recency, Frequency, Monetary) segmentation model for drivers. The goal was to enable targeted lifecycle campaigns through CRM platforms (Braze), improving driver retention, reactivation, and engagement.Initially built in DSS with Redshift and S3, the project is now being migrated to GCP with Dataform as part of a broader initiative to modernize our data platform and improve scalability, governance, and cost efficiency.<h2>Project Workflow</h2><h3>1. Data Source Integration (DSS → GCP Migration)</h3><ul><li>Legacy setup (DSS):<ul><li>Orchestrated data pipelines in DSS (Dataiku).</li><li>Integrated Redshift analytics tables and S3 raw data exports.</li><li>Built reusable SQL, Python, and Spark recipes for preprocessing, cleaning, and aggregating driver ride-level data.</li></ul></li><li>New setup (GCP + Dataform):<ul><li>Designed and built new pipelines in Dataform on BigQuery, fully replacing DSS flows.</li><li>Defined incremental ingestion logic, partitioning strategies, and deduplication windows to handle weekly refresh cycles.</li><li>Ensured data governance, reusability, and CI/CD integration in GCP.</li></ul></li></ul><h3>2. RFM Segmentation Logic</h3><ul><li>Defined driver-level segmentation based on:<ul><li>Recency → Time since the last ride.</li><li>Frequency → Number of active ride weeks.</li><li>Monetary → Driver earnings from rides.</li></ul></li><li>Developed multi-layer segmentation:<ul><li>Macro segments → Active, Hibernating, Leaving, Never Activated, Old.</li><li>Micro segments → New, Active 8w, Leaving to Save, etc.</li><li>Sub-segments → Champions, Promising, Occasional, based on earnings &amp; activity quartiles.</li></ul></li><li>Implemented segmentation logic in SQLX (Dataform) with incremental logic on BigQuery.</li></ul><h3>3. Activation to CRM (Braze Integration)</h3><ul><li>Automated API-based export of driver segments from the data pipeline to Braze.</li><li>Exposed each segment as an audience list available in Braze campaigns.</li><li>Enabled lifecycle campaigns such as:<ul><li>New driver onboarding nudges.</li><li>Retention campaigns for active drivers.</li><li>Reactivation campaigns for hibernating drivers.</li><li>Incentive campaigns for champions and high-potential drivers.</li></ul></li></ul><h3>4. Data Modeling for Analytics (Looker ML)</h3><ul><li>Modeled the output RFM segmentation table into LookML views.</li><li>Defined dimensions and measures (e.g., rides_8w, revenue_alltime, RFM_Segment_L2).</li><li>Built an Explore for Growth &amp; CRM teams to self-serve insights.</li><li>Designed dashboards to monitor segment evolution, measure retention uplift, and track campaign ROI.</li></ul><h2>Impact</h2><ul><li>Delivered a fully automated E2E RFM segmentation system, refreshed weekly.</li><li>Enabled data-driven CRM campaigns targeting different lifecycle stages.</li><li>Provided visibility to business teams via Looker dashboards.</li><li>Migrated workflows from legacy DSS/Redshift to modern GCP/Dataform/BigQuery, improving scalability, reliability, and governance.</li><li>Strengthened collaboration between Growth, Analytics, and CRM teams by providing actionable segments directly in Braze.</li></ul><h2>Skills Demonstrated</h2><ul><li>Data Engineering: SQL, Python, Spark, BigQuery, Dataform, DSS orchestration.</li><li>Data Modeling: Redshift &amp; BigQuery → LookML modeling (views, explores, dashboards).</li><li>Analytics: RFM framework, segmentation design, CRM ROI tracking.</li><li>Growth Strategy: Translating analytics into actionable CRM campaigns (Braze integration).</li><li>Cloud Migration: Redshift/S3 → GCP/Dataform migration with CI/CD and incremental logic.</li><li>Project Management: Owned the full lifecycle from data ingestion to CRM activation and business monitoring. </li></ul>

Big Data

Data Visualization

Data Analysis

RFM Driver Segmentation 

SAS Ride-hailing  

<h1>Ambassador Influencers Partnership Contest – Ride-hailing (Europe - Africa)</h1><h2>Context &amp; Challenge</h2>The Acquisition Paid team at Heetch launched influencer partnership campaigns (Ambassador Program) to boost passenger acquisition and engagement through contests and voucher-based incentives. The challenge was to measure end-to-end performance across partnership channels, track voucher usage, and assess the campaigns’ impact on activation, rides, GMV, and retention.Data was fragmented (campaign tracking files, S3 bonus records, Redshift passenger &amp; ride data), making it difficult for Growth &amp; Marketing teams to assess ROI and optimize campaigns.<h2>Objective</h2>Design and deliver a scalable end-to-end (E2E) analytics pipeline to:<ul><li>Consolidate campaign, voucher, and ride data at the passenger level.</li><li>Automate performance monitoring with Looker dashboards &amp; explores. Enable deep-dive analysis of voucher effectiveness, channel impact, and passenger engagement.</li></ul><h2>Approach &amp; Solution</h2><h3>1. Data Integration &amp; Flow Design (Dataiku – DSS)</h3><ul><li>Designed and orchestrated a DSS flow leveraging: <ul><li>Redshift tables (passenger, ride, and transactional data).</li><li>S3 tables (bonus allocation &amp; partnership records). Python and SQL recipes for business logic and transformations.</li><li>Spark recipes for processing large datasets at scale. </li></ul></li><li>Created a robust data consolidation layer merging all sources into a single passenger-level fact table, capturing KPIs such as: <ul><li>Registration → Activation funnel</li><li>First Ride → Repeat Ride behavior</li><li>Voucher redemption &amp; usage patterns</li><li>GMV uplift per partnership channel</li></ul></li></ul><h3>2. Data Modeling &amp; Business Logic (LookML)</h3><ul><li>Developed a LookML model exposing the consolidated dataset with business-ready dimensions (channel, campaign, passenger segment, voucher type) and measures (rides, GMV, redemption rate, retention uplift).</li><li>Built Explore 1: a complete analytical workspace for data analysts with all available breakdowns and metrics.</li><li>Designed Explore 2: simplified for business stakeholders, structured for intuitive navigation, ensuring consistency and governance in metric definitions.</li></ul><h3>3. Dashboarding &amp; Monitoring (Looker)</h3><ul><li>Delivered a dedicated Looker dashboard for the Paid Acquisition and Growth teams: <ul><li>Real-time monitoring of voucher usage and redemption.</li><li>Performance comparisons across influencers and partnership channels.</li><li>ROI and engagement tracking, highlighting top-performing campaigns.</li><li>Easy drill-down into passenger-level behaviors and channel contributions.</li></ul></li></ul><h2>Impact &amp; Results</h2><ul><li>Transparency: Unified monitoring replaced siloed, manual reporting.</li><li>Efficiency: Automated flow in DSS + Looker dashboards saved significant analysis time.</li><li>Actionable Insights: Teams could quickly identify high-ROI influencer partnerships and optimize spend.</li><li>Scalability: The DSS + LookML framework became a reusable template for future partnership and voucher campaigns.</li></ul><h2>Key Skills &amp; Tools Demonstrated</h2><ul><li>Data Engineering: Dataiku DSS, Redshift, S3, Spark recipes, SQL, Python.</li><li>ETL &amp; Modeling: Passenger-level consolidated table, business logic design.</li><li>Business Intelligence: LookML development, self-service Explores, automated dashboards.</li><li>Growth Analytics: Funnel analysis, voucher performance, activation-to-retention KPIs, GMV uplift.</li><li>Cross-Team Collaboration: Worked with Acquisition, Growth, and Marketing to align metrics with business goals.</li></ul>

Abdelmalek NOURI

Senior Data Analyst

About me

Work experience

Global Growth Data analyst

Business Data Analyst, BD

Data Analytics Engineer

Delivery & Support Engineer

System Administrator

Transmission Trainee Engineer

Mobile Network Trainee Engineer

Network Trainee Engineer

Education

AMOB University

National School of Telecommunications and Information and Communication Technologies (ENSTTIC)

El Ghazali High School

Portfolio

RFM Driver Segmentation

SAS Ride-Halling

Ambassador Influencers Partnership Contest Trucking– Ride-hailing (Europe - Africa)

SAS Ride-hailing

Talents

Recruiters

About