Industries
Finance decisions with inspectable assumptions
A finance team spends much of its year on decisions other people propose: a new system, a hiring plan, a capital project, a price change. Its task is to ask what each one costs, what it returns, and which single assumption the whole case depends on.
Argumont is structured decision support. It does not replace licensed professional advice. It is not investment, tax, accounting, or audit advice.
The assumption that carries the case
Most business cases rest on one or two figures that do nearly all the work. A capital request might rest on labor savings, a pricing change on retention, an acquisition on cross-selling that nobody has yet tied to named customers. Find that figure first, then ask where it came from and what the case looks like at half its value.
Sensitivity analysis does this formally. Move each input across a plausible range and watch which ones change the answer. Usually a handful matter and the rest are decoration, which tells you where the next week of diligence should go.
Return measures and their blind spots
Net present value discounts future cash at a rate reflecting the cost of capital and the risk taken, and most finance texts prefer it for choosing between projects. Internal rate of return is easier to explain but can mislead when projects differ in size or timing. Payback ignores everything after the money comes back, which is exactly where long projects earn or lose their keep.
The common mistake is false precision. A model that reports a return of 14.37 percent from inputs nobody would defend to the nearest ten percent is a story in a tidy format. A range with its drivers named is more honest and more useful to the people approving the spend.
A capital request under challenge
The Decision separates figures the team has confirmed, such as a signed quote, from figures that are estimates, such as savings. The Champion for the request makes its case using the sponsor's numbers. The Challenger presses on the one figure the return depends on, and the engine works the sums exactly from the figures provided.
Inspector marks a vendor's success story as a claim about another company, not evidence about yours. Grandfather asks what cannot be undone if the savings never arrive. When the record cannot separate the options, the Verdict can say No Strong Winner Yet rather than invent a precision the inputs do not have.
The line finance keeps
Controllers, auditors and licensed advisers carry duties that a decision tool cannot share. The room supports internal choices about spending and investment inside a business, while the Human Decision, the accounting treatment and any external reporting stay with the people responsible for them.
Automating a packing line
Operations requests $900,000 to automate a packing line, promising savings of $240,000 a year over a seven year life. At a 10 percent discount rate the net present value is about $268,000, and simple payback comes in under four years. The Challenger asks how low the savings can fall before the project destroys value, and the engine works the answer: about $185,000 a year. The savings estimate assumes two full shifts, yet the plant ran two shifts for only seven months last year. At one and a half shifts the savings fall to $180,000 and the value turns slightly negative. The Verdict favors approval only with a volume condition attached, and names the shift pattern that would reverse it.
Questions
- Which return measure should a finance team rely on?
- Net present value is the usual choice for comparing projects, with payback as a check on liquidity. Whatever the measure, show the range of results across the inputs that matter.
- Can the room build our financial model?
- The room works the sums a question asks for from the figures you provide, and labels each figure as confirmed or estimated. Your full model, your forecasts and your books stay in your own tools.
- Is a Verdict on a capital request an approval?
- No. The Verdict is analysis. Approval follows your own authority limits and is recorded as the Human Decision, with any conditions such as a volume floor.
Bring Argumont a Decision worth challenging.
Start with a paid 60 day pilot on your own real Decisions.