The Lab
This is where every hypothesis I've tested is written up, in full. What the claim was, how many citations it ran against, and the result.
Results reflect correlation strength from sample data, not proof of causation.
| # | Claim | Sample | Started | Status | Result |
|---|---|---|---|---|---|
| 2 | Do Google Page-One Rankings Predict AI Citation? Pattern scan | n = 500 | 12 October 2026 | Complete | Confirmed |
| 1 | Does Perplexity Cite the Same Top Sources Twice? Overlap test | n = 500 | 5 October 2026 | Complete | Inconsistent |
Results
Confirmed = held up · Busted = didn't hold up · Partly = true sometimes · Consistent = same sources each time · Inconsistent = different sources each time · Inconclusive = unclear
The Rules
Every threshold is fixed before the test runs. Pattern scans are graded on the gap between groups, and overlap tests on a line set in advance.
Get the Next Result Before Anyone Else
Twice a month I take one AEO claim and test it against a real sample of citations. You get the result and the dataset, straight to your inbox.
One email per experiment. No spam, unsubscribe anytime.