Do B2B Buyers Actually Use ChatGPT to Find Software? Here's the Data.
Yes. As of 2026, half of B2B software buyers start their research in an AI chatbot rather than Google, nearly three-quarters use one somewhere in the process, and AI chatbots have overtaken review sites as the single biggest influence on which vendors make the shortlist. This isn't a prediction about where buying is heading. It's survey data about how your pipeline already works.
This post gathers the numbers in one place, because at some point you'll need to justify spending time or budget on AI visibility, and "I read that buyers use ChatGPT now" won't survive a leadership meeting. These figures will.
The headline number: the starting line moved
In April 2025, 29% of B2B software buyers said they typically began their research in an AI chatbot rather than a search engine. Eleven months later, G2 asked again: 51%.
That's the stat to sit with. Not because half is a magic threshold, but because of the slope. A 22-point jump in under a year is not a channel emerging, it's a channel arriving. In the same survey, 71% of buyers said they rely on AI chatbots for software research generally, and just 3% said AI hadn't meaningfully changed how they research at all.
For anyone tempted to file this under "interesting trend to watch": the watching phase ended somewhere in 2025.
What buyers actually do in there
The lazy assumption is that people ask ChatGPT vague questions and then do their real research elsewhere. The data says the opposite: the highest-value research activities are the ones moving into chat first.
Forrester's survey of nearly 18,000 business buyers found that 94% used AI during their most recent purchase. More telling is what for: over half used AI tools to compare vendors against each other, a similar share to research products in depth, and nearly half used AI to build the internal business case, all before contacting any vendor. In the same research, AI answer engines ranked as the number one vendor research source, ahead of vendor websites and sales reps.
In G2's data, the top use case was head-to-head vendor comparison. Buyers aren't asking AI what a CRM is. They're asking which one to buy, for a company like theirs, and why.
And it changes what they buy
This is the part that turns a curiosity into a budget line. In G2's survey, 69% of buyers ended up choosing a different vendor than the one they originally had in mind, based on AI chatbot guidance. One in three bought from a vendor they had never heard of before the AI surfaced it.
Read those two numbers together. The first says AI answers routinely overturn existing preferences, including, presumably, preferences your marketing spent months building. The second says the door is open in both directions: unknown challengers are winning deals purely on the strength of being in the answer.
There's a softer effect underneath: 85% of buyers say they think more highly of a vendor when an AI chatbot recommends it. An AI mention has become a credibility signal in its own right, somewhere between a review score and an analyst nod.
The objections, taken seriously
"Our buyers are senior. They're not the ChatGPT demographic." The age data says otherwise. Bain found AI search adoption for purchase research running at 58% among Gen Z and millennials, 45% for Gen X, and 25% even among baby boomers. The Gen X figure is the one that matters for most B2B buying committees, and it's nearly half. The decision-maker on your next enterprise deal has more likely than not already used an LLM in a buying process.
"Enterprise buying is committee-driven. One person's chatbot habit doesn't decide anything." True, and it makes the problem worse, not better. Forrester's data shows buyers using AI to draft the internal business case before any vendor contact. The document circulating your buying committee may itself be AI-synthesised. If the AI's view of your product is wrong or absent, that error is now in the committee's briefing pack.
"This is a hype cycle. It'll settle." The trajectory argument cuts the other way: from 29% to 51% as a starting point in eleven months, with 3% of buyers reporting no change to their habits. Channels in a hype cycle show adoption without behaviour change. This one shows behaviour change at every stage of the journey.
The gap that makes this worth acting on now
Here's the strange part. Buyer adoption is at the levels above, yet only around a fifth of marketers currently track their AI visibility at all. Most companies have never once checked what AI assistants say about them, while half their buyers start there.
Gaps like that don't stay open. This one exists because the shift outpaced the playbook: analytics can't see conversations inside ChatGPT, so nothing in the standard dashboard forced the issue. But a channel that builds the shortlist, invisible or not, is a channel you're competing in, either deliberately or by default. Right now most of your competitors are competing by default. That's the window.
What to do with all this
Three steps, in order of effort.
First, find out where you stand. Ask the major AI assistants the questions your buyers ask and see whether you appear; I've written the full 20-minute method here [link: DIY check post], or my free mini audit will score you in about a minute [link: mini audit].
Second, if the terminology is getting in your way internally (GEO, AEO, the rest of the acronym soup), send your team the plain-English guide [link: pillar page].
Third, if the score comes back ugly, that's fixable, and knowing is the expensive part. The companies losing deals in AI answers today aren't losing because the problem is hard. They're losing because they haven't looked.
FAQ
How many B2B buyers use AI chatbots to research software? As of 2026, 71% of B2B software buyers rely on AI chatbots for research, and 51% start their buying process in a chatbot rather than a search engine, according to G2's survey of over 1,000 B2B decision-makers. Forrester's larger survey found 94% of business buyers used AI during their most recent purchase.
Do AI recommendations actually influence B2B purchase decisions? Yes, measurably. G2 found that 69% of buyers chose a different vendor than originally planned based on AI chatbot guidance, and one in three purchased from a vendor they had never heard of before an AI surfaced it. AI chatbots now outrank review sites as the top influence on vendor shortlists.
Which AI tools do B2B buyers use for software research? ChatGPT dominates, used by around 63% of B2B software buyers for research, with Perplexity, Gemini, Claude and Google's AI Overviews making up most of the remainder. Coverage varies significantly between engines, which is why visibility should be checked across several rather than ChatGPT alone.
