A different kind of HR analyst.

The problem with most HR analytics

Most HR analytics is descriptive. It tells you last quarter's turnover rate, how engagement scores trended this year, which departments have the most open reqs. This is useful for reporting — but it isn't analytics in any meaningful sense. It's measurement without inference.

Real analytics asks different questions: Who is likely to leave in the next six months, and why? What predicts high performance in this role? How much does a bad hire actually cost, and how much would better selection reduce that?

Those questions require models, not dashboards. They require rigor about what we're measuring and why. And they require being honest about uncertainty rather than projecting false confidence.

My influences

My approach is shaped by a few thinkers who changed how I see measurement and uncertainty:

• Douglas Hubbard — How to Measure Anything. The core insight: measurement means reducing uncertainty, not eliminating it. Anything can be measured if you define what you mean. Most things HR claims are "unmeasurable" are just unmeasured.
• Sam Savage — The Flaw of Averages. Plans based on average assumptions are wrong on average. Working with distributions rather than point estimates changes how you see risk and makes decisions dramatically better.
• The IO psychology tradition — validity, reliability, criterion-related prediction. The discipline that figured out how to measure human behavior scientifically, mostly ignored by the rest of the HR world.

What I bring to an engagement

I'm a solopreneur by design. You work with me directly — not a team that hands off your project to junior staff. I bring a combination of:

• Applied statistics and predictive modeling
• Industrial-organizational psychology and psychometrics
• Decision analysis and calibrated estimation
• The ability to ask good questions before picking up a tool