How to Get Your Company Cited by ChatGPT and AI Search: An Earned Media Playbook for B2B Leaders (2026)
To get cited by ChatGPT and other AI search tools, earn coverage in trusted publications, publish data with a clear view, and keep it fresh.

- Muck Rack's May 2026 analysis found earned media behind 84% of AI citations, while paid and advertorial content made up just 0.3%.
- More than half of the journalism that AI tools cite was published in the past 12 months, so cadence matters more than one big hit.
- Releases with an original number and a clear point of view get picked up; awards and hiring announcements rarely do.
- Track a fixed set of buyer questions in ChatGPT, Gemini, Claude and Perplexity every month, and treat the answers as your new share of voice.
To get your company cited by ChatGPT and other AI search tools, you need other trusted publications to write about you, with something worth quoting: original data, a clear expert view, and a steady flow of fresh coverage. Paid placements barely register; earned media does most of the work.
That is the question I kept seeing last week. A Forbes Agency Council piece on October 23 asked how PR builds authority for AI visibility, and trade coverage this week tested Gartner’s forecast that PR budgets will double by 2027. B2B founders and C-level executives are asking me a plainer version: “When a buyer asks ChatGPT who to hire in our space, why aren’t we in the answer?”
In 20+ years of running content and PR, I have watched search change many times. This shift is different because the buyer may never reach your website. The answer is the shortlist. Here is how I approach it, and how my Global Media PR & Influencers service is built around it.
Why does earned media decide what ChatGPT says about your company?
AI assistants answer by summarizing sources they trust, and they mostly trust what independent publications say. Muck Rack’s May 2026 Generative Pulse analysis of more than 25 million links across 17 industries found earned media accounted for 84% of citations in ChatGPT, Claude and Gemini, journalism alone for 27%, and paid or advertorial content for 0.3%.
That last number is the one I show to boards. You cannot buy your way into these answers the way you can buy a search ad. The same Muck Rack report found that across four editions since July 2025, earned media’s share stayed between 82% and 89%, so this is not a one-quarter blip (Muck Rack via GlobeNewswire, May 2026).
Gartner has made the budget argument too. In its Top Predictions to Inform 2026 Comms Strategies, it predicts that by 2027 mass adoption of public LLMs as a replacement for traditional search will drive a 2x increase in PR and earned media budgets (Gartner, 2026). Treat that as a forecast, not a fact, but the direction matches the citation data.
What do the AI tools actually cite?
They cite recent journalism, trade publications, credible data and clear answers to specific questions. Muck Rack found more than half of journalism citations came from articles published in the past 12 months, and journalism citations spread across more than 20,000 outlets, which means niche trade titles count, not just national headlines.
Three details from the same report shape how I plan a program:
| Finding (Muck Rack, May 2026) | What it means for a B2B PR plan |
|---|---|
| ChatGPT cites sources in 96% of responses, averaging about five citations | Only a handful of sources make each answer, so being one of the five matters |
| Gemini cites in 82% of responses (about eight sources); Claude in 55% (about 13 when it cites) | Each tool behaves differently, so track all of them, not just one |
| Industry trend questions pull journalism at more than double the rate of how-to questions | Commentary on where your market is going is your best route into answers |
| Press releases appear 3.5 times more often in best-of queries | Releases still matter for “best X for Y” shortlists, if they carry real news |
How do I get my company cited by ChatGPT? A 7-step playbook
Start with the questions your buyers ask, then earn coverage that answers them in trusted outlets, back it with your own clear page, and repeat on a steady cadence. The steps below are the order I run them in for B2B companies and executives in the US and globally.
- List the 20 questions your buyers ask an AI tool. Write them the way a CFO or VP would type them: “best fractional PR partner for a SaaS company,” “how much does B2B PR cost,” “who are the experts on supply chain risk.” Run each in ChatGPT, Gemini, Claude and Perplexity. Record who is named and which sources are cited. That is your baseline.
- Map the cited outlets. You will usually see the same trade publications, analyst blogs and a few national outlets repeat. Those are your target list, ahead of any vanity title.
- Build one piece of original data per quarter. A short survey, an anonymized benchmark, or a cost index. AI tools and reporters both need numbers to quote. Lead with the number and your view on what it means.
- Pitch commentary on trends, not features. Since trend questions pull journalism at twice the rate of how-to questions, put your executive forward on where the market is heading. Reporters need sources who will say something specific.
- Publish contributed bylines in trade titles. Bylines let you set the framing when other coverage is thin. Aim for one a month in a publication on your cited-outlet map.
- Make your own site the clean reference. A clear “about,” service pages that answer cost and fit questions directly, and FAQ sections. Check that your site does not block AI crawlers. My free website audit template covers the basics.
- Keep the cadence and re-test monthly. Re-run the same 20 questions every month. Count how often your company is named and cited. That is your new share of voice.
Which press releases actually get picked up by AI tools?
Releases built around a clear point of view on an economically meaningful number get picked up. Announcements about awards, new hires and routine company news rarely do. Write each release so the headline number and your interpretation could be quoted on their own.
One agency’s own quarter-long test is useful here, with the caveat that it is a single company’s experiment in one sector. LeadCoverage published one release a week for a quarter and reported that releases leading with a view on a relevant number earned 3.5 times more AI citations than those that didn’t, while personnel and award releases earned almost none (LeadCoverage via GlobeNewswire, July 2026). That matches what I see: reporters and AI tools reward the same thing, which is something worth repeating.
A quick test before you send any release:
- Is there a number in the first two lines?
- Would a reporter quote the second sentence without editing it?
- Does it answer a question a buyer would actually type?
- Would it still be useful to read six months from now?
If the answer to two or more is no, it is an internal announcement. Post it on LinkedIn and save your media relationships for the stronger story.
What role should the CEO or founder play?
The executive is the most quotable asset a B2B company has. AI tools often answer “who is the expert on X” questions by citing people who are quoted in trusted outlets and who publish consistently under their own name, so the CEO’s visibility directly shapes the company’s.
In practice that means three things. First, a short list of topics the executive owns and will comment on quickly. Second, a consistent LinkedIn presence that repeats and extends the media coverage; the Forbes piece I mentioned makes the same point. Third, a profile page on your site that states the executive’s expertise plainly, so AI tools have a clean source to match against coverage. I wrote about the day-to-day side of this in Executive LinkedIn Management, and it is the core of my Executive Visibility work.
Influencers and creators help too, particularly on LinkedIn, where industry voices are often quoted in trade coverage. My guide to B2B influencer marketing on LinkedIn covers how to choose and brief them.
How do I measure whether this is working?
Measure share of answer: how often your company is named, and how often your sources are cited, across a fixed set of buyer questions in each AI tool, every month. Pair it with branded search, direct traffic and inbound leads that mention AI tools. Rankings and clicks alone will undercount the effect.
A simple monthly scorecard:
| Metric | How to collect it (free) | Healthy direction |
|---|---|---|
| Named in answers (out of 20 questions x 4 tools) | Manual test, logged in a spreadsheet | Rising quarter over quarter |
| Cited sources that are about you | Click through the citations in each answer | More earned, fewer outdated |
| New earned placements in mapped outlets | Your media log | At least two to four a month |
| Branded search and direct visits | Google Search Console, Google Analytics | Rising |
| Leads who say “I found you through ChatGPT” | Add the option to your contact form | Rising |
You can plan the cadence in my free content calendar template, and the rest of my free tools are in the marketing toolkit.
What mistakes should I avoid?
The common mistakes are chasing volume, buying sponsored placements, and treating this as a one-off campaign. Flooding the wires with thin releases irritates reporters and earns nothing durable. Sponsored content barely registers in AI citations. And one big hit fades, because AI tools favor recent coverage.
I would add one more: letting your own site contradict your coverage. If a trade article calls you “the B2B PR partner for logistics firms” and your homepage says “full-service agency,” AI tools have two stories to choose from. Keep the positioning consistent everywhere.
Where to start this week
Run your 20 buyer questions through the four main AI tools today and write down who gets named. That one hour will tell you more about your PR priorities than any pitch deck. If you want help turning it into a 24/7 client acquisition engine built on earned media, content and executive visibility, see my Global Media PR & Influencers service or book a free discovery call.
Sources: Muck Rack, “What Is AI Reading?” Generative Pulse findings, via GlobeNewswire (May 2026); Gartner, Top Predictions to Inform 2026 Comms Strategies (2026); LeadCoverage, “PR Drives AI Citations, Traffic & Search Visibility,” via GlobeNewswire (July 2026); Forbes Agency Council, “AI Visibility Starts With Authority: 7 Ways PR Can Build It” (October 2026).
Frequently asked questions
Can I pay to get my company cited by ChatGPT?
Not in any reliable way. Muck Rack's May 2026 analysis of more than 25 million cited links found paid and advertorial content made up 0.3% of AI citations. Earned coverage, journalism and credible third-party sources carry the weight.
How long does it take for PR coverage to show up in AI answers?
It varies by tool and topic. Coverage that tools can crawl tends to surface within weeks for news-style questions, but a steady monthly cadence over two to three quarters is what moves the answers buyers see for category questions.
Do press releases still matter for AI search?
Yes, when they carry real news. Muck Rack found press releases show up 3.5 times more often in best-of queries than elsewhere, and one agency's test found releases built around an original number were cited far more than announcements about awards or new hires.
Is this different from SEO?
It overlaps. Your own site still needs clear, crawlable answers, but AI tools lean heavily on what other trusted sources say about you. SEO gets your page found; earned media gets you named in the answer.