This project focuses on building and serving an end-to-end product to forecast sales for Rossmann Pharmaceuticals stores across several cities. The goal is to provide accurate six-week sales predictions, empowering the finance team and store managers with data-driven insights.
The project incorporates analysis of factors such as:
- Promotions
- Competitor Activity
- Weekend and State Holidays
- Promotion Analysis: Evaluate the effectiveness of promotional campaigns.
- Competitor Analysis: Assess the impact of competitor proximity and openings on store performance.
- Customer Behavior Trends: Explore customer activity during weekends, holidays, and store operational hours.
- Seasonality Insights: Identify patterns in sales based on weekend.
The dataset contains the following columns:
- Date: Daily data for each store.
- Store: Unique ID for each store.
- DayOfWeek: The day of the week.
- Sales: Sales for the store on the given day.
- Customers: Number of customers for the store on the given day.
- Promo: Whether a store is running a promotion that day.
- CompetitionDistance: Distance to the nearest competitor store.
- Assortment: Assortment type ('a', 'b', 'c').
- Promo2: Whether the store is part of a continuous promotion.
-
Exploratory Data Analysis (EDA):
- Analyzed sales trends based on store type, promotions, and competitor activity.
- Visualized customer behavior trends during weekends and holidays.
-
Data Preprocessing:
- Merged multiple datasets (store details, sales data, promotional data).
- Handled missing values and ensured consistency in time-series data.
-
Promotion Analysis:
- Evaluated the sales uplift during promotional periods.
- Assessed the effectiveness of promotions on customer engagement.
-
Competitor Impact Analysis:
- Correlated sales with the proximity and opening of competitors.
- Investigated the impact of competition on store performance.
- Promotion Effectiveness: Promotions significantly boosted sales, particularly in stores with extended assortments.
- Competitor Impact: The presence of nearby competitors slightly reduced sales, though city-center stores were less affected.
- Customer Trends: Weekend sales were normal like other days, with promotions.
-
Clone the repository:
git clone https://github.com/helinatefera/10xWeek4.git
-
Navigate to the project directory:
cd 10xWeek4
- Promotion Insights: Refer to the
notebooks/tasx_1.ipynb
notebook for insights into promotional effectiveness. - Competitor Impact: Explore the impact of competitor activity using the
notebooks/task_1.ipynb
notebook.
Contributions are welcome! Please fork the repository, create a feature branch, and submit a pull request.
This project is licensed under the MIT License.