Why is our financial model wrong when the maths is right?

A model can balance perfectly. Every formula correct, every tab linked, every number tying out, and still be useless, because the assumptions inside it do not reflect how the business actually works.

The spreadsheet cannot tell you this. It does exactly what it is told, with complete confidence, to two decimal places.

The four places models go wrong

Growth that compounds forever

A model assuming 15% month on month growth for three years is not wrong arithmetically. It is wrong because almost no business sustains that rate for that long, and the model has no way of knowing.

Compounding is what makes this dangerous rather than merely optimistic. Fifteen percent monthly is roughly a fivefold increase over a year. Held for three, it is a number nobody in the room believes but everybody has signed off on, because it arrived one cell at a time.

Costs that stay linear

Models routinely understate what scale costs. Headcount is stepped rather than smooth: you do not hire a third of a manager. Acquisition cost rises as you exhaust the cheapest channel first. Support load grows with customers, not with revenue, so a shift toward smaller accounts raises cost per rupee even while revenue improves.

Revenue timing treated as cash timing

This is the most common structural error, and in India it is the most expensive one. A model that shows revenue in the month it is invoiced, with no collection lag, will overstate the cash position by an entire working capital cycle.

Add to that the dues the P&L does not show. GST is payable on invoices raised rather than payments received. Loan principal leaves the bank without appearing as a cost. A model that ignores both can show comfortable cash in a month the business cannot actually fund.

One case, presented as the plan

A single set of numbers is a hope, not a plan. What makes a model useful for deciding is knowing which variable you are most exposed to, and a single column cannot tell you that.

The test that actually matters

A good model is not judged by whether it balances. It is judged by whether its assumptions survive someone who knows your market pushing back on them.

How to make one that holds up

1. 01Put every assumption on one tab, each in its own cell, each with a note saying where it came from. No hardcoded numbers buried inside formulas.
2. 02Ground growth rates in something observable: your own last four quarters, or what comparable businesses have actually achieved. Ambition belongs in a scenario, not in the base case.
3. 03Model collections separately from revenue, using how your customers actually pay rather than the terms on the invoice.
4. 04Run three cases. Not to predict, but to see which assumption the outcome is most sensitive to.
5. 05Check it against reality every month, and write down which assumption moved rather than only how far the total missed.

The adjustments an India-based model needs

Most model templates in circulation were written for a market with different cash mechanics. Four adjustments matter, and leaving them out is the commonest way a technically correct model overstates the cash position.

• ·GST falls due on invoices raised rather than payments received, with the monthly return payable on the 20th of the following month. A model that nets GST against collections will show cash the business does not have.
• ·TDS deducted by your customers means you receive less than you billed and recover the difference against your own tax liability later. Model the receipt, not the invoice.
• ·Payments to registered micro and small suppliers carry a statutory time limit, so those costs cannot be stretched the way a generic model assumes payables can.
• ·Loan principal leaves the bank without appearing anywhere in profit, so a model built off the P&L alone will miss it entirely.

That last step is what turns a model from a document into an instrument. A forecast that is never compared against what happened cannot get better, and the value of the exercise was never the number in the final cell.

About Simplify

Simplify is a finance clarity and investment readiness practice working with founders across India, built on six years inside startups. We write about the questions founders bring before a decision, not after it.