The Quantium Data Analytics Virtual Experience simulates the real responsibilities of a data analyst in a commercial analytics team. Over approximately 4–5 hours of tasks, the simulation builds practical skills in data preparation, exploratory analysis, experimentation evaluation, statistical testing, and synthesizing insights into a professional report for stakeholders.
Skills demonstrated in this project include:
- Data cleaning and preparation
- Exploratory data analysis (EDA)
- Data visualization
- Statistical analysis and hypothesis testing
- Commercial thinking and insight communication
In this task, I analyzed transactional and customer demographic datasets to identify purchasing behavior trends and derive insights that would inform the supermarket’s strategic plan for the product category.
Activities included:
- Cleaning and validating raw transaction and customer data
- Deriving new features (e.g., product size, brand indicators)
- Segmenting customers based on demographic and purchasing traits
- Generating summary metrics to understand sales distribution
- Creating visualizations to support insights
- Preparing strategic recommendations for the client
The second task focused on evaluating the impact of store layout changes on store performance using controlled experimentation techniques.
Core steps:
- Identifying trial stores and potential control stores
- Computing metrics (sales revenue, customer counts, transaction averages)
- Applying similarity measures (e.g., Pearson correlation, magnitude distance) to select control stores
- Performing comparison analysis to assess whether performance differences were statistically significant
- Interpreting results in a business context
The final task brought together insights from Tasks 1 and 2 into a consolidated report for the category manager.
Deliverables:
- A structured report leveraging the Pyramid Principle framework
- Data visualizations integrated into the narrative
- Clear business context, insights, and tactical recommendations
- Executive summary designed for non-technical stakeholders
This project uses the following technologies:
Primary Languages & Environments
- Python (with pandas, NumPy, Matplotlib, seaborn, SciPy)
- Jupyter Notebook
Other Tools
- Excel (for supplementary exploration & pivoting)
- PowerPoint (for final presentation and reporting)