Grounded in your truths
It starts from your model and your data: the actuals, the drivers, the contracts, the documents. Nothing is invented to fill a gap it could have looked up.
Ask anything, including things your model was never built for. The answer is assembled from what you actually have, and every proxy, approximation and assumption it took to get there is named on the face of it.
It starts from your model and your data: the actuals, the drivers, the contracts, the documents. Nothing is invented to fill a gap it could have looked up.
Where something genuinely is not there, it reaches for a proxy or states an assumption and carries on. A question you need answered today gets answered today.
Every proxy, approximation and assumption is named on the answer: what it stands in for, who set it, when. You always know which parts are judgement.
Approximations use your agreed definitions and stay inside your guardrails, so an analysis run in an afternoon reconciles with the one that took a month.
Six months later, in a post-mortem or an audit, the question is not what the number was. It is how you got there. Every step of the derivation is kept, in order, with its basis.
The interesting test is a question your model was not built for.