The Great GEO Gold Rush

Hi everyone, and welcome to the latest Napier webinar. It’s great to have you here. Today’s session focuses on what we’ve been calling internally the “great GEO gold rush.”

To begin, let’s clarify what we want to achieve. Many people are now talking about generative engine optimization, or GEO, which is essentially about getting your brand included in the answers produced by AI tools. In this session, we’ll explain why this is becoming so important and how you can start influencing those AI-generated answers.

We’ll start by looking at what’s happening with search. Search is changing quite dramatically, and it’s important to understand that this isn’t simply a shift from traditional search engines to AI chatbots. We’ll also explore how to measure GEO performance and, most importantly, how to actually get your brand into AI-generated results. Along the way, we’ll share research Napier has conducted across around 30 B2B technology companies. These insights are particularly relevant because getting into AI answers in B2B markets is quite different from doing so in consumer markets. We’ll highlight both the differences and the similarities, and briefly touch on GEO tools and their capabilities at the end.

Looking at search trends, recent research gives us useful context. A study by OneLittleWeb covering April 2024 to March 2025 found an 81% increase in visits to AI chatbots, while traditional search declined by only 0.5%. Search is clearly not dead—search engines still receive 34 times more visits than AI chatbots. This suggests that AI is currently additive rather than a direct replacement. However, it’s also true that platforms like Google and Bing increasingly show AI-generated summaries alongside traditional results, so generative AI is already influencing classical search experiences. The familiar “blue links” still matter, but the landscape is evolving.

More recent research from Ahrefs shows continued growth, with ChatGPT prompts now representing about 12% of Google’s search volume—up from just 0.5% in less than a year. That’s a remarkable increase. Despite this, Google still drives around 190 times more traffic to websites. This highlights a key distinction: SEO has traditionally focused on driving website traffic, whereas GEO is about visibility and brand perception within AI-generated answers. While AI may cite your content and occasionally drive clicks, the volume is far lower than traditional search. GEO should not be expected to deliver significant inbound traffic; its value lies in shaping how your brand is presented.

In some industries, particularly publishing, increased AI usage has significantly reduced website traffic. In others, the impact has been smaller. Whether users click through or rely on summaries depends heavily on the sector and audience behavior. This variability is important to keep in mind when evaluating GEO’s impact.

Our research also shows that B2B brands currently have relatively poor visibility in AI answers. Across roughly 1,000 prompts and 29 brands, only 21% of companies appeared in more than 25% of relevant responses. Even more striking, about one-third of brands appeared in less than 5% of answers. These were prompts the brands themselves considered highly relevant. This indicates a major opportunity: many B2B brands are simply not showing up, leaving significant room for improvement through better GEO strategies.

AI is also changing how people search. Traditional SEO relies heavily on keywords and relatively short queries, such as “best running shoes for beginners.” In contrast, AI queries are much longer and more complex. Users often describe their situation, provide context, and then ask a detailed question. These interactions are frequently conversational, with users refining their queries over multiple steps. This creates a funnel-like process, and brands need to appear consistently throughout that journey.

In B2B contexts, especially for technical audiences, queries can be highly sophisticated. For example, an engineer might ask an AI to compare FPGA vendors for a specific application with constraints like low power and connectivity. These are the kinds of queries where visibility is critical, yet they may not explicitly name vendors. Without a strong GEO strategy, your brand may be excluded entirely from these high-value decision-making moments.

Measuring GEO effectiveness is more complex than traditional SEO because of the variability in prompts. Instead of focusing on keywords, measurement revolves around broader concepts. Visibility refers to how often your brand appears in responses. Share of voice measures how frequently your brand is mentioned relative to competitors. Citations track how often AI pulls information from your website or related sources. Sentiment and thematic associations help you understand how your brand is being described, while mention position indicates where you appear in lists or comparisons.

To measure these effectively, you need a structured process. First, identify which AI models matter most to your audience. For example, software engineers may favor Anthropic’s Claude, while enterprise users may rely on Microsoft Copilot. Next, define segments based on product categories, applications, or customer problems. Then, run a large number of prompts—often 50 to 100 per segment—and collect data regularly, ideally daily. AI outputs vary due to randomness and frequent updates, so consistent tracking is essential. Once you have the data, you can analyze it and refine your content strategy.

When it comes to improving GEO performance, the core principle is simple: answer questions. Unlike SEO, which focuses on keywords, GEO is about providing clear, authoritative answers to the questions your audience is asking. High-quality content remains essential, but it must be structured in a way that aligns with how AI models process information. Using question-based subheadings and including FAQ sections can be particularly effective.

Platform presence also matters, but not always in the ways people expect. For example, our research found that Wikipedia was not as influential in B2B AI answers as commonly assumed. In many cases, it didn’t appear at all. LinkedIn, on the other hand, was surprisingly influential—especially LinkedIn articles, which are relatively easy to publish and control. Reddit also played a smaller role than expected. These findings highlight the importance of relying on data rather than assumptions.

Video content is another powerful factor. YouTube appeared in the top cited domains for 75% of brands analyzed, particularly in Google-related AI results. Creating video content that answers key questions can significantly improve visibility.

It’s also critical to mention your brand explicitly in your content. If your content is cited but doesn’t include your brand name, you may gain no visibility benefit. This represents a shift in content strategy, requiring more deliberate brand inclusion within informational content.

There are many GEO tools available, but they vary widely in capability and cost. Effective tools measure performance, provide insights, and support content development. However, high-quality tools can be expensive due to the cost of running large volumes of AI prompts. Free tools may offer basic insights but are often too limited for serious campaigns.

Some tools offer technical features such as schema generation, which is highly recommended. Schema helps AI models understand your content more clearly and is actively encouraged by AI companies. Other features, such as serving different content to AI crawlers, carry risks and should be approached with caution, as they may lead to penalties similar to those seen in early SEO practices.

Content generation tools are also common, but relying entirely on AI-generated content can be problematic. AI-generated text often performs worse than human-written content, partly because AI systems are designed to avoid reinforcing their own outputs. A better approach is to use AI for research and structure, then have humans create the final content.

In summary, GEO is becoming increasingly important as AI usage grows. There is a significant opportunity, particularly in B2B, where many brands currently have low visibility. Success requires careful analysis, the right tools, and a strong focus on creating high-quality, question-driven content. The encouraging news is that results can appear quickly—often within days rather than months.

Before moving to questions, a quick note about our next webinar at the end of April. It will focus on research into email marketing performance across different industries, highlighting best practices and areas for improvement. If you’re interested, you can sign up using the QR code or link provided.

Thank you very much for joining. We appreciate your time and welcome your questions. The webinar will be available on demand, and attendees will receive a link to access it. On the topic of timescales, unlike SEO, which can take months to show results, GEO can deliver impact much faster. We’ve seen content appear in AI answers within a week of publication.

Regarding tools, the key is to use a comprehensive, integrated platform that supports the full GEO process, from identifying queries to creating content. While many tools are available, it’s important to choose one that provides reliable data and actionable insights. Expertise is also essential, whether in-house or through an agency.

Finally, thank you again for attending. If you have further questions or would like to learn more, feel free to reach out via email. And if you have a moment, we’d appreciate it if you could complete the webinar evaluation.