Writing

Kahneman's work is routinely used to dismiss human judgment. But that wasn't his conclusion. The real lesson is about knowing when to let data lead — and when to let judgment lead.

Probability distributions are more than mathematical constructs learned in statistics classes. They help make the predictions that decisions are built on — and knowing how to use them is one of the most practical skills a decision maker can develop.

Traditional analytics answers questions about the past. Decisions are about the future. This is the story of how I found the bridge between them — and why it changed the way I work.

When models fail, we blame the forecast. But the real issue is almost always in how we interpret it. There are two distinct layers of uncertainty in any model — and most decision processes only account for one of them.

The debate between data and intuition is the wrong debate. Expert judgment isn't the opposite of rigorous analysis — it's a form of data that can be extracted, structured, and combined with quantitative evidence.

A young manufacturing company faces a decision that could determine its survival. With no historical data to rely on, the question shifts from "What will happen?" to "How confident do we need to be to act?"

Forecasting is expensive — in time, attention, and capital. A simple two-dimensional framework for knowing when to invest heavily in it, and when to just decide and move on.

The law of large numbers works both ways. Group statistics describe what happens across thousands of observations — but they say surprisingly little about the outcome of any one specific case.

The standard weighted pipeline method for valuing a sales funnel has three serious flaws. A probabilistic simulation approach lets PE due diligence teams quantify their assumptions and explore realistic scenarios.

A VC firm can quantify market size and growth benchmarks — but not the one thing its partners consider most important: their judgment about a team's ability to execute. Until now.