Lyufa Z.

Key Skills

Software

Other

Top Subject Matter

FMCG marketing analytics

customer segmentation and conversion prediction

FinTech customer analytics

Top Data Types

Text

Document

Top Task Types

Classification

Freelancer Overview

Data Analytics Consultant (market research analytics; predictive modeling and classification for marketing campaigns). Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include N, A, and Other. Education includes Master of Science, Lancaster University (2014) and Bachelor of Science, University of the West of England (2013). AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Classification, Computer Programming, and Coding.

Labeling Experience

Data Analytics Manager (growth analytics with experimentation and KPI optimization)

Classification

Led analytics initiatives focused on optimizing conversion, repurchase, and marketing effectiveness through data-driven experimentation and predictive analysis. Mapped customer journeys, identified bottlenecks, and designed A/B tests to lift conversion without increasing ad spend. Used cohort and metric-based decision frameworks to define goals, build dashboards, and iterate based on observed performance. • Analyzed payment cycles and reallocated channel budgets based on marginal revenue impact to improve cash-flow metrics. • Designed and evaluated A/B tests for Xiaohongshu payment conversion using funnel and journey analysis. • Conducted cohort analysis to identify repurchase drivers and created targeted re-engagement campaigns. • Established KPI-driven decision frameworks with dashboard tracking to guide continuous optimization.

2021 - 2025

Managing Data Specialist / Data Scientist (predictive modeling and data validation for financial products)

Other

Delivered data science solutions requiring data preparation, validation, and predictive modeling for financial services use cases. Built logistic regression and random forest models in Python for conversion improvement, including data preprocessing, feature selection, and model diagnostics. Ensured data integrity by validating migrated datasets with SQL and creating reconciliation reports for compliance and reporting. • Developed logistic regression (interpretability) and random forest (accuracy) models to predict second-loan likelihood. • Performed end-to-end data processing: extraction, cleaning, transformation, outlier handling, and imputation. • Led SQL-based data validation and reconciliation across 10M+ records for asset-backed securities readiness. • Visualized findings and presented to stakeholders using Power BI and Tableau while managing delivery progress.

2017 - 2020

Data Analyst (customer segmentation and cohort analysis for targeting)

Classification

Performed customer segmentation and cohort-based analytics to identify high-value groups and optimize acquisition channels. Used clustering to segment customers and SQL cohort analysis to track retention across channels, then recommended strategy changes based on outcomes. Supported marketing campaign design and evaluated response and retention metrics. • Clustered customers with K-means (R) to find high-value segments for targeted messaging. • Collaborated with marketing to run personalized text/SMS campaigns and measure response. • Conducted SQL cohort analysis to compare retention across acquisition channels. • Identified underperforming channels and recommended adjustments to improve 30-day retention.

2016 - 2016

Data Analytics Consultant (market research analytics; predictive modeling and classification for marketing campaigns)

Classification

Provided predictive modeling and customer propensity classification to support marketing and business decision-making for FMCG clients. Built logistic regression models and regression models to estimate sales outcomes and customer likelihoods, then validated results against post-campaign or tracking data. Translated model outputs into actionable channel and targeting recommendations for senior stakeholders. • Modeled customer propensity (logistic regression) for Starbucks campaign targeting. • Built regression model (SPSS) to predict sales across city-category combinations for P&G channel expansion. • Produced market share tracking and reporting synthesis to support segmentation-level marketing adjustments. • Presented results for stakeholder approval and adoption into multi-million-dollar plans.

2015 - 2016

Education

Work History