What gets B2B brands cited in genAI answers

A study of 1,000+ prompts across four AI engines reveals what drives B2B brand visibility in generative AI answers — and where common GEO advice falls short.

Table of Contents

    Spy on Any Website

    Get traffic data and keyword intel on competitors instantly.

    One of the biggest priorities for marketing teams is managing the transition from conventional search to customers and prospects using AI for research. Although SEO remains essential, it’s critical that your brand appears in AI search summaries and chatbot answers.

    Marketers are being bombarded with advice and recommendations on how to run generative engine optimization (GEO) campaigns. Some of this advice is great, but other recommendations are simply snake oil. We decided to conduct some research to determine what really affects whether a B2B brand appears in generative AI responses.

    How we analyzed B2B AI visibility

    Our research used one of the leading GEO tools, Brandi, to analyze brand mentions and website citations in AI answers. The analysis was extensive, tracking more than 1,000 prompts across 29 brands using four different AI engines (ChatGPT, Perplexity, Grok and Google AI Mode/Gemini).

    We analyzed a mix of B2B tech companies, including electronic components, semiconductors, industrial automation, software, logistics, electrification and engineering. There was a mix of company sizes, though we predominantly focused on enterprise and midsize companies.

    The prompts were created by Brandi’s intelligence engine based on the key topics and questions each brand felt it should rank for.

    All analyses used English-language prompts and answers.

    800×440 1

    What 1,000 prompts tell us about B2B AI visibility

    Here’s what the data revealed about how B2B brands appear — or fail to appear — in AI-generated answers.

    Many B2B brands have low visibility

    AI visibility is the frequency a brand is cited in response to a set of prompts, expressed as a percentage. Our research showed that only one-fifth (21%) of the brands appeared in more than 25% of the prompt answers that mattered to them. One-third of the brands appeared in less than 5% of the AI answers.

    This isn’t due to the research focusing on companies with low GEO. Two-thirds of the time, the top brand in the B2B category we analyzed had less than 25% visibility, and nearly one in five (17%) of the most visible brands appeared in fewer than 10% of the AI responses.

    This is clearly concerning. Today, B2B brands don’t appear in the AI responses to prompts used by their customers and prospects. As B2B customers move from conventional search to AI chatbots, the visibility of most suppliers will decrease.

    Having the right content on your website is a critical factor in ensuring your brand is visible. Brands whose domain was in the top 10 domains cited by AI answers averaged 25% visibility, while those outside the top 10 were seen in AI answers only 7.6% of the time on average. 

    Interestingly, there were diminishing returns when a domain was cited most frequently: brands whose domain received the most citations achieved an average of only 31% visibility — only a slight improvement over the top 10 brands — showing you can’t build a GEO strategy purely on owned content.

    10X your SEO with Semrush for Enterprise.

    The world’s most powerful SEO platform, purpose-built for Enterprise.

    Request demo

    Wikipedia and Reddit aren’t as important as you think

    Despite accepted wisdom that Wikipedia and Reddit are major contributors to GEO visibility, for B2B brands, they have a limited impact. For 60% of brands, Wikipedia didn’t make the top 25 cited websites, and for 30% it wasn’t cited in any answers to the prompts created. This could change as brands invest time and money in influencing Wikipedia content. It also suggests that building content for other platforms may be more important.

    Reddit was even less influential and wasn’t in the top 25 cited domains for more than four-fifths (83%) of the brands analyzed. This likely reflects the nature of B2B tech brands, which tend not to be discussed on Reddit as much as consumer brands.

    Because of commercial arrangements, ChatGPT has near-exclusive access to Reddit. If this AI model matters to you, it’s still worth considering a Reddit strategy — although you must have products or services the Reddit community will engage with and discuss.

    Getting visibility in AI answers isn’t just about written content

    Video should be a key part of any GEO strategy, as platforms like YouTube are used extensively by AI models. About 75% of brands had YouTube among the top 25 cited domains in AI answers.

    It’s no surprise Google’s Gemini model — including AI answers at the top of search pages — was most likely to cite a YouTube video. However, all AI models use video, meaning written content alone isn’t enough to ensure strong AI visibility.

    The return of the LinkedIn article?

    LinkedIn was surprisingly influential, appearing in the top 25 domain citations for 37% of brands. LinkedIn articles — even some posted a few years ago — are appearing in AI answers, despite LinkedIn itself apparently losing interest in the format. A strong LinkedIn presence is therefore critical for B2B brands, and writing long-form articles on the platform is a sensible GEO strategy.

    Owned media trumps earned

    Despite leaders of large AI companies promoting the authority of earned media, we found that owned media is critical in determining which B2B technology brands appear in AI answers. More than twice as many owned websites are cited as earned media sites. Although many of the owned sites will be competitors, B2B brands must create content for their website, their channel partners and other partner brands in the ecosystem if they want to be visible in AI answers.

    Interestingly, the earned media that impacts AI answers is frequently different from a brand’s “top tier” media list. Often, a small number of titles drive most citations of trade publications, and sometimes publications dismissed by PR specialists due to a perceived lack of influence over the brand’s audience had a substantial impact on the answers that the audience will receive when using AI.

    Given the hyping of the importance of earned media by the leaders of the AI companies, we would expect the balance between owned and earned media to move toward earned in the future.

    What this means for your GEO strategy

    Ensuring that your B2B brand is visible in the AI answers provided to your customers and prospects is critical for long-term success. Yet most answers don’t mention the leading brand, and most companies are only visible in a small percentage of AI responses. Building an effective GEO strategy, however, isn’t straightforward. The data don’t support much of the accepted wisdom promoted by GEO gurus.

    To be successful, B2B companies must conduct a detailed analysis of their market to understand the factors that influence the major AI models. This data-driven approach is critical, as this study highlights that accepted wisdom is frequently wrong.


    Contributing authors are invited to create content for MarTech and are chosen for their expertise and contribution to the martech community. Our contributors work under the oversight of the editorial staff and contributions are checked for quality and relevance to our readers. MarTech is owned by Semrush. Contributor was not asked to make any direct or indirect mentions of Semrush. The opinions they express are their own.

    Mike Maynard
    Managing Director, Napier

    Mike Maynard is a renowned PR professional based in Chichester, UK. He is the CEO and Managing Director of Napier, a leading PR agency that specializes in the technology sector. Since October 2023, he has been Deputy Chairman of Eurocom Worldwide. With over 30 years of experience in the industry, Mike has worked with some of the biggest names in tech, including ABB, Avnet, Microchip, Nokia, NTT and Yokogawa. 

    Mike is a self-confessed geek who loves talking about technology. He believes that combining the measurement, accountability and innovation that he learnt as an engineer with a passion for communicating ensures Napier delivers great campaigns and tangible return on investment. Mike began his career as an electronics design engineer, working for companies ranging from GEC-Marconi to DDA, developing products from complex radar systems to Kim Wilde’s mixing desk. Mike later joined IDT Inc, an American semiconductor company, making his move into marketing when he switched from European Applications Manager to European Marketing Manager.

    He acquired Napier in 2001 with his business partner, Suzy Kenyon. Since then, he has been instrumental in growing the agency into one of the most respected names in the tech PR industry, building a team of over 45 professionals and working with clients across Europe and North America. Napier’s unique value is walking the line between engaging storytelling and holding the team accountable for delivering tangible campaign results for clients.

    Mike is known for his expertise in B2B technology PR, content marketing, and integrated marketing communications. He is a regular speaker at industry events and hosts a B2B podcast focused on hot topics and important trends in PR and marketing such as AI, content marketing and the value of data in marketing campaigns. In addition to his work at Napier, Mike is a short track speed skater and keen cyclist.

    View Author Profile