How we work

We start with the estimand

An estimand is the precise quantity a study sets out to measure: the effect of what, on what outcome, for whom, compared with what, over what period. Writing it down first forces the important decisions into the open, before they are made silently by whatever method or dataset happens to be convenient. Read more on why this matters.

Methods follow the question

We use the method that can answer the question with the data available. That might be a quasi-experimental evaluation, a health economic model, an agent-based simulation or a straightforward descriptive analysis. When the data cannot support a confident answer, we say so, and explain what would.

Uncertainty is reported, not hidden

Every estimate comes with its uncertainty and the assumptions it depends on. Decision-makers need to know not only the best estimate but how much it could move, and which assumptions would move it.

Fixed fees and written scopes

Each engagement begins with a 30-minute conversation and a written scope memo within two business days. The memo states the question, the method, the deliverables, the timeline and a fixed fee. If the scope changes, we agree the change in writing first.

Reproducible handover

At the end of an engagement you receive the report, a technical appendix, and the code, data specifications and documentation needed to reproduce every result. Your team, or a reviewer, can rerun the analysis without us.

Data handling

We work with de-identified or aggregated data wherever the question allows. Data access terms are agreed in writing before any data is shared, and data is held only in environments the client approves and only for as long as the engagement requires.

Independence

We are paid for the quality of the analysis, not for a particular answer. Findings are reported as they are, including findings that are inconvenient. We decline work where we cannot give an honest answer with the data available.