I learned more about modelling from one experience than from any other: having my first funding model assessed by an adviser the lender nominated. It showed me things I could improve, and it held up. Both were worth knowing.

‍

The model is not where boards, investors and lenders evaluate the geology or the flowsheet. It is where they judge whether the economic case is coherent, transparent and defensible. That comes down to one question:

‍

Does the model support a decision under scrutiny?

‍

What Due Diligence Actually Tests

‍

Due diligence spans a number of workstreams — technical, legal and market among them. The model review is another, and within it, structure is systematically assessed via actions such as:

• take an output and trace it back to the assumption that produced it
• change one input and check whether everything downstream moves with it
• flex the case to a downside and see whether it stays internally consistent

‍

A model can calculate perfectly and fail all three.

‍

‍

Where Models Come Apart in Review

‍

A number that cannot be traced may point to a hardcode, or a calculation done off-model and pasted back in. An input that does not propagate suggests the connection was never built — change the mining rate and the cost base sits still. A downside case that breaks apart can mean scenarios were layered on top of the model rather than designed into it.

‍

These failures are structural, and they bite when the model is being relied upon.

‍

What Withstands Examination

‍

The model that survives is not necessarily the most detailed. It is the one where:

• every output can be traced to the assumption behind it
• geology, production, capital and cost are genuinely connected
• the case still holds together when it is flexed

‍

Complexity that cannot be explained is a liability, not a feature.

‍

Investment-grade is not a standard a model claims for itself. It is granted by the people who examine it, and it holds only for as long as the model keeps answering their questions.