Research

A transparent research method

How to evaluate sources and keep an evidence trail for AI data market claims.

Sources and guidance · Reviewed October 2026

Prefer original evidence

Read the buyer’s own programme page, the actual agreement where public, a regulator’s guidance, a court filing or a company announcement. Use secondary reporting to locate evidence and context, while keeping its interpretation distinct.

Classify every claim

Tag statements as a disclosed fact, a company claim, an estimate, a reported figure or an editorial recommendation. Record publication and access dates. A page remaining online does not prove a programme remains open.

Make comparisons reproducible

Preserve the unit, denominator, currency, time period and inclusion rules. Never combine an acquisition price, a settlement, a subscription and a per-hour rate into an unexplained average.

What you can download here

The datasets page provides this site’s editorial page catalogue and source list in JSON and CSV. It is not a copy of another publisher’s proprietary research database. The catalogue can be used to audit coverage and navigate the site.

Use this as a preparation framework. It is not legal advice, a valuation or a promise that a buyer will accept your data.