For years, search marketing followed a fairly predictable path. Someone had a question, typed it into Google, looked through the results and clicked on the page that seemed most useful. If your company ranked well enough, that search could bring a potential buyer directly into an environment you controlled.
Once they arrived, you could do something with that visit. They might read another article, download a report, subscribe to a newsletter or look at a product page. If they were further along, they might book a demo. Even when they left without converting, you’d still learnt something about how people were finding you.
AI search is starting to interrupt that journey. Google’s AI Overviews can answer the question before someone visits a website at all, while tools like ChatGPT can research products, compare options and recommend vendors without sending the user anywhere near the companies being discussed.
We already know what happens to clicks when this becomes part of ordinary search. Pew Research found that people clicked a traditional Google result in 15 per cent of searches without an AI summary, but only eight per cent when one appeared. Ahrefs has since found that an AI Overview correlates with a 58 per cent lower click-through rate for the number-one organic result.
For B2B technology companies, there’s another part of this shift to consider. AI search isn’t staying at the top of the research journey anymore. It’s moving closer to the products, brands and buying decisions vendors actually want to be found for.
AI Search Is Moving Into B2B Product Research
AI Overviews initially seemed easier to understand as an SEO problem when they mostly appeared for informational searches. If someone wanted to know what a technology did or how something worked, Google could summarise the answer. Useful for the person searching. Less useful for the websites hoping they’d click through.
But those boundaries are changing. Semrush tracked 458 manufacturing and industrial searches and found AI Overviews covering 38 per cent of their search volume in January 2026. By July, that figure had reached 57 per cent. The searches gaining AI Overviews included specific products and branded terms, not only broad educational questions.
That fits a much wider pattern. In a separate study of more than 600,000 keywords across ten industries, Semrush found that AI Overviews appearing on commercial-intent searches increased by 71 per cent over six months. These are searches made by people researching and comparing their options, rather than simply looking for a definition.
B2B buyers are also taking this behaviour beyond Google. G2’s 2026 Buyer Behavior Report found that more than 80 per cent of software buyers had sourced software recommendations from an AI chatbot during the previous two years. For those buyers, AI had its greatest influence during shortlisting and evaluation almost half the time.
This makes product discovery a very different problem. AI isn’t simply helping someone understand what identity governance or observability means. It can help them decide which vendors to investigate. And 6sense found that the eventual winner is already on the buyer’s Day One shortlist 95 per cent of the time.
Getting into that early research process has always been important. Now some of it can happen without generating a visit to your website at all.
The Lost Click Is a Bigger Problem Than Lost Traffic
Organic traffic is easy to reduce to a number on a dashboard. If visits fall from 50,000 to 40,000, marketing has lost 10,000 sessions. That’s useful information, but it doesn’t explain what disappeared along with them.
A website visit creates possibilities. Someone can subscribe to an email list, download a report, register for an event or request a demo. They can move through several pages and tell you something about what they’re interested in. Depending on your tracking and consent setup, that activity can also support retargeting and future marketing.
If AI answers the buyer’s question somewhere else, none of those interactions happen. The person may still have learnt something about your company. They may even be more interested in it than they were five minutes earlier. From the vendor’s perspective, however, they’re largely invisible.
AI referral traffic isn’t making up the difference yet. Semrush found that just 0.48 per cent of sessions to the manufacturing websites it studied came directly from AI search tools and assistants. Similarweb, meanwhile, found AI referrals growing 117.4 per cent year on year, but warned that the channel can still be small enough for companies to overestimate its current contribution.
So the commercial problem isn’t simply that AI search can reduce organic traffic. Fewer visits also mean fewer opportunities to turn anonymous research into an interaction the business can see, understand and potentially convert.
Niche B2B Vendors Have Less Traffic to Lose
That becomes particularly important in enterprise technology because most vendors aren’t trying to attract everyone on the internet. A cybersecurity platform designed for large financial institutions might only have a relatively small number of organisations it can realistically sell to. The same applies across infrastructure, data, AI and other specialist markets.
Within those organisations, the relevant audience gets smaller again. A buying group may involve security, IT, finance, procurement and senior leadership, but only a limited number of people are actively researching a particular problem at any given time.
That changes how we should think about declining search traffic. Ten thousand lost visits to a consumer publisher and 100 lost visits from people researching a specialised enterprise product aren’t commercially equivalent. The smaller number may include people from organisations capable of becoming significant customers.
And buyers do a lot of work before vendors know they exist. 6sense found that buying groups had already filled roughly four of the five places on their shortlist on the first day of the buying journey. In 94 per cent of cases, they’d also ranked their shortlist before speaking to sellers.
If AI is increasingly involved in helping buyers decide who deserves one of those places, B2B visibility has to extend further than getting the right product page onto page one of Google.
Your Website Can’t Carry the Entire Buyer Journey Anymore
None of this means company websites or SEO have stopped being useful. Buyers still visit vendor sites, search remains an enormous source of traffic, and owned content gives businesses somewhere to explain their products in as much detail as necessary.
The problem is expecting the website to do all the work.
A buyer can now ask an AI assistant for suitable vendors, encounter a company in an AI Overview, read independent reviews, watch an expert discussion and search the company by name later. Similarweb found that people exposed to AI recommendations were 2.5 times more likely to subsequently visit a recommended brand, while 55.9 per cent of AI-influenced traffic eventually arrived through search.
The vendor may see that final Google visit and assume search deserves the credit. What its analytics won't necessarily show is everything that helped create the interest beforehand.
Google itself is beginning to separate these ideas. Search Console now includes dedicated reporting for appearances within generative AI features, including AI Overviews and AI Mode. In other words, businesses can increasingly measure whether their content appeared even when that appearance didn't produce a click.
For B2B marketers, the practical response is to stop thinking about visibility as something that happens only on channels the company owns. As we explored in EM360Tech’s Approval Economy report, much of the enterprise buying journey already happens beyond the vendor’s view. AI search is pushing even more research into that wider information environment, which means brands need credible ways to exist there too.
Human Expertise Becomes More Valuable as AI Content Multiplies
There's a strange side effect to making content easier to produce. The internet gets more content, but that doesn't necessarily give people more worth paying attention to.
AI can explain a concept, summarise a report and turn the same basic information into a blog post, social update or email within seconds. For straightforward information, that's incredibly useful. But enterprise buyers aren't only looking for information. They also want judgement from people who understand the decisions they're making.
A CISO who's spent years responding to incidents can explain why a security strategy failed in ways a generic summary can't. A data leader can talk about the compromise nobody anticipated during an implementation. Put those people into a genuine conversation with someone who knows the industry well enough to challenge them, and their expertise becomes much harder to reproduce with generic content.
There’s evidence that buyers still actively seek this kind of thinking. LinkedIn and Edelman's research found that 56 per cent of target buyers and 55 per cent of less visible internal stakeholders use thought leadership when evaluating vendors. G2 found something similar from another direction: review sites slightly outranked AI chatbots as an influence on software shortlists, at 38 per cent versus 37 per cent.
AI can make finding possible vendors much faster. It hasn't removed the human need to work out which ones are actually worth believing.
Podcasts Turn Human Expertise Into Marketing Collateral
This is where podcasts become particularly useful. Not because AI systems have some special preference for them, but because people do consume them and the conversation itself is difficult to fake well.
A good podcast captures things polished marketing copy often removes. There are follow-up questions, personal experiences, disagreements, pauses to think and moments where someone has to explain what they actually mean. The audience gets access to the person behind the expertise rather than another carefully controlled company statement.
Podcast consumption also continues to grow. Edison Research's Infinite Dial 2026 found that 58 per cent of Americans consumed podcasts monthly and 45 per cent did so weekly, both record highs. It also found that 57 per cent had both listened to and watched podcasts, which shows how far the format has moved beyond audio alone.
For B2B marketers, though, the value doesn't end when somebody finishes an episode. It’s something we see directly through EM360Tech’s own analyst-led podcasts, where one expert conversation can become the starting material for a much wider content programme.
One conversation can become many assets
A recorded expert discussion can exist as a full video and audio podcast, but those are only the original formats. The strongest moments can become short videos, social posts, quotations and newsletter content. The ideas can support articles, while useful explanations can become material for sales teams working through the same questions with buyers.
Content Marketing Institute has made a similar distinction. Its B2B research found that only 27 per cent of marketers were using podcasts, but noted that companies often treat the format only as a distribution channel rather than a source for content that can be reused across blogs, newsletters and social media.
That makes the economics of a podcast very different from producing one asset for one audience on one platform. The conversation becomes the source material, while the individual pieces created from it give the expertise more opportunities to reach people.
Human content can travel further than the original format
Those different formats also solve a fairly ordinary problem: people don't all consume content in the same way. One buyer might watch a full discussion over lunch. Another might encounter a two-minute clip on LinkedIn, while someone else reads an article based on the same conversation weeks later.
Video already has a substantial role in B2B marketing. LinkedIn's 2025 benchmark found that 78 per cent of B2B marketers were using it, with more than half planning to increase their investment. Marketers with mature video strategies were also 2.2 times more likely to describe their brands as well trusted.
One genuinely useful conversation can therefore keep working long after the recording ends. More importantly, every format still leads back to the same thing: real expertise from identifiable people with something worth saying.
Visibility, Citations and Traffic Need Different Measures
All of this also creates a measurement problem. For years, search marketing gave us a fairly understandable sequence. A page ranked. Someone clicked. Analytics recorded the session. Maybe that visitor converted. We couldn't see everything happening around those steps, but at least the basic path was easy to follow.
AI search adds several outcomes that can look similar on a dashboard while meaning very different things. Semrush's manufacturing study found that only two domains appeared in both its top 15 most-mentioned brands and top 15 most-cited websites. It also found no clean relationship between receiving more AI citations and receiving more AI referral traffic.
That gives B2B marketers four useful distinctions:
- Mentioned: Does the AI system talk about the brand?
- Cited: Does it use the company's content as a source?
- Clicked: Does someone actually visit the website?
- Converted: Does that visit become an identifiable commercial action?
None is meaningless. They simply measure different things.
A citation can show that content is being used as a source without producing significant traffic. A brand mention can influence someone who later arrives through Google. A click can generate engagement without creating pipeline. And a conversion is much closer to commercial value, but tells you little about all the interactions that came before it.
GEO measurement therefore can't become a race to replace organic rankings with citation counts. Marketers need to understand where their brands appear, whether people reach them afterwards and, eventually, whether any of that influence contributes to business outcomes.
Build Authority Where Buyers Are Already Looking
Once buyers can discover and evaluate companies without going directly to them, publishing more content on the company website isn't a complete answer. Vendors need a presence in the wider places where their markets learn, compare ideas and decide who deserves attention.
That can mean independent industry publications, expert commentary, analyst conversations, podcasts and other content where the company's expertise appears alongside voices buyers already recognise. The goal isn't to abandon owned content. It's to stop asking owned content to carry a journey that increasingly happens elsewhere.
That's also where EM360Tech can help. We work with enterprise technology companies to turn internal expertise into thought leadership, analyst-led conversations, podcasts and supporting content designed for the audiences they're actually trying to reach. Those conversations can then travel across formats and platforms instead of remaining trapped on a corporate website.
For vendors competing in specialised markets, that creates more opportunities to become familiar before the buying process becomes visible. It puts real people and useful ideas in front of enterprise audiences while giving the business content it can continue using across marketing, sales and wider distribution.
AI search is changing how buyers find information. The answer isn't to try to drag every part of that journey back onto your website. It's to make sure your expertise can travel without it.
Final Thoughts: Authority Has to Travel Further Than Your Website
The old search journey isn't disappearing completely. People will still Google products, visit company websites, read blogs and fill in forms. But marketers can no longer assume that being discovered means getting the visit, particularly as AI Overviews and AI assistants become more involved in B2B product research.
That leaves vendors with two jobs rather than one. They still need strong owned content for the people who arrive, but they also need authority that can reach buyers before that happens. Human expertise, independent publishing and expert-led media give them ways to build familiarity and confidence beyond their own digital properties.
The companies that stay visible won't necessarily be the ones filling their websites with the most content. They'll be the ones whose useful ideas and credible experts keep appearing wherever their buyers are learning, comparing and making decisions.
If your buyers are spending more of that journey elsewhere, your expertise needs to meet them there. EM360Tech helps enterprise technology brands turn the people and knowledge inside their businesses into credible thought leadership, expert conversations and content that can travel further than their own websites.