Getting Started
Quick Start Guide
Business Value in 30 Seconds
Installation
Your First Analysis in 3 Steps
Step 1: Prepare Your Data
Step 2: Run the Analysis
Step 3: View Results
What You’ll Get
Next Steps
Sample Output
User Guide
Business Value & Positioning
What Problem We Solve
Why This Matters for Your Business
Where GeoLift Fits in Your Measurement Stack
Investment & ROI
Overview
The 3-Step Workflow
Step 1: Find a Fair Comparison (Donor Evaluation)
Data Requirements
Running Donor Evaluation
What to Look For
Step 2: Check if the Test is Strong Enough (Power Analysis)
Power Analysis Outputs
Step 3: Measure the Lift (GeoLift Analysis)
Understanding Your Results
Key Metrics Explained
Causal Impact
Statistical Significance
Business Impact
Interpreting Results
Visual Diagnostics
Data Preparation Best Practices
Data Quality Requirements
Common Data Issues
Data Validation
Configuration Options
Analysis Parameters
Advanced Options
Reporting and Export
Generate Business Reports
Custom Reporting
Troubleshooting Common Issues
Poor Pre-Period Fit
Low Statistical Power
Implausible Results
Next Steps
Frequently Asked Questions
Business Questions
Q: What problem does GeoLift solve?
Q: Why is this important for my business?
Q: Is this the only measurement tool I need?
Q: What’s the cost and expected ROI?
Q: How does the process work?
Getting Started
Q1: What data do I need to run a GeoLift analysis?
Q2: How long should I run my campaign to get reliable results?
Q3: How many control markets do I need?
Analysis Issues
Q4: My pre-period fit looks poor. What should I do?
Q5: My results show no significant effect. What went wrong?
Q6: The effect size seems too large to be believable. Is this normal?
Interpretation
Q7: How do I interpret the confidence intervals?
Q8: What’s the difference between absolute and relative lift?
Q9: How do I calculate ROI from the results?
Q10: When should I trust the results vs. be skeptical?
Need More Help?
Technical Reference
API Reference
Core Classes
GeoLiftAnalyzer
Methods
DonorEvaluator
Methods
PowerCalculator
Methods
Result Classes
GeoLiftResults
Attributes
Methods
DonorResults
Attributes
Methods
Utility Functions
Data Validation
Configuration Management
CLI Commands
Basic Analysis
Power Analysis
Donor Evaluation
Configuration Schema
YAML Configuration
Error Handling
Common Exceptions
Advanced Usage
Custom Synthetic Control
Batch Processing
Custom Inference Methods
Methodology Comparison: GeoLift (SparseSC) vs. Google’s GeoX (GBR/TBR)
Understanding Google’s GeoX (GBR/TBR) Approach
Why GBR/TBR (Older GeoX-style) Can Be Suboptimal Compared to SparseSC
Conclusion
Advanced Topics
Mathematical Background
Synthetic Control Method
SparseSC Enhancement
Statistical Inference
Bootstrap Inference
Placebo Inference (Permutation Test)
Advanced Configuration
Custom Donor Selection
Multiple Treatment Cohorts
Regularization and Model Selection
Performance Optimization
Large-Scale Analysis
Memory Optimization
Caching and Persistence
Production Deployment
Azure Batch Processing
Docker Deployment
API Service
Custom Extensions
Custom Inference Methods
Custom Outcome Models
Debugging and Diagnostics
Advanced Diagnostics
Sensitivity Analysis
Research Extensions
Difference-in-Differences Integration
Machine Learning Integration
GeoLift
Index
Index