Break-even under uncertainty

Most break-even tools ask you to predict your sales. This one asks what has to be true for the business to cover its costs, and how much of that you actually know.

Fact. It's on a quote, lease, or contract, or it's your own decision. One number.
Evidence. Your own records, your own count, or industry data you checked. A range.
Guess. Your estimate, or anything from AI or the web you haven't checked. Testing it is the work.

You don't sort these yourself. Tell the tool where each number came from, and it sorts them.

How this tool thinks

A product idea isn't a business. A business is the system that makes selling it repeatable, and most of what sinks a new one is a constraint nobody looked at in time. So this tool doesn't ask you to predict your sales. It works out what has to be true for the business to cover its costs, keeps track of where every number came from, and points you at the guess that matters most.

There are no odds anywhere, on purpose. Most early numbers are guesses, and putting probabilities on guesses makes them look like knowledge. When a result looks too good, the tool treats that as a sign of a missing constraint, not as good news.

One tool in a small business discovery toolkit in development. See the rest of my work with AI.

$
$

Decide this before you open. It becomes your stop point.

Monthly fixed costs

What you pay whether or not anyone buys. Click a name to rename it.

How the business earns

Limits

What has to be true

Volume needed to break even Volume you expect Capacity

Cash over 18 months

Cash before taxes and before reinvesting anything.

The good and tough cases put every range at its favorable or unfavorable end at once. They mark the edges of what your numbers allow. They are not predictions, and nothing here assigns odds.

Good case Middle Tough case Stop point

Question the good case

A result that looks too good usually means a constraint is missing. For the good case to happen, all of this has to be true at the same time:

    What it hinges on

    Each bar swings one range from its unfavorable end to its favorable end while everything else stays in the middle. The measure is your lowest cash point. Facts don't appear, because they don't move.

    Red-team it with AI

    Copy this into any free AI tool (ChatGPT, Claude, Gemini, Copilot). It carries your numbers and where each one came from, and it tells the AI not to turn your guesses into fake certainty.

    When the answer comes back

    • If it gives one precise number for something that's still a guess, ask what the number is based on.
    • If it cites a statistic, ask for the source and check it yourself before you use it.
    • If it says the plan looks great, ask for the three most likely ways it fails.
    • If it points you to real data for one of your guesses, check the data, then change where that number comes from.