A simple web app that helps coffee shop owners predict their daily sales using machine learning.
Coffee shop owners never know how much money they'll make each day. This makes it hard to:
- Plan how many staff to schedule
- Know how much food to prepare
- Decide when to run promotions
- Manage daily cash flow
My app solves this by predicting tomorrow's revenue based on simple inputs like weather, date, staff count, and expected sales.
Input your details:
- Pick a date from calendar
- Enter weather info (temperature, rain)
- Add operational details (staff count, promotions)
- Estimate product sales (coffee, pastries, sandwiches)
Get instant prediction:
- Shows predicted revenue amount
- Gives advice (good day, slow day, etc.)
- Helps plan staffing and inventory
- Easy date picker - Just click calendar instead of entering complex date info
- Weather consideration - Accounts for rain and temperature effects
- Staff planning helper - See if you need more or fewer staff
- Promotion insights - Test if promotions will boost revenue
- Clean interface - Simple design anyone can use
Performance:
- Trained on 292 coffee shop records
- 94.7% accuracy in predictions
- Average error: only $25
- Most important factor: Coffee sales volume
What it considers:
- Day of week and season
- Weather conditions
- Staff count and operational issues
- Customer satisfaction
- Product sales numbers
- Python - Main programming language
- Streamlit - Web app framework
- Machine Learning - Scikit-learn for predictions
- Deployment - Hosted on Render with GitHub
├── app.py # Main web application
├── coffee_sales_model.pkl # Trained model file
├── scaler.pkl # Data processing file
├── feature_selector.pkl # Feature selection file
├── requirements.txt # Required packages
└── README.md # This file
- Download the project files
- Install Python packages:
pip install -r requirements.txt - Run the app:
streamlit run app.py - Open web browser to see the app
For coffee shop owners:
- Save money on unnecessary staff
- Reduce food waste
- Make better promotion decisions
- Plan cash flow more accurately
Example: Predict Saturday will be busy ($8,450 revenue) → Schedule 4 staff instead of 2 → Better customer service → Actually hit revenue target
Technical skills:
- Building complete machine learning projects
- Creating user-friendly web apps
- Deploying projects to the cloud
- Working with real business data
Business skills:
- Understanding small business challenges
- Creating solutions people actually need
- Measuring real impact on profits
- Designing for non-technical users
- Add more detailed analytics
- Connect with weather APIs for automatic updates
- Support multiple coffee shop locations
- Add mobile phone optimization
- Include seasonal trend analysis
Sankaran S
This project shows how data science can solve real problems for small businesses. Built to help coffee shop owners make smarter decisions with their daily operations.
⭐ If you find this project useful, please star it!