03/06/2026
Most models don't break because of a wrong formula.
They break because of bad architecture. And architecture is decided before you type a single equation.
Here are the 6 rules I'd say to every analyst if I could.
1. Separate Inputs, Calculations, and Outputs physically.
Inputs are the assumptions you change. Calculations are the engine. Outputs are what the decision-maker sees. Mix them and you've built a machine where the dials are hidden inside the gears. One input section. Clearly colored. Everything that can change lives there, and nowhere else.
2. No hardcoding inside formulas. Ever.
=A5*1.15 is a landmine. Six months later, nobody knows what 1.15 is or why. Put the 15% in a labeled input cell and reference it: =A5*(1+$B$2). A model should be editable without anyone opening a single formula.
3. One row, one formula, all the way across.
A calculation row should use the identical formula in every period. If column F does something column G doesn't, you've created a silent inconsistency that someone will find at the worst possible moment. Consistency across the row is how you make a model auditable at a glance.
4. Every output traces back to a source in under 10 seconds.
The test: someone points at a number and asks "where did this come from?" If you can't walk them from output → calculation → input in seconds, the model isn't reviewable. And what can't be reviewed can't be trusted.
5. Build the checks before you need them.
A dedicated checks section: does the balance sheet balance? Does the cash flow tie to opening and closing cash? Green means go. Red means stop and find it before the meeting, not during it. A model that can fail silently will fail loudly, in front of the board.
6. Be clear, not clever.
The "genius" formula only you understand is not a strength. It's fragility. The best modelers write formulas a tired colleague can read at 11pm. Clarity is the professional standard. Cleverness is a liability with good PR.
None of these are about Excel skill.
They're about discipline. And discipline is what separates someone who builds spreadsheets from someone who builds models people stake decisions on.
