I tried to forecast a business I built from scratch
What happened when I simulated a software company, forecast its revenue two ways, and used the forecast to decide whether to hire more salespeople.
UFC fight forecast
A model that estimates each fighter's chance of winning a UFC matchup, using only what was known before the fight, and shows which factors pushed the forecast one way or the other.
I wanted to see if I could predict a UFC fight
I'd wanted to build a UFC forecasting tool for years. This is how it came together with AI, what I learned about leakage and calibration, and why I stopped trying to improve the model.
What does it actually take to make a sale?
A VP of Inside Sales once asked me what it takes to make a sale. This is what I learned trying to answer that with data, and how AI changed the way I spent my time on the analysis.
What actually drives sales performance?
I built a simulated SaaS company with 13,107 opportunities and looked at which activities and deal characteristics go with winning, and how that changes by customer segment.