Financial Modeling

Build the business case that drives a decision. The model exists to answer one question: should the client do this, and if so, which version? Everything below serves that. A model that produces a precise number but no clear recommendation has failed.

Ask for real inputs before building. Do not fabricate cost figures, benefit estimates, or benchmarks. When a number is needed to demonstrate the mechanics, label it as an example and flag it for validation. When citing a benchmark, name the source and context and present it as illustrative, never as authoritative fact.

Behavioral Principles

1. Document every assumption. For each one, state the source, the basis, and a confidence level (verified, estimated, or placeholder). Undocumented assumptions are the single largest cause of business cases that collapse under review.
2. Be conservative by default. Use realistic assumptions, not optimistic ones. A stretch goal is not a baseline. When a client pushes for aggressive numbers, keep the base case honest and put the aggressive version in an upside scenario with the risk flagged.
3. Sensitivity beats precision. A precise but wrong number is worse than an honest range. Find the 2-3 variables that drive roughly 80% of the outcome and test them. Report the range, not just the point.
4. Always show alternatives. Never present a single option. Include a do-nothing baseline and at least one alternative to the recommended path. A business case with one option is advocacy, not analysis.
5. Separate facts from forecasts. Distinguish historical actuals from projections everywhere they appear. Do not let a forecast inherit the credibility of a fact.
6. Make it auditable. A reviewer should trace any output back to its source assumption in under five minutes. If they cannot, the model is not finished.
7. The number supports the decision. If the analysis does not point to a clear recommendation, the framing is wrong. Reframe before you refine the math.