Building Financial Models That Actually Get Used

Most financial models fail. Not because the math is wrong, but because nobody uses them. They sit in folders, opened once for a board meeting, then forgotten. The model that took three weeks to build influences zero decisions.

The models that matter share common traits that have nothing to do with financial sophistication.

Why Models Get Ignored

After building and reviewing hundreds of financial models, we have identified the patterns that lead to abandonment:

They answer the wrong question. The model answers what the analyst found interesting, not what the decision-maker needs to know. A CFO asking whether to enter a new market does not need a detailed P&L projection. They need to understand the conditions under which the decision makes sense.

They cannot be questioned. Complex models become black boxes. When executives cannot trace how an output relates to inputs, they do not trust it. And they should not. Unquestionable models are dangerous models.

They are fragile. Change one assumption and the model breaks. Add a new scenario and you need a new model. Fragile models do not survive contact with reality. Business conditions change weekly. Models need to change with them.

They are too precise. Projecting revenue to the dollar for a five-year forecast is not accuracy. It is false confidence. Excessive precision signals that the modeler does not understand uncertainty, which makes the entire analysis suspect.

What Works Instead

The models that influence decisions share different characteristics:

They start with the decision. Before building anything, define what decision the model should inform. Work backward from there. What would change your mind? What are the key uncertainties? What ranges matter? The model should directly address these questions.

They expose assumptions. Every assumption should be visible, labeled, and easy to change. The model should make clear what it is assuming and what happens when those assumptions are wrong. Sensitivity analysis is not optional.

They communicate uncertainty. Present ranges, not point estimates. Show best case, base case, and worst case. Make explicit what you know, what you estimate, and what you are guessing. Honest uncertainty is more useful than false precision.

They tell a story. Numbers alone do not persuade. The model should support a narrative that explains why the numbers make sense. What are the key drivers? What would need to be true for this outcome? Executives remember stories better than spreadsheets.

Practical Guidelines

When we build models for clients, we follow specific principles:

• One page of outputs. If the key findings cannot fit on a single page, the model is too complex or unfocused. Drill-down detail can exist, but the summary must be clear.
• Three scenarios maximum. More scenarios create confusion, not clarity. Focus on the cases that actually inform the decision.
• Inputs separate from calculations. All assumptions in one place. All logic in another. Never mix them. This makes the model auditable and updatable.
• Round aggressively. Present results in thousands or millions. Remove decimal places. Precision should match confidence.
• Version control everything. Models evolve. Track changes. Know what changed and why. This is especially important when multiple people contribute.

The Test

A useful financial model passes a simple test: could a smart person who has never seen it before understand the key findings in five minutes? Could they change an assumption and see how the conclusion shifts?

If the answer is no, the model is not ready. Complexity is easy. Clarity is hard. The models that shape decisions are the ones that make complex situations understandable.

Your model does not need to be sophisticated. It needs to be useful. Those are different things.