The practice
Advisory
Threshold Data Sciences takes a small number of engagements. The published research demonstrates the method; the advisory work points it at your question.
Applied empirical analysis
Study design, causal inference, measurement, and the data engineering underneath it — answers that hold up to scrutiny, not charts. The method is the product: sources documented, assumptions stated, results reproducible.
Regulatory program analysis
What a mandate will actually produce in a given organization or region — as opposed to what it requires on paper. Grounded in the research program this practice publishes: disclosure rules, reporting requirements, and infrastructure programs, analyzed through the organizational capacity that determines their outcomes.
Governance assessment
Where an organization's rules and its behavior have come apart, and why. Shadow IT, insider risk, and AI governance — the areas where policy travels top-down as a document while behavior travels peer-to-peer as a habit.
Engagement shapes
Two forms, both bounded:
- Fixed-scope assessment. A defined question, a defined period, a written finding. Scoped and priced before work begins.
- Commissioned analysis. Data work on a question you name — sources documented, method stated, and the result delivered whether or not it is the result anyone hoped for.
Findings are what the evidence supports. They are not negotiable, and no outcome is promised in advance.
To start a conversation, use the contact page.