// data desk — 2026-08-16
A synthesis of four peer-reviewed studies on how AI search decides what to cite — covering visibility instability, the discovery gap, content structure, and the earned-media bias. Every figure sourced and dated.
// gaming data desk
Compare six CS2 trading and marketplace platforms, with source-checked bonus and promo-code information.
We compared the top CS2 case-opening sites on the numbers that matter — bonus value, minimum deposit, withdrawal speed, user base and age. Key-Drop leads on bonus size with code LOOT. Full data table, verified promo codes, and the house-edge math.
A 112-product study found AI models recognize products by name 99% of the time but surface them in category searches just 3–8% of the time — and on-page GEO optimization did nothing to close the gap.
A St. Gallen study tracked four AI engines for six weeks and found the sources they cite overlap only 34–42% between consecutive days. AI visibility is a distribution, not a ranking — and needs 7–8 measurements to pin down.
A six-engine study held meaning constant and changed only document structure — headings, chunking, tables. Citation rate rose 17.3%, with heading hierarchy alone responsible for nearly half the gain.
A p-value is not the probability your result is true, and 'not significant' is not 'no effect.' A plain-language guide to what a p-value actually means, the traps, and how to check significance yourself.
Technology
Adoption curves, benchmarks and the measurable reality behind the hype.
Markets
Technical and quantitative analysis grounded in published price and flow data.
Science
What the peer-reviewed literature actually found — read past the press release.
Economy
Macro indicators, labour data and the numbers governments publish.
Methods
How to read a statistic: significance, sampling, bias and uncertainty.
Gaming
The economics and data of games — item markets, odds, RTP and platform metrics.