AI-visibility platform — citation tracking, decay, persona testing, agents — and the vendor shipping measurement features fastest on this record.
Nine items, and Profound appears in two roles: the vendor shipping measurement features faster than anyone on this record, and the source of studies with stated methods and large samples. The features: Citation Decay (August 13), tracking week-over-week counts for every cited URL; a YouTube citation breakdown by channel and video (August 19); Adobe Experience Manager nodes for its agents; and a Context Manager for its AI Marketer. The studies: commercial intent in ChatGPT conversations rose from 13.9% to 19.2% in a year across 7.5 million U.S. conversations; inaccuracy rates of 4.3–6.8% across engines in more than 158,000 brand claims, with Claude more than 1.3 times as likely as ChatGPT to produce an inaccurate one; Claude and Claude Code behave as distinct engines, searching the web in over 93% versus 13% of responses; feed-integrated retrieval reached roughly 65% of ChatGPT Shopping recommendations; and across 71,147 persona-varied responses only about one in four brand picks matched across income, age and gender. Every figure is vendor-published; the record carries no outside replication.
10 items on the wire, newest first.
Elsewhere: official
Profound tested ChatGPT, Gemini, and Claude with personas built from combinations of income, age, gender, and occupation, asking each about credit cards…
Profound's analysis of 24,135 responses collected July 13–23 found Claude searched the web in over 93% of responses versus just 13% for Claude Code, and…