AI Leads 94% of B2B Buying, Forcing Rethink of Search Visibility

Ninety-four percent of business buyers now use AI in their buying process, and twice as many say generative AI or conversational search is a more important source of information than any other source, Forrester says.

If buyers are getting answers from AI before they ever speak to sales, brands need to show up in the sources those systems cite, or they get left out before the real work even starts.

Ash Lockett, head of B2B marketing at Babel, says the answer isn’t a total reinvention of marketing.

Instead, he believes in clear positioning, visible experts, and content that is open, structured, and easy for both people and AI systems to find and trust.

“I'm going to challenge this question and say I'm not sure it is redefining what brands should do. I would more likely say it's kind of refocusing on getting the fundamentals right.”

In this DesignRush interview, Lockett argues that gated content and generic messaging are quietly weakening visibility in AI-driven discovery.

Who Is Ash Lockett?

Ash Lockett is Head of B2B Marketing at Babel, with more than 15 years of experience across PR, marketing, and brand roles in agency and in-house environments in the UK, Australia, and the US.

He started his career at WPP agencies in the UK, including an earlier role at Babel PR, before relocating to Sydney, where he spent 10 years working across major WPP agencies.

His work during this period focused on B2B communications, brand campaigns, and client strategy across multiple sectors. He later co-founded a brand engagement agency, which operated for four years and earned industry recognition, including being named Holmes Report’s “Pound for Pound” most creative agency in the world.

After his agency leadership experience, Lockett moved in-house to a venture builder, where he led marketing across a portfolio of businesses.

He then joined a fintech scale-up as an early employee, initially leading marketing before expanding his remit to include sales and customer success as the company scaled across Australia and entered the US market. He later served as Global Head of Marketing, focusing on growth and market entry efforts in the US.

Generic Messaging Is Part of the Problem

Lockett is blunt about how much B2B messaging still tries to speak to everyone at once.

“The reality is, we also wanted to capture as broad a market as possible because we wanted as many of these e-mail addresses as possible, right?” he says.

“So this led to kind of a spray-and-pray vanilla content, targeting these really generic buyer personas.”

For example, Lockett tells me that, hypothetically, the target is the technology decision-maker at companies with 1,000 to 500,000 employees, typically aged 25 to 54.

“The only thing that's nuanced about that is that they're in tech, which is still kind of a large segment,” he adds.

In practice, the brand ends up sounding like every other vendor in the category, which makes it harder for buyers to understand what it’s actually known for.

Lockett’s point about discoverability also speaks to how content is packaged for modern discovery channels.

Research from 10Fold’s AI-First, Buyer-Ready study shows that reaching B2B tech buyers now depends on social media, paid media, video, and research-led content.

Yet only 11% of B2B organisations say most of their content is actually structured for AI discovery.

That’s problematic.

If a brand can’t define its audience and point of view clearly, it becomes so much harder to earn attention during the research phase or to show up with any authority when AI systems scan for relevant sources.

Gated Content Is Working Against Brands

One of Lockett’s strongest criticisms is aimed at gated content, which he believes is still hurting both discoverability and user experience.

“The number of brands I’m actively telling on a weekly basis, please take down your gates, is still mind-blowing.”

If content sits behind a form that needs to be filled in, AI systems can’t access it. It’s that simple.

It means any white paper, guide, or report that took weeks to produce might be invisible in the very systems that buyers are using to research vendors.

Not to mention that it frustrates human users. During the early stage, buyers don’t want to hand over their email addresses just to look around.

They’re still trying to understand the market, not book a demo.

This makes gated content a bad trade. Brands lose reach at the exact moment buyers are most open to learning.

“Whilst LLMS can obviously crawl PDFs and they're pretty good at doing that, you lose all the structured content, the semantic HTML for better citation extraction.”

He explains that a PDF traps information in a closed container.

Once content is inside it, there’s no real structure for a system to explore beyond that document. It’s static, isolated, and hard for external tools to understand in relation to other material.

A website built in HTML works differently.

Pages connect through links, forming a web of connected pages, where each click leads to another piece of related information, like moving through rooms in a building instead of standing in a single locked box.

Within this structure, systems like LLMs can travel across related pages, pick up additional context, and build a fuller picture instead of relying on a single file.

The argument here is that HTML gives content more context and discoverability.

It’s easier to access, easier to interpret, and better suited for generating accurate references because it supports interconnected information rather than locking everything into one fixed document.

“Not only that, it's a better user journey for the reader itself, like you don't have to sit there and have a PDF on your laptop,” he adds.

The point is, PDFs still work for distribution, but they’re not ideal when the goal is discoverability, context, and richer citation paths.

Authority Depends on People, Not Just Brands

Lockett thinks B2B companies have also leaned too hard on brand-level content and enough on visible experts.

“I see it time and again. That's not going to help with citations at all,” he says.

Content written by an expert inside the organisation who is active in relevant online communities is more likely to be cited because it signals experience, expertise, authority, and trust that can be validated across those networks.

Lockett points to the need for subject-matter experts to actively shape and publish thinking, not just appear as names on brand content.

That includes work such as Babel’s The Brave Benchmark AI Audit, which looks at how AI systems influence visibility in B2B research.

That matters because trust is forming earlier than most teams account for.

Edelman found that 95% of hidden buyers are more receptive to sales and marketing outreach when they have already engaged with strong thought leadership content, since it builds credibility before any direct contact happens.

Lockett also points to LinkedIn and Reddit as especially important for B2B discovery.

Those are the platforms where people talk openly about their work, ask questions, and build credibility over time.

The same logic applies to PR.

A strong media placement is useful, but it delivers more value when the spokesperson is also visible in the places where buyers and AI systems are already looking.

Content distributed through earned media channels has been linked to a median 239% lift in AI search visibility when surfaced across AI systems.

This is according to Stacker’s Generative Engine Optimization study, based on 87 stories and more than 2,600 prompts across multiple AI platforms.

For Lockett, trade media plays a strong role in reinforcing authority within a specific market.

“Trade publications are becoming super important because they're not covered in the mainstream media,” he says.

They give context, detail, and relevance that a general business publication often does not.

When a brand shows up only on its own website, it’s relying on self-assertion. But when it appears in independent coverage, that adds outside validation.

This points back to what he prioritizes with PR.

“What we need to do is get as much coverage as we can on the topic that matters to your business, the lanes that you pick, the message that you actually want to promote to that target segment,” he explains.

That approach breaks down into three actions:

1. Build a concrete narrative.
2. Reinforce it across PR, content, and social channels.
3. Stay consistent about the topics the brand wants to own.

This will help buyers understand what the brand stands for, and also help AI systems connect the dots across the sources and attribute expertise to the right people.

What Brands Need to Focus On

Even with that focus in place, there is still no clear way to see how that visibility translates into performance across AI-driven discovery and research.

The challenge is knowing that what works is still not fully solved in this new environment.

“Everyone's looking for the silver bullet. I'm pretty certain if there was a simple answer to measurement, like we'll be talking about the next unicorn decacorn in the press constantly,” Lockett notes.

That means teams need to be more disciplined about the basics. Clear positioning, accessible content, visible experts, and consistent messaging are still the missing pieces that matter most.

This is reflected in Babel’s A Brave Framework for B2B Marketing Strategy.

“The winners will really be those with a strong brand, a unique message, an active bench of experts that know what they want to be famous for, that aren't everything to everyone, that really have picked their lane, pick their audience and are doubling down on that,” Lockett believes.

That requires knowing the audience, owning a topic, and producing content that’s useful enough to act on.

“What can you write today that your audience can read and implement tomorrow?”

That question points to the kind of content buyers use and the kind of content AI systems are more likely to surface.

Brands that get this right are easier to find when decisions are being made.