Approach
Work is organized to be robust and review-ready: clear scope, documented assumptions, and reproducible results. The method and delivery level are adapted to the engagement’s risk level and requirements.
Principles
- Traceability: inputs, assumptions, methods, and versions are documented.
- Reproducibility: results can be repeated with the same inputs and tools.
- Plausibility checks: consistency checks, sensitivity, and sanity checks.
- Review tailoring: report/appendix structure adapted for technical review.
Typical workflow
- Scoping – goals, boundaries, data status, assumptions, delivery format.
- Baseline – reference case and verification of data/configuration.
- Analysis – runs, comparisons, parameter studies, interpretation.
- Quality assurance – plausibility checks, sensitivity, peer review when needed.
- Delivery – report + appendices + optional reproducible calculation package.
Low-risk engagements can be compressed into a rapid expert intervention. Higher requirements add verification and documentation.
Quality assurance
QA is adapted to the engagement. When needed, checklists are used for assumptions, unit checks, run logging, and versioning of delivered artifacts.
- Version IDs on deliveries
- Change log (Release Notes) for iterative deliveries
- Structured evidence handling when required
FAQ
What information is needed to start?
Ideally: goals/questions, system/process description, available inputs, and preferred delivery format. Incomplete inputs are fine — a practical scope can be defined in step 1.
Can you work under confidentiality?
Yes. When needed, examples and presentations are generalized, and deliveries are structured so that sensitive data is separated from general methodology.
What do I receive at the end?
Typically a report (PDF/DOCX) with assumptions, method, and results, plus appendices (tables/figures). When relevant, a reproducible calculation package (scripts/notebooks) can be included.
How do you ensure results are reasonable?
Through baseline verification, unit/consistency checks, sensitivity analysis, and explicit assumptions. For higher requirements, peer review and expanded verification can be included.
How long does an engagement take?
It depends on goals, data status, and verification requirements. Often a quick scoping and plan is followed by iterative partial deliveries per step or milestone.
Can work be split into smaller deliveries?
Yes. A common structure is: (1) scoping, (2) baseline, (3) analysis, (4) delivery — with check-ins after each step.