Some examples are:

  • Developing a credit model to predict borrower delinquency and default.
  • Estimating the impact of interventions using causal inference methods.
  • Developing a reinforcement learning1 system for automated and adaptive decisioning.
  • Designing and building A/B testing infrastructure to assist go/no-go product change decisions.
  • Forecasting unit demand for improved inventory management.
  • Developing a dynamic pricing model where prices respond in real time to changes in supply and demand.
  • Estimating customer lifetime value (‘LTV’) to determine spending on marketing channels.

Contact us to explore how we can help you.

Separately, you may be interested in our educational content (see left side, or below if you’re on mobile). We’ve written about a variety of topics to spread an understanding of how these techniques work and foster a discussion of their application.

Footnotes

  1. Reinforcement learning is a nascent technology for commercial applications. Clients should be aware that there is a long trial-and-error period for developing RL systems materially better than simpler operations research alternatives. ↩