hackathon

Code for the DSBA Eleven Strategy Hackathon Group 2

Marketing Campaign Planner (Proof of Concept)

Overview

This project provides a Streamlit-based platform for planning marketing campaigns with personalized recommendations for customers. Users can set campaign parameters, generate recommendations, and access insights on campaign performance.

Note: This is a Proof of Concept (PoC) with limited functionality.

Installation

Prerequisites

• Python (>=3.8)
• Install dependencies:

  pip install -r requirements.txt

Running the App

Run the following command in your terminal:

streamlit run app.py

Important: The app may initially show a streamlit.errors.StreamlitDuplicateElementId error. Simply refresh the page to proceed.

Features

Once the app is running, you will have access to two tabs:

1. Campaign Settings – Upload data and configure campaign parameters.
2. Campaign Overview – View campaign details.

How to Run a Campaign

Step 1: Upload Data

Click “Browse Files” and upload the following CSV files:

• transactions.csv
• stocks.csv
• stores.csv
• clients.csv
• products.csv

Step 2: Configure Parameters

Adjust the following settings:

• Number of Recommendations per Customer
• Recommendation Conversion Rate (%)
  • This defines the expected conversion rate (probability of purchase per recommendation).
  • Example:
    • If set to 100% (1.0) → Each recommended product is expected to be sold. Thus, recommendations cannot exceed stock availability.
    • If set to 10% (0.1) → Each product is expected to be sold once per 10 recommendations, meaning we can recommend it up to 10 times the available stock.

⚠️ Country and Customer Type settings are hardcoded and do not affect recommendations.

Step 3: Generate Campaign

Click “Generate Campaign”. Once you see “Campaign Generated Successfully!”, navigate to other tabs for insights.

Step 4: Restarting a Campaign

To launch a new campaign:

• Refresh the page and re-upload data before generating a new campaign.
• Skipping this step may cause issues.

License

This project is a personal proof of concept intended for evaluation purposes only. It is not licensed for public distribution, modification, or commercial use.

Contact

For questions or contributions, open an issue in this repository.