This project was part of a Data Science Bootcamp at WBS Coding School.
Base of the project was a collection of data (e.g. orderlines, products, brands) from a real but anonymized company. To make things simpler, it shall be called Eniac from now on.
The project was structured in the following way:
- Data preparation and cleaning
The provided data had a lot of inconsistencies, sometimes missing information etc. So the first part of the project was to provide a consistent and reliable dataset.
- Data analysis
The now reliable dataset was analysed regarding customerbehaviour. Goal was to find out if and how strongly discounts on product prices boost revenue and/or sales.
- Cluster products
Products were clustered using a product-type attribute already present in the data.
Further clustering was done using Google Gemini AI controlled via Python code.
- Discounts, revenues and sales over time
Product-clusters were analyzed over time to get an overview and find out general patterns. - Find correlations between discounts and revenue/sales
Linear-regression models between discounts and revenue/sales were setup per product-type.
- Cluster products
- Present results
Results of the data analysis were ordered by importance and impact on the company. Finally results were presented using Google Slides.
The following shows some images from the final presentation.
See the full presentation over on [Google Slides].
