Free tool · runs in your browser · nothing stored
Financial model health check
Twelve questions about how your model is built. An honest answer takes three minutes and tells you whether it survives diligence.
- 01
Are the profit and loss, balance sheet and cash flow statement fully linked?
- 02
Does the balance sheet tie to zero without a plug or balancing figure?
- 03
Are all inputs in labelled cells, with no constants typed inside formulas?
- 04
Are inputs, formulas and cross-sheet links visually distinguished (for example blue, black, green)?
- 05
Is there a checks tab that fails loudly when something breaks?
- 06
Is revenue built from drivers (volume, price, conversion, retention) rather than a growth rate?
- 07
Are churn and expansion applied to the correct cohort base rather than to total revenue?
- 08
Is operating expense built from a role-level headcount plan with start months?
- 09
Do receivables, payables and inventory move with volume on real terms?
- 10
Is there a single scenario switch driving base, upside and downside?
- 11
Is the model free of circular references and iterative calculation?
- 12
Is there a version log recording what changed, when and why?
Your score
—
0 of 12 answered
Full PDF report
All twelve items with a recommended fix and a worked example for each, plus the Investor-Ready Model Checklist.
What this check measures
Not whether your forecast is right — nobody can assess that from twelve questions. It measures whether the model is structurally trustworthy: whether outputs derive from inputs, whether errors surface, and whether someone other than the author can operate it. Those three properties are what a diligence team tests in the first hour, and they are largely independent of how good the business is.
How the score is weighted
The twelve questions carry different weights, from five to twelve points, reflecting how much damage each failure does in practice. Statement linkage, a balance sheet that ties without a plug, and a driver-based revenue build carry the most weight because failing any of them makes every downstream number unreliable. A missing version log matters, but it does not invalidate the arithmetic.
The score bands are calibrated against what actually happens in a process. 85 and above generally passes diligence with minor comments. 65 to 84 generates findings but survives. 40 to 64 is where processes stall while the model is reworked under time pressure. Below 40 is usually cheaper to rebuild than to repair, and an honest audit will say so.
The three risks you get back
Rather than listing everything you answered no to, the output ranks failures by weight and returns the three that matter most. Fixing three things is an action; fixing eleven is a project nobody starts. The full PDF report covers all twelve with a recommended fix and a worked example for each.
Nothing is stored
Your answers never leave the browser. There is no submission, no account and no retention — the score is computed locally as you answer. The email field exists only for the PDF report, and it is used once.
Two ways to start
Book the call, or start with the checklist.
If you know what you need, book the scoping call. If you are still deciding, take the checklist investors effectively run your model against and see where it stands.
Book a 30-minute scoping call
Thirty minutes on what exists, what it needs to do, and who has to be convinced. A fixed-scope proposal follows within 48 hours.