The Idea Graveyard — 135 monetisation ideas and why 46 of them died

Every guide to making money from a creative business is a list of things that worked for somebody. This is the other list.

135 monetisation ideas were generated across three rounds, put in front of 15 inspectors — an engineer, a CFO, a tax lawyer, a consumer-protection officer, a brand editor, a first-customer sceptic — and any one of them could kill an idea outright. 46 died. This is the record of the killing, with all 79 refusals quoted in full.

Why the dead ones are the product

A surviving idea is worth something only to a business shaped like the one it survived in. A dead one carries a reason, and reasons transfer. "Don't do escrow" is useless. "Escrow means holding other people's money, which inverts a payment design where you never touch it" is something you can apply to your own case tonight.

What is in it

• graveyard.csv — all 46 killed ideas, each with its round, its grouping, the pitch as written, and every inspector's verdict verbatim
• survivors.csv — the 36 that lived, with price, grouping and whether the manuscript already existed
• inspectors.md — the 15 lenses, and the real sentence each one kills with
• walls.json — the rejection counts as data
• README.md — how it was run, and what the shape of the graveyard says

The finding you cannot get from a list of successes

43% of all 79 rejections were one wall, said three different ways. Two inspectors who never saw each other's verdicts killed the same eight ideas — when independent lenses converge on the same set, the constraint is real rather than a preference. Reading only what worked would never surface that number.

The method, including where it failed

Each round changed the search axis, because a generator asked the same question twice starts producing synonyms that look like new ideas. Round one's duplicate check reported zero duplicates out of twenty-four — because it compared names, and every name was different. A check that cannot fail is not a check, and that failure is documented rather than tidied away.

Two of the fifteen inspectors were added after a round, because the round before had missed something. The inspection standard was itself an output of the process.

What is deliberately not in it

No revenue projections. The harness produced month-one dollar estimates to keep scoring honest, and they are guesses about ideas that were never run. Shipping a guess inside a product whose claim is "measured" would spend the only thing worth having. Counts, reasons and verdicts are records; the projections stay out.

Who this is for

Anyone deciding how a small creative operation should make money, who would rather read forty-six refusals with reasons than one more list of tactics. The constraints that did the killing were ours — one operator, no third-party sellers, a small catalogue. Read the reasons, not the rulings.

English. Markdown, CSV and JSON — no software to install.

—

Where this sits

• The Prompt Atlas — $49 — the corpus the pack's grammar was derived from — 1,297 shipped prompts
• Six Months of Spend — $29 — what running this standard actually cost: 2,464 clips, itemised
• The whole catalogue — teardowns, references, skill packs and free calculators, with what each one is for