At Better B2B South, the afternoon panel discussion ‘Beyond SEO: Winning visibility in the new search landscape’ was led by search experts explaining how brands can stay ahead when AI is changing everything.
Search has left the building
Not long ago, being found meant ranking on page one of Google. You picked your keywords, optimised your landing pages, published your blog posts and watched the clicks roll in.
That model has not disappeared, but it is no longer the whole story. Buyers now discover brands on LinkedIn, validate them on Reddit and ask ChatGPT or Perplexity to build their shortlist before they ever visit a website. Many never click at all.
At our recent Better B2B South event, we brought together four search specialists to unpick what AI search visibility really means for B2B brands in 2026:
- Billie Geena, SEO Lead at Tutorful
- Flo Guerrouani, Global Search Performance & AI Lead at Cushman & Wakefield
- Rejoice Ojiaku, Senior Business Content Specialist at Wise
- Radina Ivanova, AI Search Lead at Hallam
This article pulls together what they shared: what has changed, what still matters and what you can do about it.
What does search look like in 2026?
The first question to the panel was deceptively simple. How do we define search when people can find you on Google, LinkedIn, Reddit and ChatGPT?
Rejoice summed it up well. The old journey was linear: type a keyword, get a result, click. Now, before a buyer searches for your brand directly, they have probably already met it somewhere else. They saw a LinkedIn post, read a Reddit thread or noticed you mentioned in an industry publication.
Longer, messier B2B journeys
For Flo, working in commercial real estate, the buying journey can run for months. Investors and occupiers start in similar places but quickly fragment across LinkedIn, trade publications and professional communities. Only when they get close to a decision do they start looking hard for trust and authority signals.
Her team has even started looking at TikTok, because that is where younger audiences in the sector are beginning to discover content.
Your website is now the last stop on the journey, not the first. Visibility has to be earned on every platform that feeds the buyer’s research, including the ones you do not control.
One practical tip from the session: because AI search encourages longer, more conversational queries, you may already be seeing these situational searches appear in Google Search Console. Have a look. They tell you a lot about the problems your buyers are trying to solve.
Why reviews and sentiment now shape what AI says about you
When someone asks an AI tool for a recommendation, you no longer just need to rank. You need to be mentioned, and ideally cited. That makes brand reputation a search issue.
Billy shared an example. Her business provides tutoring, so GCSE results day in August is peak season for reviews. As soon as exams finish, a wave of feedback arrives from parents, students and tutors. After that influx, her team saw a clear improvement in AI sentiment and a sharp spike in referrals from AI platforms.
But the effect cuts both ways. For years, reviewers described the business as a “marketplace”, a label it is now actively trying to move away from because customers want a personalised service. Despite new messaging, that old sentiment lingers in how LLMs describe the brand.
LLMs hold on to old narratives. Audit what your reviews actually say about you, and update your messaging so the language customers use matches the position you want to own.
Consistency is the new authority
Rejoice built on this. AI tools assemble a narrative about your brand, and that narrative often comes from third parties rather than your own website. If publishers, directories and partner sites describe you differently from how you describe yourself, the machine has to pick a version.
She described a client that offered guided tours but was not described that way on the attraction websites that listed it. As a result, it was grouped with the wrong type of business and appeared lower in AI answers than its competitors. Fixing the entity and narrative across third-party sites became the priority, ahead of anything on the client’s own website.
This matches what we have seen in our own research. In our Share of LLM analysis, brands in the top half for Share of Search were 2.5 times more likely to also rank in the top half for Share of LLM, but the brands that outperformed were those with consistent, authoritative content ecosystems. We explore this further in our piece on why brand mentions now matter more than backlinks for AI search.
A slow win worth starting now
Billy’s tip here was refreshingly practical. Search for your brand (or your client’s) on Google and look for directories, listings and mentions. Where you can easily contact the site owner, ask for corrections. It is not a quick win. It is a very slow one, but it gradually changes how you are recognised.
The panel also suggested paying attention to the platforms AI tools rely on. If you have a local presence, make sure you are accurately listed on Google, Bing and relevant directories such as Tripadvisor. Radina noted that partner platforms like these are likely to be prioritised in AI responses.
Is traditional SEO still enough?
The panel’s answer was unanimous: keep it all. AI has not replaced SEO. It has shown that SEO works in layers, and every layer still counts.
Content that machines can extract
Rejoice argued that AI has taken us back to basics. Rankings and keywords still matter, but the question has shifted from “what has the highest search volume?” to “what does the user actually want to know?”
Write for humans, but make content easy to extract. In practice that means:
- Clear FAQ sections that answer real questions directly
- Lists and bullet points for steps and comparisons
- Infographics and visual summaries that stand on their own
- Short, specific answers near the top of the page
If you want to go deeper, our guide on how to optimise for AI search covers the content and technical signals in more detail.
Technical SEO: the always-on foundation
For Flo, technical SEO is non-negotiable. Her site runs on one domain across multiple markets and languages, so if the technical foundation is weak, crawlers and bots simply will not find the content. It is background work that rarely gets applause, but without it the site would not be visible at all.
Billy agreed, with a caveat. Many clients tell her they have already “done” technical SEO by running their site through an AI tool, then get lost in complex, low-value fixes. Her advice is to get a sheet of paper and stick to the basics:
- Can search engines and AI crawlers access your site?
- Are the pages you want indexed actually indexable?
- How quickly can a user, or an LLM, get to the information they need?
Most modern websites built on platforms such as Shopify, Wix or Salesforce are already in decent shape. Chasing marginal technical issues will frustrate your developers without moving the needle.
Get the technical basics right and keep them right. Then put your energy into brand, content and third-party visibility.
Watch out for JavaScript and AI crawlers
Two technical points stood out. First, if your content depends on JavaScript to render, you have a problem. Google can render it, but many AI crawlers cannot, so fixing JavaScript dependency should be a top priority.
Second, AI crawlers do not always play by Google’s rules. The moderator shared findings from server log analysis showing ChatGPT’s crawlers visiting pages tagged noindex. If a page is live, assume AI tools may read it, and review what is on there. OpenAI documents its different crawlers and user agents, which is a useful starting point for log analysis.
How do you measure AI search visibility?
This was the question everyone wanted answered, and the honest answer from the panel was that nobody has fully solved it yet.
Flo was candid. With teams tracking data across dozens of markets, simply getting everyone to report on the same thing is a challenge. Her focus is on a single source of truth: one dashboard, with the main data sources connected, and agreement across the organisation on which KPIs matter.
Billy described modern acquisition as a mess. A buyer might see an AI Overview, leave, return weeks later via ChatGPT, then come back again through social while they do their due diligence. Every tool reports share of search, impressions and LLM sentiment differently, so you will never get one perfect number.
Tools and metrics the panel uses
Despite the uncertainty, the panel shared what is working for them:
- Prompt tracking. Rejoice uses SE Ranking to set up prompts and see where her brand appears in AI answers. Position matters: being the fourth recommendation in a list of ten is still valuable, and higher placements are more likely to earn a link.
- Cross-platform sentiment. Tools such as Promptwatch show how sentiment differs between ChatGPT, Claude and other models.
- Share of LLM. A share-of-voice style measure of how visible your brand is across a defined set of category prompts. This is the approach behind Hallam’s own Share of LLM framework.
- CRM and GA4 data. Look for correlations between AI referrals and leads, MQLs and SQLs, and track conversion journeys across paid, organic, AI referral and email.
There is no single metric for AI search yet. Agree your KPIs internally, keep one source of truth and use the data to tell a story rather than chase a perfect number.
Report less often, with more story
In the Q&A, an audience member asked how often the panel reports. Flo reports quarterly, across three regions rather than 35 individual markets. Weekly reporting, in her view, is mostly noise. Longer windows reveal trends.
Above all, she said, focus on what the report changes. What actions will the team take together, and what can individual markets or clients act on? Stakeholders rarely care about graphs on their own. They care about the story and the next step.
Fewer clicks, better leads
Another audience question tackled the obvious worry. If AI answers reduce clicks, is all this effort worth it?
The evidence on clicks is clear. Pew Research Center found that Google users clicked a traditional search result in 8% of visits when an AI summary appeared, compared with 15% when it did not.
Billy’s team has lived this. Their average ranking position improved from around 14 to around seven, yet clicks have not followed, because users increasingly get what they need from AI Overviews and other search features. What has followed is lead quality. Conversions are up, and the visitors who do arrive are far more motivated.
Her advice is to sit down and define what a qualified lead means for your business, then build a custom report in GA4 to track it. If those leads are not coming through, that is the problem to solve, not the click-through rate.
Owning your topics beyond your website
Should technical SEO now take priority to win in LLMs? Rejoice argued the opposite. Get the basics right, then focus on owning your topics.
LLMs prioritise credible sources. If they are pulling information about your topic from everywhere except your website, someone else is telling your story better than you. Brands that simply use AI to churn out content are not owning anything. Meanwhile, many companies sit on excellent case studies they never talk about.
She pointed to Monzo as an example of a brand that owns its topics in B2B, and to the growing number of brands building communities on Reddit, one of the most frequently cited sources across AI tools.
A tip for regulated industries
For brands in finance, insurance, healthcare or education, Rejoice shared a simple way to stand out. Every year, government bodies publish updates on dry regulatory topics. Read them, rewrite them in plain English and repurpose the explanation across your channels. It positions you as the expert and fills a genuine gap.
Break down the silos
Flo admitted she had not known who sat in her company’s PR and social teams until recently, despite seven years in the business. Now they collaborate on content, repurpose LinkedIn material and share market data packed with the kind of statistics that get cited.
Rejoice agreed: AI has effectively forced collaboration. Because AI tools draw from PR, social, content and the website at once, it no longer makes sense for SEO to work in isolation.
Should you make bulk content changes?
The final audience question came from a one-person marketing team. Do large content updates help or hurt in LLMs?
Billy’s team recently updated more than 120,000 pages to fix their brand messaging. Some areas improved, while others dipped. It takes time for LLMs to pick up changes, so expect short-term volatility, and do not panic if numbers drop for a month or two.
Rejoice, working in a regulated field, recommended a more strategic approach. Group bulk updates into topic clusters that clearly relate to each other, rather than changing unrelated pages at random.
Radina added a final tip, test. Unique content such as white papers and case studies can appear in Gemini and Perplexity within 24 hours. Server logs will show whether AI crawlers have visited your updated pages, and for what purpose.
Winning AI search visibility: where to start
The panel’s message was consistent. AI search visibility is not a separate discipline from SEO. It is SEO with a wider lens, where your brand, your reviews and your third-party presence matter as much as your website.
If you are wondering where to begin, start here:
- Audit how AI tools currently describe your brand, and compare it with how you want to be positioned.
- Check that reviews, directories and publisher mentions use consistent language about what you do.
- Fix the technical basics, especially crawlability and JavaScript dependency.
- Structure content so answers are easy to extract.
- Agree a small set of KPIs and one source of truth before building new reports.
- Get SEO, PR, content and social teams talking regularly.
As Billy put it, good SEO is good GEO, AEO or whatever acronym comes next. The fundamentals have not changed. The places buyers find you have.
If you want to understand how visible your brand is in AI search today, get in touch with us or join the conversation in the Better B2B community.
