An Analysis of Consumer Behavior, Shopping Methods, and Promotion Engagement
This project dives deep into customer analytics, leveraging a comprehensive dataset to uncover actionable insights for businesses. It identifies distinct customer segments to understand who buys what, how they shop, and their engagement with promotions, providing a holistic view for targeted marketing strategies.
Key Research Objectives:
Identify customer segments based on product purchasing patterns and preferences.
Differentiate customer groups by their preferred purchasing channels.
Categorize customers by their responsiveness to marketing campaigns and promotions.
Skills Demonstrated:
Data Cleaning, Exploratory Data Analysis (EDA), Feature Engineering, Outlier Detection & Handling, Encoding, Feature Scaling, K-Means Clustering, Cluster Interpretation & Profiling, Statistical Analysis, Data Visualization.
Methodology & Context:
This cross-sectional research employs an exploratory and descriptive inquiry into consumer behavior. Using a quantitative and qualitative encoded dataset sourced from primary ad hoc online research, the project systematically segments customers to reveal actionable patterns.
Project Deliverables:
All findings are clearly documented with the Jupyter Notebook. The complete code and project assets are publicly available on GitHub.
This project analyzes a sample dataset of digital ad campaign performance to compare the effectiveness of Facebook Ads (Meta Ads) against Google Ads (AdWords Ads) in driving conversions. Research Objective: To determine if the Facebook Ad campaign is more effective in terms of conversions compared to the Google Ad campaign, utilizing an A/B testing methodology. Skills Demonstrated: Data cleaning, A/B testing concepts, statistical analysis, data visualization, and reporting. Methodology: This exploratory and descriptive quantitative research project utilized primary data from a continuous online data collection method. The longitudinal research design allowed for the comparison of ad campaign performance over time.
All findings are clearly documented, and the code is available on GitHub.
This web application allows users to create and participate in online polls, visualizing the results in real-time. It serves as a practical demonstration of my technical skills in web development and my foundational understanding of market research principles, particularly in online surveys and data collection. Research Objective: To develop a robust platform for efficient online data collection and real-time visualization. Skills Demonstrated: Web Development: Django, Bootstrap, Gunicorn; Data Visualization: Bokeh; Database Management: PostgreSQL; Programming: Python, SQL; Market Research: Online Survey Development, Primary Data Collection, Cross-Sectional Research Design Methodology: This project utilizes an ad-hoc online data collection method, gathering qualitative insights from participants through user-generated polls. The primary research data is collected directly from users interacting with the platform.
Research Context: B2C / B2B | Nature of the Research Enquiry: Exploratory | Status of Source of the Data: Primary Research (Field Research) | Type of Data: Qualitative | Mode of Data Collection: Ad Hoc Research | Method of Data Collection: Online | Type of Research Design: cross-sectional
