About Failure Forecast

Why it exists

Most advice around a new idea answers the wrong question. Friends, forums and general-purpose AI chat tools are good at answering “is this a good idea?” — a question that rewards optimism and produces encouragement. Encouragement feels like progress, but it does not tell you where the idea breaks.

The more useful question is “what could kill this?” That question is harder to ask about your own idea, because by the time you are ready to build, you are already invested in the answer being “nothing.” Failure Forecast exists to ask it on purpose, in a structured way, before money and time are on the line.

Who it is for

It is built for anyone who is about to commit real time, money or reputation to an idea that has not yet been tested — not only venture-style startup founders.

• People about to buy inventory or place a first stock order
• People about to pay a developer or agency to build software
• People about to record a course or launch a paid community
• People about to sign a lease or take on fixed costs
• People about to leave a job, or spend months of evenings, on one idea
• Operators inside a company proposing a new product or service line

Product philosophy

• Failure first. The default output is the failure story, not a pitch deck. Optimism is easy to find elsewhere.
• Assumptions over opinions. An idea is a stack of assumptions. The job is to surface the ones that are load-bearing and untested.
• Decisions, not documents. Every report ends in something you can act on: a test to run, a signal to watch, a threshold that would make you stop.
• Honest limits. The analysis is decision support. It does not know your market better than you do, and it says so.

If you want the mechanics rather than the intent, see how the Failure Forecast methodology works.

How the guides are reviewed

The written guides on this site are drafted in-house, then reviewed before publishing against a fixed editorial standard: every claim must be either observable, testable or clearly labelled as opinion; no invented statistics; no fabricated case studies; and every recommended test must be cheap enough to run before committing money. Guides are revised when the product or the advice changes, and the published date is kept on the page.

Who made it

Failure Forecast was created by Daniel McKee, who writes and maintains the product and its guides.

Questions, corrections or feedback: hello@failure-forecast.app.