🏢 Company Overview
Swiggy is one of India’s largest food delivery platforms, competing with Zomato.
- Founded: 2014
- Cities served: 500+
- Daily orders: ~2.5 million
- Delivery partners: 300,000+
🎯 Business Problem
Swiggy faced 3 major data challenges:
- High order cancellation rate
- Increasing delivery time in peak hours
- Declining customer retention rate
👉 Key question:
How can Swiggy use data to reduce delivery time and improve customer retention?
📂 Dataset Overview (Sample Data for Analysis)
📅 Orders Dataset (1 Month Sample)
| Metric | Value |
|---|---|
| Total Orders | 7,50,000 |
| Completed Orders | 6,30,000 |
| Cancelled Orders | 1,20,000 |
| Avg Order Value (AOV) | ₹320 |
| Peak Hours | 7 PM – 10 PM |
🚚 Delivery Dataset
| Metric | Value |
|---|---|
| Avg Delivery Time | 38 minutes |
| Target Delivery Time | 30 minutes |
| Late Deliveries | 2,10,000 |
| On-Time Deliveries | 4,20,000 |
👤 Customer Dataset
| Metric | Value |
|---|---|
| Total Customers | 2,00,000 |
| Repeat Customers | 1,10,000 |
| New Customers | 90,000 |
| Retention Rate | 55% |
| Churn Rate | 45% |
📊 Data Analysis Performed
1. 📉 Cancellation Rate Analysis
Formula:
Cancellation Rate = Cancelled Orders / Total Orders
= 1,20,000 / 7,50,000 = 16%
👉 Insight:
- High cancellation during peak hours
- 60% cancellations due to late delivery
2. ⏱️ Delivery Time Analysis
- Avg delivery time = 38 min (Target = 30 min)
- Peak hour delay = +12 minutes
👉 Root causes:
- Traffic congestion
- Poor delivery partner allocation
- Restaurant preparation delays
3. 🔁 Customer Retention Analysis
Retention Rate = 55%
👉 Segment Analysis:
| Segment | Retention |
|---|---|
| Fast delivery (<30 min) | 72% |
| Late delivery (>40 min) | 38% |
💡 Insight:
Delivery time directly impacts retention
4. 💰 Revenue Impact Analysis
Revenue = Orders × AOV
= 6,30,000 × ₹320 = ₹20.16 Crore
👉 Loss due to cancellations:
= 1,20,000 × ₹320 = ₹3.84 Crore lost revenue
🧠 Advanced Analytics Used
🔮 Predictive Model (Churn Prediction)
- Features used:
- Delivery time
- Order frequency
- Order value
- App usage
👉 Output:
- High-risk churn users identified = 35,000 users
🧪 A/B Testing
Test: Faster delivery vs Normal delivery
| Group | Avg Delivery | Retention |
|---|---|---|
| A (Control) | 38 min | 55% |
| B (Improved) | 28 min | 70% |
👉 Result:
+15% increase in retention
📈 Key Insights
- 📌 Delivery delay is the biggest driver of churn
- 📌 Peak hour inefficiency causes revenue loss
- 📌 Faster delivery increases repeat orders
- 📌 Predictive analytics helps identify churn early
🚀 Recommendations
- Dynamic Delivery Allocation
- Use AI to assign nearest delivery partner
- Peak Hour Surge Planning
- Increase delivery partners from 7–10 PM
- Restaurant SLA Optimization
- Penalize slow restaurants
- Customer Retention Campaign
- Offer discounts to high-risk churn users
📊 Business Impact (After Implementation)
| Metric | Before | After |
|---|---|---|
| Delivery Time | 38 min | 29 min |
| Cancellation Rate | 16% | 9% |
| Retention Rate | 55% | 68% |
| Revenue | ₹20.16 Cr | ₹24.5 Cr |
Small improvements in delivery time can lead to massive revenue growth.
