Your buyers ask ChatGPT before they ask your sales team

Picture a head of operations at a 200-person company. Her team spends too long on support tickets, and her CEO wants to know what AI can do about it. Before she opens a single vendor website, she types a question into ChatGPT: what are the best AI tools for customer support for a mid-sized B2B company?

In about ten seconds she has a list of five products, a sentence on each, and a view on which one fits. That list is now her starting shortlist. If your product isn’t on it, you may never hear that she was looking.

This is how a growing share of B2B software gets bought, and the numbers have moved quickly.

The research moved into the chat window

In March 2026, G2 surveyed 1,076 B2B software buyers. 51% now start their research in an AI chatbot more often than in Google, up from 29% a year earlier. 71% use a chatbot at some point in their research.¹²

Other surveys land in the same place with different numbers. 6sense found that 94% of buyers used large language models somewhere in their buying process.³ TrustRadius, surveying 1,862 tech buyers in January 2026, put it at 63%.⁴ The spread comes from different samples and different questions. The direction is the same in all three.

The scale behind it is enormous. OpenAI reported 900 million weekly ChatGPT users in February 2026.⁵ Google says AI Overviews reach more than 2.5 billion people a month, and its newer AI Mode more than a billion.⁶ Even buyers who never open a chatbot now read an AI-written answer at the top of their search results.

The shortlist forms before you hear about it

Buyers have done most of their deciding before they talk to anyone. 6sense’s 2025 research, built on nearly 4,000 buyers, found:³

• Buyers make first contact with a vendor about 61% of the way through their journey.
• They pick a winner from their day-one shortlist 95% of the time.
• The vendor they favoured before contacting anyone won 77% of deals.

Gartner adds that 61% of B2B buyers prefer an overall rep-free buying experience.⁷ TrustRadius found that 83% of buyers shortlisted three products or fewer.⁴

Put those together and the shortlist is where the deal is won. More and more, an AI assistant is helping write it.

For a startup, there’s good news inside the G2 data. 69% of buyers said a chatbot recommendation led them to a different vendor than they had planned on, and one in three bought from a vendor they had never heard of before.¹ AI answers reward the products that are described well and mentioned often. A two-year-old startup can earn that faster than it can earn a brand.

Fewer clicks, better visitors

The cost of all this lands on your website traffic.

Pew Research tracked the real browsing of 900 US adults. When Google showed an AI summary, people clicked a result on 8% of visits, compared with 15% when it didn’t. They clicked a link inside the summary itself on just 1% of visits.⁸ Ahrefs estimates that an AI Overview cuts the click-through rate of the top-ranked page by about 58%.⁹ SparkToro’s data shows 68% of US Google searches now end without a click to any website.¹⁰

The visitors who do arrive from AI tools tend to be further along. Ahrefs found that AI search sent just 0.5% of its own site’s traffic but 12.1% of its signups.¹¹ Semrush estimated that the average visitor from AI search converts at 4.4 times the rate of an organic search visitor.¹² Methods vary, so treat the exact multiples with care. The pattern holds: buyers who come from an AI answer have already been told you might be the right fit.

What AI assistants actually cite

If the answer decides the shortlist, the useful question is what shapes the answer. Three findings stand out.

Mentions matter more than links. Ahrefs studied 75,000 brands and how often they appeared in Google’s AI Overviews.¹³ The strongest signal was how often a brand was mentioned across the web.

Signal	Correlation with AI Overview mentions
Branded web mentions	0.664
Branded anchor text	0.527
Branded search volume	0.392
Domain Rating	0.326
Backlinks	0.218

“Best of” lists carry a lot of weight. In a study of 750 ChatGPT prompts asking for software, product and agency recommendations, Ahrefs found that “best X” list articles made up 43.8% of the pages ChatGPT cited. About a third of those lists sat on low-authority sites.¹⁴ A small publisher with a well-made comparison page can shape what a model says.

Fresh pages win. Across 17 million citations, the pages ChatGPT, Perplexity, Gemini and Copilot cited were about 26% newer on average than the pages ranking in normal search results.¹⁵

The usual community and reference sites matter too. Reddit and Wikipedia are among the domains ChatGPT cites most often, and Semrush found LinkedIn content in about 11% of AI answers in its 2026 study.¹⁶¹⁷ G2 reports that buyers see review-site citations as the strongest trust signal inside an AI answer.¹

A plan for a B2B AI startup

None of this needs a big team. It needs a short list of buyer questions and the discipline to work on them every month.

1. Run the prompts your buyers run

Write down the 20 questions a buyer would ask before choosing a product like yours: best AI tool for X, X vs Y, alternatives to [the market leader], is [your category] worth it for a company our size. Ask them in ChatGPT, Perplexity, Gemini and Claude once a month. Log who gets named, how each product is described, and which pages get cited. That log is your new keyword report.

2. Get mentioned where the models read

The cited pages in that log are your target list. Usually that means a few “best X” roundups, review sites like G2, Reddit and community threads, LinkedIn posts, and comparison articles. Pitch the roundup authors with a clear reason to include you. Ask happy customers for reviews. Have the founder answer real questions in the communities your buyers use, under their own name.

3. Publish the pages buyers ask about

Models can only describe what they can find. Give them clear, current pages for the questions in your log: an honest comparison page for each main competitor, use-case pages in your buyers’ language, an explanation of how pricing works, and integration docs. Put a date on them and update them. Freshness is a ranking signal now.

4. Say the same thing everywhere

If your site calls you an “AI agent platform”, your LinkedIn says “workflow automation” and your directory listings say “chatbot”, a model has three stories to choose from. Pick one positioning sentence and repeat it on your homepage, founder profiles, review listings and press. Consistency is how a model learns what you are.

5. Measure it

Add an AI option to your “How did you hear about us?” field. Track referral traffic from chatgpt.com and perplexity.ai. Re-run your prompt log monthly and count how often you’re named. TrustRadius found that 94% of buyers who use AI fact-check its answers at least some of the time, and that demos, free trials, prior experience and user reviews rank among the most influential resources when they choose a vendor.⁴ So make sure every claim an AI repeats about you holds up when a buyer clicks through.

What this means

The first sales conversation for your product now happens without you, in a chat window, and it ends with a shortlist. The work is making sure the answer names you, describes you the way you’d describe yourself, and sends the buyer to a page that confirms it.

A note on the sources: G2 sells review-based marketing and 6sense sells buyer-intent software, so both have an interest in these findings. I’ve leaned on them because their samples are large and recent, and checked them against Pew, Gartner, TrustRadius and Ahrefs wherever I could.

Footnotes

1. G2, “The Answer Economy” (April 15, 2026). Survey of 1,076 B2B software buyers, March 2026. company.g2.com ↩ ↩² ↩³

2. G2, “2025 Buyer Behavior Report” (May 14, 2025). Survey of 1,169 B2B decision-makers. company.g2.com ↩

3. 6sense, “2025 B2B Buyer Experience Report.” Nearly 4,000 global B2B buyers. 6sense.com ↩ ↩²

4. TrustRadius, “2026 B2B Buying Disconnect” (July 15, 2026). 1,862 technology buyers and 444 vendors, surveyed January 2026. hginsights.com ↩ ↩² ↩³

5. TechCrunch, “ChatGPT reaches 900M weekly active users” (February 27, 2026). techcrunch.com ↩

6. Google, “I/O 2026: Welcome to the agentic Gemini era” (May 19, 2026). blog.google ↩

7. Gartner, “Gartner Sales Survey Finds 61% of B2B Buyers Prefer a Rep-Free Buying Experience” (June 25, 2025). 632 B2B buyers. gartner.com ↩

8. Pew Research Center, “Google users are less likely to click on links when an AI summary appears in the results” (July 22, 2025). pewresearch.org ↩

9. Ahrefs, “AI Overviews reduce clicks” (February 4, 2026 update). 300,000 keywords. ahrefs.com ↩

10. Search Engine Land, on SparkToro’s 2026 zero-click study (January–April 2026 data). searchengineland.com ↩

11. Ahrefs, “AI search traffic converts” (June 2025). Data from Ahrefs’ own site. ahrefs.com ↩

12. MarTech, on Semrush’s AI search study (June 2025). martech.org ↩

13. Ahrefs, “AI Overview brand correlation” (May 26, 2025). 75,000 brands. ahrefs.com ↩

14. Ahrefs, “Best lists research” (December 4, 2025). 750 ChatGPT prompts. ahrefs.com ↩

15. Ahrefs, “Do AI assistants prefer to cite fresh content?” (July 28, 2025). About 17 million citations. ahrefs.com ↩

16. Semrush, “Most cited domains in AI” (November 2025). semrush.com ↩

17. Semrush, “We analyzed 89K LinkedIn URLs cited in AI search” (March 10, 2026). semrush.com ↩