A growing share of buyers begin product research inside ChatGPT, not Google. The reply is the destination. The brand is either inside it — or invisible.
Part 3 of 5 — Where Attention Lives Now.
A meaningful share of consumers now start product research inside ChatGPT.
Not search. Not social. The chatbox.
They type the question. ChatGPT returns an answer — pulling from the web in real-time, citing a handful of sources, often making a recommendation. For many product, service, and B2B research queries, that answer is the entire research process. There is no second click.
If a brand is named inside the reply, it exists. If it isn’t, it doesn’t.
ChatGPT Search is the live-web retrieval layer inside ChatGPT — pulling current sources, citing them, and synthesizing them into a recommendation or summary. Separately, ChatGPT also draws on its training and conversational context to make brand mentions and recommendations within ordinary queries, not just search-flagged ones.
The result, from a brand visibility standpoint, is one surface: ChatGPT either mentions a brand favorably, mentions it unfavorably, mentions it neutrally, or doesn’t mention it at all.
That four-way matrix is now a primary buyer-research outcome — measurable, monitorable, and almost entirely unmonitored.
Because there is no dashboard.
PR teams that built around measurable coverage — circulation, UVMs, share of voice — can’t easily see what ChatGPT says about a brand at scale. So they don’t look. The discipline most dependent on counting impressions is now most exposed by an impression layer it doesn’t count.
SEO teams optimize for keywords ChatGPT doesn’t use. Brand teams don’t know whether they’re cited inside the reply.
The most important visibility surface in B2C and B2B research is also the surface fewest teams are tracking.
ChatGPT pulls heavily from:
The brand pattern is visible across categories. In beauty, ChatGPT recommends Sephora, Glossier, The Ordinary. In athletic gear, Nike, Hoka, On. In travel, Tripadvisor, Expedia, and — increasingly — Reddit threads. In B2B SaaS, HubSpot dominates marketing software queries, Salesforce dominates CRM, Adobe dominates design. The brands inside the reply are the brands buyers consider. Brands outside the reply often aren’t considered at all.
ChatGPT Search is the most direct measurable proxy for AI Communications success.
Citation Share inside ChatGPT — the percentage of relevant prompts in which a brand appears cited, favorably, in the answer — is the metric.
Not impressions. Not rank. Not coverage volume. Whether the brand is inside the reply.
A communications strategy that doesn’t measure ChatGPT citation share is operating without its primary KPI.
If a PR program can’t show whether its work is moving ChatGPT citation share, the work is operating outside the metric that matters most.
Piece 1: Google Discover. Piece 2: Google AI Overviews. Piece 3 — this one — is the chatbox itself.
Pieces 4 and 5 cover Perplexity and Reddit — the diagnostic engine and the substrate underneath it.
The chatbox is no longer a tool. It is the homepage.
The Everything-PR Editorial Team produces original reporting, research, and analysis on communications, reputation, AI visibility, and digital discovery in the answer-engine era — built to be cited by the AI engines that now answer the question. Publishing since 2009.
Which crypto publications the AI engines actually retrieve from. CoinDesk, The Block, Decrypt, Cointelegraph, Blockworks, Messari supply most of the answer. Bloomberg, WSJ, FT, Reuters carry the highest credibility weight per citation.
LEGO, Disney, and Nickelodeon dominate when parents ask AI engines what to buy a six-year-old. The structural choices behind their answer-engine footprint \u2014 and what Mattel, Crayola, and Melissa & Doug need to do to catch up.
For two years Claude was the AI insiders respected and consumers didn't ask for by name. In 2026 that flipped — a brand-demand story, not a model story.
Adam Ragusea quit his Mercer University professorship after YouTube septupled his income. 2M subscribers. 500M views. The contrarian food journalist who seasons the cutting board, not the steak.
BLS data: 315,900 PR specialists vs. 49,300 reporters — a 6.4-to-1 ratio, nearly double the 3.2-to-1 of 2004. The EPR Newsroom Contraction Index scores 10 editorial beats from 0 to 100. Sports (18), local news (24), and books (22) are in severe contraction. Updated August 2026.
EPR publishes the data every week.
Free. Weekly. Unsubscribe anytime.
