As an MBA student specializing in Business Analytics, I've always been fascinated by a simple question: "Why do Uber and Amazon change their prices every minute, while most retailers only change them once a season?"
I wanted to move beyond theory and build a tool that actually does this.
My goal was to create a Dynamic Pricing Engine that acts like a digital analyst—constantly watching demand trends and competitor moves to set the perfect price in real-time.
I deployed this project as a live web application so you can play with the market simulation yourself.
👉 Launch the Live Dashboard Here (If the app is sleeping, give it 30 seconds to wake up!)
Most traditional retailers use "Cost-Plus Pricing" (e.g., It cost $50, let's sell it for $70). The flaw? It's blind to the market.
- Problem A: When demand explodes (e.g., Black Friday), you run out of stock too fast and leave profit on the table.
- Problem B: When a competitor undercuts you by $1, you lose all your customers.
My Solution: An algorithm that prioritizes Revenue over static margins.
I didn't want to use fake numbers. I used the Olist Brazilian E-Commerce Dataset (100k+ real orders) and focused on the "Bed, Bath & Table" category.
Using Python (Scipy), I ran a Linear Regression on the historical sales to find the Price Elasticity of Demand (PED).
My Key Insight: The model revealed a Price Sensitivity of -0.64. Translation: For every $1.00 I raised the price, I only lost 0.64 sales. This proved that customers were loyal enough that we could safely raise prices during surges to increase total revenue.
I built a Python loop that acts as the "Server." It monitors two live signals:
- Demand Pulse: If traffic spikes >10%, the engine activates Surge Pricing (+5%).
- Competitor Watch: If a rival drops their price, the engine intelligently undercuts them by $0.50 to defend market share.
| The Dashboard | The Elasticity Analysis |
|---|---|
 |
 |
| Real-time revenue tracking | Calculating the demand curve |
- Python: The core logic.
- Streamlit: For building the interactive frontend (no HTML/CSS needed!).
- Pandas/NumPy: For data wrangling and simulation math.
- Scipy: For the statistical regression models.
If you want to look at the code or run it on your own machine:
# 1. Clone this repository
git clone [https://github.com/your-username/pricing-engine-demo.git](https://github.com/your-username/pricing-engine-demo.git)
# 2. Go into the folder
cd pricing-engine-demo
# 3. Install the requirements
pip install -r requirements.txt
# 4. Run the app
streamlit run app.py