Financial Modeling

The problem

Most financial spreadsheets fail the same way. Assumptions get typed directly into formulas, so the interest rate lives in nineteen different cells and changing it means hunting every one of them down.

The result is a model nobody trusts enough to alter, which defeats the purpose. A forecast is only useful if you can ask what happens when the inputs move.

What I built

A set of linked workbooks covering cash flow projection, long-horizon investment growth, and loan amortization, each built on the same discipline: inputs in one clearly marked place, calculations downstream of them, outputs at the end.

Changing a rate, a term, or a contribution amount updates every dependent figure and chart at once, because nothing downstream contains a hardcoded number.

How it works

The amortization model builds a full period-by-period schedule rather than jumping straight to a payment figure, which makes the split between principal and interest visible over the life of the loan. That split is usually the part people have not internalized, and seeing it laid out changes decisions about extra payments.

The investment model compounds contributions across a long horizon and is deliberately built to be run several times at different return rates. A single projected number invites false confidence, so the model is designed to produce a range instead.

Scenario toggles sit at the top of each workbook. Switching between a conservative, expected, and optimistic set of assumptions is one change, and every schedule and chart follows.

What it produced

Models that survive being questioned. Because the assumptions are exposed rather than buried, a disagreement about the output turns into a specific conversation about which input is wrong.

They are also reusable. The same amortization workbook answers the next loan question without being rebuilt, which is the whole difference between a model and a one-off spreadsheet.