As consumers turn to AI tools like ChatGPT and Google’s AI Overviews to find the information they need, traditional marketing search strategies – that prioritised keywords and SEO – essentially become redundant. In this article, six executives at the forefront of Generative Engine Optimisation (GEO) shed light on what brands must do to rank highly not in search, but in AI answers and recommendations.
Julia Pilkes, Senior Director of Marketing, HubSpot
“What do you do as a brand when you lose 140 million website visitors overnight? That happened to HubSpot – not through any failure of our own SEO strategy, but because the way people find information had fundamentally changed. The brands that are losing visibility in AI-generated answers right now often haven’t done anything wrong – they’ve just been optimising for the wrong thing.
“Experiencing it at HubSpot gave us a front-row seat to what GEO actually demands. For instance, traditional SEO rewards precision i.e., the right keywords, the right technical structure and the right backlinks. GEO operates on entirely different logic because large language models don’t crawl pages looking for ranking signals. They synthesise brand presence across the entire digital ecosystem, from reviews and media coverage to social authority, community mentions and owned content. If that footprint is narrow, even a technically perfect website won’t save you.
“The most impactful thing a brand can do today is audit its presence beyond its own channels. Not only reviewing website performance but overall reputation. What do public reviews tell you about your brand? Is it cited in credible third-party media? Does it feature in genuine online conversations? These are the signals LLMs value most heavily. That realisation is what led to HubSpot developing its own Answer Engine Optimisation (AEO) capability.
“The mindset shift is significant for leaders to stop asking, “Where do we rank?” and start asking “How is our brand understood?” Brands that have built genuine, consistent authority over time will win. There are no shortcuts in the world of GEO.”
Teddie Cowell, Global VP, SEO & GSO,WPP Media
“The single biggest thing a brand can do today is build a strong, credible content ecosystem across owned and third-party websites that clearly answers the questions its audience are asking. We call that a System of Influence. In simple terms LLMs can only surface what they can find, interpret and trust, so if your brand does not give LLMs something to cite by publishing useful, authoritative and accessible content, it is far less likely to appear. Many brands overlook this because they are still treating AI visibility purely as a technical issue, when it is really a content and authority issue. Some are starting to adapt by making content easier for LLMs to retrieve and understand, but most are still at the stage of working out where to begin.
“Whilst the underlying foundation models might be updated less frequently, the responses consumers see can draw on data retrieved in real time from across the web, search results, product feeds, agent to agent protocols like MCP and UCP etc.
“Consistent visibility in LLMs comes from sustained credibility rather than one-off optimisation, so brands should think about integrated short-term and long term tactics building durable authority. Short term, fresh content is often weighted more favourably by LLMs so keeping content updated means that it’s more likely to be recognised and surfaced.
“Long term keep core content accurate and up to date, showing up consistently across trusted sources, and create enough depth for foundation models to associate your brand with a topic over time.”
Chris Shelbourn, Organic Search Project Delivery Director, IDHL
“The most impactful thing a brand can do is to build consistent, third-party validation across the web. LLMs don’t rely on a single source, they look for repeated, corroborated signals across trusted environments. That means being cited in reputable publications, high quality media, social platforms, forums and relevant communities, so your brand is consistently associated with key topics.
“A strong website alone rarely translates into consistent visibility across AI outputs. Where brands begin to “stick” in LLM responses is when their narrative is reinforced externally, across multiple trusted sources.
“For fixed facts that do not change, for example “When was Google founded?” LLMs often rely on their trained knowledge base or embedded memory layers. In these cases, visibility is harder to influence in the short term. However, for more opinion based, comparative or evolving information, models typically draw on live or recently published sources. This is where brands have a clear opportunity to shape how they appear. In practice, we see that short-term tactics rarely compound. Visibility in LLMs comes from consistency, not just publishing useful content, but reinforcing it across trusted environments over time. Brands that treat this as a campaign tend to see fragmented or short-lived visibility, whereas those investing in sustained authority-building see far more stable and repeatable presence.”
Helene Simon, Head of EMEA, Webflow
“The most impactful thing that brands often ignore is refreshing content they already have. LLMs heavily favour freshness, in fact 95 percent of ChatGPT citations point to pages updated in the last ten months. When we accelerated our own content refreshes, we saw a 40% traffic lift within days. Creating net-new content feels more exciting, but a stale answer just means LLMs cite someone else’s fresher version instead.
“SEO remains essential – but it was built for a transparent, auditable system where rankings are visible and trackable. GEO operates by an entirely different set of rules. Where SEO earns you a position on a results page, GEO earns you a place in the answer itself.
LLMs don’t see your search ranking. They see what the rest of the web says about you. If that picture is sparse or outdated, you’re invisible, and you won’t even know it
The winners will be the ones becoming primary sources, not the ones chasing placement. That requires a fundamentally different investment: original research, third-party presence, and content that earns citations rather than content that targets keywords.”
Cathy Farmer Head of GEO, Third City
“LLMs synthesise patterns across everything that’s been written about a brand. If your core positioning appears consistently across owned, earned, and third-party channels, the model builds a coherent picture. If it doesn’t, you’re noise. Most brands overlook this because there’s no dashboard flagging the problem. They invest in their own content while neglecting the external signals – media coverage, customer forums, Reddit, YouTube – that training data draws heavily from.
“Understand that everything you’re doing publicly right now is potentially influencing future training data. That’s a mindset shift most brands haven’t made yet.
“Training data pulls significantly from media owners and from trust-based discussion in places like Reddit, YouTube, and established forums. So the answer is essentially the same as the first question: earn credible third-party coverage, generate genuine customer conversation in the right places, and do it consistently over time. Brands chasing short-term visibility tactics are optimising for a snapshot. The ones building durable authority are playing a longer game – and they’ll be the ones the models cite by default in two years’ time.”
Rich Harper, Head of Digital Marketing, Brew Digital
“Some of the most influential search activity now happens in online communities, Reddit threads, Discord groups, LinkedIn discussions, and niche Facebook forums. These spaces have become trusted sources of insight, where real people share authentic recommendations. And AI-driven search engines are paying attention, surfacing user-generated content, expert takes, and social proof more than ever.
“Brands must view AI visibility much like academic reputation. You do not build lasting authority through a single article or a quick optimisation tactic. You build it through consistent contributions, independent citations, and peer recognition over time.
“Modern AI systems are designed to identify credible sources rather than just well-optimised text. They place the highest confidence in organisations with accumulated evidence of expertise. Your most valuable assets for long-term AI visibility are original research, proprietary data, expert commentary, and consistent third-party validation.
“These credibility signals persist across training cycles and retrieval systems. Every research report you publish, media mention you earn, and industry citation you secure strengthens your position in the broader information ecosystem.”
