How we research

How a build-or-kill verdict is made

Every verdict in the product and in the Verdict Library comes from the same pipeline. Three research agents run against a specific idea: one mines Reddit and niche communities for the complaint patterns behind the problem, one maps the competitive field and how defended it is, and one pulls search volume to size expressed demand. The signals are scored against a fixed seven-dimension rubric, and the total lands in one of three bands: build it, look closer, or kill it.

The rubric is fixed on purpose. It means a beloved idea and a boring idea get graded on the same scale, and it means we publish kill verdicts. A validation tool that never says kill is a cheerleader.

What promising, crowded, and trap mean

Idea lists and cost pages carry one of three labels. Promising: real demand, a reachable buyer, and an opening a small operator can defend. Crowded: the demand is real but the field is saturated; you need an edge, not just effort. Trap: the economics quietly fail most people who try it, usually because the people marketing the opportunity earn their money from the entry fee, not the operation.

The trap label is the one competitors will not print, because trap categories (franchises, vending routes, turnkey stores) are where affiliate commissions live. We take none, so we can say it.

Where the money numbers come from

Cost and profit figures on our pages are ranges compiled from public operator reports, industry association data, franchise disclosure documents, equipment vendor pricing, and marketplace listings, cross-checked against each other. We publish ranges instead of single numbers because single numbers on the internet are almost always someone rounding a guess to look authoritative.

Two rules apply everywhere: profit figures assume you pay yourself for the hours you actually work, and every page shows an updated date. If a number ages badly, it gets revised or the page gets pulled.

What we will not publish

No fabricated testimonials, user counts, or star ratings. No affiliate-ranked "best of" lists. No income-claim screenshots. No single-number startup costs presented as fact. Where we recommend against something, the reasoning and the numbers are on the page so you can check the work.

Corrections

Spot a number that is wrong or stale? Email hello@getolune.com with the page and the source. Verified corrections ship within a week and the page's updated date reflects it.