We log what could change a decision: price updates, score changes, corrections and changes to the method. Routine wording edits are not listed.
- Rubric v1.1: the "Data, compliance and evidence" ladder was rewritten. It had two faults. It treated a document we could not retrieve as a document the vendor does not publish, which scored vendors down for our own fetch failures; a sub-processor list now only earns credit at level 5, and only when the vendor publishes it openly. And it made named-customer case studies a gate on a security score, which the criterion's own question does not ask about; case studies are still reported on each review page but no longer affect this level. Re-applying the ladder moved Manychat 3 to 5, and HubSpot, Landbot and Tidio Lyro 3 to 4. No other criterion changed.
- Site published with rubric v1.0 and nine products. All prices and limits checked against their sources on this date. Editions: US English, UK English, German, Spanish.
- Reference-bot build test protocol published. Status: not run.
Researched and drafted with AI assistance from the sources listed on this page. We have not built or tested bots on these platforms yet. Method: How we review