The purpose of this repository is to build reproducible data mining pipelines using analytics data made available from the Google Merchendise Store. More information on the BigQuery source data is available here.
Topics include:
- Data driven motivation for big-picture marketing efforts
- Justify marketing spend and be accountable for wins and losses
- Inform when and which products to promote with product-specific forecasts
- Anomaly detection to identify and fix issues fast
- Use knowledge of prior purchases for targeted marketing campaigns
- Identify promotional bundle opportunities
- Implement / audit realtime recommendation engine
- Inform UX design
- Visual representation of campaign success in terms of user acquisition
- Compare long-term value of different marketing campaigns you have run in the past
We are also exploring topics as follows on our develop branch. We encourage pull requests! Please fork our code if you are interested in contributing to this project.
- User / session clustering
- Page sequence associations
- Feature importances for converting segments
- Predictive conversion models
- Graph visualizations