How we check a claim
1. One concrete claim
We take a number from a popular video title or statement, e.g. "$1,557 per month". If a video makes several claims, we test the most modest concrete one – the one most favourable to the idea.
2. Test plan before data
Before collecting any data, we write down what we measure, for whom, and which result busts the claim. If we change the plan later, we mark it openly as an addendum and say why. Corrections in favour of the claim are allowed and documented.
3. Typical, not top
Gurus show their best month. We report the median: what a typical beginner gets. Top results are shown next to it, never instead of it.
4. Thresholds
| Verdict | Rule |
|---|---|
| BUSTED | typical result under 10 % of the claim, or a kill criterion is met |
| CAVEAT | 10–50 % of the claim, or it works only for a specific group |
| SURVIVED | 50 % or more of the claim for a typical beginner |
5. Data sources
Only public data and official interfaces: marketplace rankings and listings (Amazon best-seller ranks, Etsy Open API), the YouTube Data API, app store and plugin directories, public statistics (e.g. the Federal Reserve). Comments under the claim videos are read and sorted by hand: no results, small results, meaningful results, account problems. We never bypass bot protection and we never publish other platforms' raw data – only our own aggregates.
6. Estimates
Sales and revenue are often estimated, e.g. from best-seller ranks or views. Every check names the model and its limits. Treat our numbers as order of magnitude, not as accounting.
7. Corrections
Found a mistake? Send it to [[CONTACT_EMAIL]]. We fix it publicly in the check's changelog.