AI SEO and the Power of Brand Mentions

How Reddit, G2 and Other Review Sites Influence Large Language Models

Search is changing. For decades, search engine optimisation (SEO) revolved around keyword rankings and backlinks. Today those classic metrics still matter, but they no longer dominate the conversation. In 2025, people increasingly interact with AI‑powered search tools—Google’s AI Overviews, Bing’s “Copilot” results, ChatGPT and competitors such as Perplexity— that synthesize answers instead of simply listing web pages. These systems are driven by large language models (LLMs) that learn by ingesting vast amounts of content. As a result, a new discipline, AI SEO, is emerging: optimising your brand for citations and mentions in AI‑generated answers.

This article explains what AI SEO is, why brand mentions on platforms like Reddit, G2, Capterra and other review sites matter, and how you can position your company to be surfaced in AI‑generated answers. It draws on recent data and research published in 2024–2025, and it illustrates how YXR (a Reddit marketing agency) can help brands adapt to this changing landscape.

---

From Traditional SEO to AI SEO

Traditional SEO focuses on improving a website’s visibility in search engine results pages (SERPs) through on‑page optimisation, backlinks and technical factors. The goal is to get as high as possible in the “blue links” list of results. AI‑generated search results upend that model. Google’s AI Overviews (AIO) are answer boxes summarising information from multiple sources; they appear at the top of the SERP and can answer the user’s question without a click. G2 reports that AI Overviews were triggered for about 13.14 % of queries in March 2025, a sharp rise from 6.49 % in January. When AI Overviews appear, click‑through rates (CTR) drop by 34.5 % for the top organic listing, which means that even ranking first in the blue links yields far less traffic when an AI answer is present.

AI models often pull sources that are not in the top organic results. BrightEdge’s 2024 analysis found that 66 % of the URLs cited in AI Overviews do not rank in the traditional search results. In e‑commerce queries, about 75 % of cited sources are not ranking organically. In education queries, more than 90 % of citations are from pages that do not rank on page 1. These findings highlight a fundamental change: AI Overviews anticipate the next question and pull information from pages that answer related queries, not just the highest‑ranking pages. To influence AI search, brands must think beyond their own domains and consider how they are talked about across the web.

---

How LLMs Learn and Why Brand Mentions Matter

Large language models are not search engines; they do not crawl and index pages the way Googlebot does. Instead, they learn by consuming vast corpora of text and extracting statistical patterns and associations. Fatjoe’s analysis explains that LLMs “read and interpret massive volumes of text, building associations between words, topics and brands; if your brand name appears frequently in positive, relevant contexts, the AI starts to associate your business with certain topics or industries.” In other words, brand mentions act like implicit links: signals that help LLMs understand how your brand relates to a subject.

Not all mentions carry equal weight. High‑quality citations in respected publications, niche blogs and curated listicles reinforce authority. Off‑topic or negative mentions may be ignored or even harm sentiment. The quality and sentiment of reviews, comments and discussions therefore influence the associations that LLMs form around your brand.

---

Evidence that rankings correlate with LLM mentions – but they’re not everything

Seer Interactive ran a large‑scale study in early 2025 to determine what drives brand mentions in AI answers. They pulled 300,000+ keywords in finance and SaaS, extracted nearly 600,000 “People Also Ask” questions and queried the OpenAI GPT‑4o API to see which brands were mentioned. Their findings:

• Brands ranking on page 1 of Google had a strong correlation (~0.65) with LLM mentions (Seer Interactive).
• Bing rankings had a moderate correlation (~0.5–0.6) (Seer Interactive).
• Backlinks had weak or neutral impact (Seer Interactive).
• Content variety (images, videos, multi‑modal content) did not significantly increase mentions (Seer Interactive).

Seer’s authors emphasised that rankings are only part of the story; factors such as on‑page quality, PR features, partnerships and user reviews are likely to play important roles (Seer Interactive).

A Reddit post discussing an Ahrefs data analysis makes a similar point: websites with more organic traffic are mentioned more often in AI search (Reddit). Participants noted that LLMs are probabilistic and therefore cite sites that “appear more frequently across the web within a given niche” (Reddit). In other words, popularity begets more mentions.

---

User‑Generated Content (UGC) and Review Platforms as AI Training Data

Reddit – the largest source of citations for AI Overviews

A study by Profound found that Reddit is the single largest source of citations for Google AI Overviews (21 %), followed by YouTube (18.8 %) (G2). G2’s article notes that this reflects a strategic shift: authentic, user‑generated stories carry weight because AI systems aim to capture diverse perspectives and personal experiences. Jessica Hill of Zendesk notes that her team is exploring how to utilise these platforms to increase visibility (G2). Leandro Perez of Salesforce says his marketing team amplifies user stories rather than sanitising them because that authenticity builds buyer trust (G2). Google emphasises that its systems reward content created “to help people, not content made to rank well” (G2). This is equally true for AI Overviews.

Reddit’s influence is growing rapidly. In February 2024, Google announced an expanded partnership with Reddit that includes new licensing agreements and direct API access (Foundation). This partnership is already driving more traffic to Reddit: Google’s Q3 2024 investor report noted that the platform averaged 97.2 million daily active users (Foundation). Machine translation has allowed Reddit content to reach more languages (Foundation). Sistrix data shows that Reddit is now the third most visible domain in Google’s US search index, with a click‑potential of more than 573 million monthly clicks (Foundation). The number of queries where Reddit ranks in the top ten is increasing, and the platform holds over 10 million top‑10 keywords and nearly 20 million top‑100 keywords (Foundation). Review sites like G2, Capterra and TrustRadius have lost significant traffic after Google’s Helpful Content update, while Reddit has surged (Foundation). G2’s traffic reportedly dropped from 5.4 million to 1 million visitors, Capterra from 1.4 million to 990,000, and TrustRadius from 850,000 to 310,000 (Foundation). This shift means that UGC communities may outrank polished review sites for bottom‑of‑the‑funnel (BOFU) keywords (Foundation).

G2, Capterra and review sites in flux

G2, Capterra, TrustRadius and other software review sites still play a role in B2B buying decisions. However, they are no longer as dominant in AI search results. According to the Foundation report, third‑party review sites have experienced steep traffic declines and are often outranked by Reddit for important queries. Part of the reason is that LLMs learn associations from across the web rather than weighting heavily curated review content. Traditional review platforms may also appear less authentic or less up‑to‑date compared with real user discussions. That doesn’t mean you should ignore them—positive reviews still contribute to trust—but it underscores the need to diversify your presence beyond review aggregators.

Why user reviews and sentiment matter

Fatjoe emphasises that positive brand mentions in user reviews, testimonials and third‑party assessments help AI tools understand not just what your brand does but how people feel about it. Sentiment therefore becomes a ranking factor in the AI era. Conversely, negative sentiment or a lack of reviews can harm brand associations.

---

AI SEO Tactics: How to Get Your Brand Cited in AI Answers

1. Understand how AI Overviews select sources

Google’s AI Overviews are built on patents that outline how sources are chosen. The system uses a combination of query‑independent measures (popularity, trustworthiness, freshness), query‑dependent measures (relevance to the topic) and user‑dependent measures (past searches and location) (Foundation). Multiple LLMs interpret the query’s context and gather content from relevant pages before a specialised model condenses this data into a single answer (Foundation). It then learns from user interactions and adjusts what it surfaces (Foundation). Optimising for AI Overviews therefore involves building trust and relevance, ensuring your content is up‑to‑date and being mentioned in contexts where AI systems can easily extract your brand.

2. Build high‑quality, context‑rich mentions

According to Fatjoe, brand mentions can now be more valuable than backlinks. Structured content such as “Top 10” lists, comparisons and roundups are often the first resources AI tools reference. Being featured in listicles and reviews increases the chance that your brand will be recommended by AI overviews. But quality matters: mentions in respected or niche publications are more impactful than those on low‑quality sites. Consider digital PR campaigns to secure features and interviews, and use guest posts or outreach articles to reinforce your brand identity.

3. Leverage user‑generated communities (Reddit, Quora, Discord)

UGC sites are where authentic conversations happen. The G2 article notes that Reddit provides 21 % of AI Overview citations, and Foundation’s study shows that Reddit’s SERP visibility is soaring. To influence these conversations:

• Create native content on Reddit. Foundation recommends identifying relevant subreddits, paying attention to their rules, and focusing on education over promotion. Offer how‑to posts, best practices, or case studies rather than direct pitches. Use an authentic voice and include personal stories to build credibility.
• Engage in comments. The success of a post often depends on follow‑up engagement. Answer questions, provide additional insights and show gratitude for feedback. Active participation increases the likelihood that your posts will stay visible and be cited.
• Monitor share of voice. Tools like STAT or Semrush can track how often Reddit pages appear for your keywords and how AI Overviews are affecting your SERP footprint. This data informs where to allocate resources.

4. Optimise on‑site content for AI parsers

G2’s article argues that we now write for three audiences: humans, search crawlers and AI parsers. AI parsers look for content that is structurally summarizable: concise paragraphs, descriptive headings, bullet points and schema markup. Semantic SEO and structured data help AI understand your content’s meaning. Abhishek G.P., Vice President of Growth at Atlan, says his team now thinks “like educators and product storytellers” to be viewed as a trusted source.

5. Prioritise authenticity via UGC and reviews

AI Overviews prize authenticity. Industry leaders from Zendesk and Salesforce highlight that the most persuasive stories come directly from customers, and that marketing teams are amplifying real‑world experiences rather than sanitising them. Encourage satisfied customers to share experiences on Reddit, G2, Trustpilot or your own forums. Respond to reviews—both positive and negative—to show that your brand cares about feedback.

6. Monitor AI citations and measure influence

As Ross Briggs (G2) notes, the ROI of search marketing is becoming more abstract because AI Overviews may yield brand impressions and sentiment rather than direct clicks. Monitoring AI citations is therefore crucial. Keep an eye on how often your brand appears in AI answers and which queries trigger these mentions. Adjust your content strategy accordingly. Tools like BrightEdge’s Generative Parser, SeerSignals or AI monitoring services can help track citations across AI platforms.

7. Address multiple questions and adjacent topics

BrightEdge’s analysis shows that AI Overviews anticipate related questions. Long‑form, comprehensive content that answers primary and secondary questions increases the chances of being cited. For example, an article about “best CRM tools” should also cover pricing, integration options, alternatives and user sentiments. This multi‑faceted approach helps AI models find more relevant snippets to cite.

8. Build E‑E‑A‑T and partner with trusted sources

Experience, Expertise, Authoritativeness and Trustworthiness (E‑E‑A‑T) remain important. BrightEdge notes that trusted and authoritative websites are more likely to be cited. Seek partnerships with reputable sites in your industry. For YMYL (Your Money or Your Life) queries, such as finance or healthcare, citations often come from highly authoritative domains. Aligning with respected sources increases your credibility in the AI era.

---

The New Currency: Brand Mentions Over Backlinks

How brand mentions outperform links in AI SEO

Historically, backlinks were the primary off‑page signal. AI search is shifting that priority. Andy Crestodina of Orbit Media Studios tells G2 that AI Overviews are making marketers think about brand mentions instead of links; marketers must focus on building a larger digital footprint and being included in directories and forums. Ross Briggs adds that brand mentions and sentiment within AI answers have tremendous value even without a link. Because AI models learn through language, increasing the frequency and positivity of your brand mentions helps train AI on your expertise and relevance.

Fatjoe summarises the new approach: Brand mentions “act as signals to AI that your brand is trustworthy, authoritative and worth recommending.” They note that structured listicles and product roundups are particularly effective at earning citations, and that digital PR and strategic link building help amplify these mentions. Their research emphasises that Reddit and Quora are the two most cited domains in Google’s AI Overviews; building positive sentiment on these platforms is therefore vital.

Managing sentiment and quality

With brand mentions as the new currency, sentiment management becomes a core function of SEO. Encouraging customer reviews, responding to feedback and providing support in public forums signals that your brand cares. Negative or unaddressed criticism can influence AI models to associate your brand with problems rather than solutions.

---

YXR.io: Helping Brands Thrive in the AI Search Era

YXR is a Reddit marketing agency that specialises in building authentic brand presence in the communities that matter. Our team understands the unique culture and rules of different subreddits, and we design strategies that educate, engage and build trust rather than spam. With AI Overviews now pulling heavily from Reddit, a robust and positive presence there is essential.

Why work with YXR?

• Experience in UGC channels: We’ve been helping brands build communities on Reddit, Discord and other forums long before AI Overviews put them in the spotlight.
• Data‑driven approach: We monitor your brand’s share of voice in SERPs and track how often it’s mentioned in AI answers. This allows us to adapt strategies quickly.
• Content and engagement: We craft native posts, run AMAs (Ask Me Anything), respond to questions and nurture discussions that provide value. Our focus is on education over promotion, which aligns with AI systems’ preference for helpful content (Foundation).
• Reputation management: We help you manage reviews and sentiment across Reddit, G2, Capterra and other platforms. Positive engagement drives the associations AI models form about your brand.

AI SEO is not a passing trend. As AI‑driven search experiences become mainstream, brands that cultivate authentic mentions and expertise across the web will stand out. YXR can guide your organisation through this new landscape and ensure your brand is part of the conversation.

---

Conclusion: Preparing for the Future of Search

AI search has already changed how people discover information. AI Overviews appear for a growing share of queries (G2), pull sources that often aren’t ranking organically (BrightEdge), and prioritise authentic, helpful content (G2). Reddit and other UGC platforms are now among the most cited sources (G2), while traditional review sites have lost visibility (Foundation). Studies show that search rankings correlate with LLM mentions (Seer Interactive) but do not guarantee them; backlinks are less influential (Seer Interactive) and multi‑modal content alone doesn’t move the needle (Seer Interactive). Instead, brand mentions, sentiment and authenticity are becoming the new currency.

To succeed in this environment, marketers should build high‑quality mentions across diverse platforms, optimise content for AI parsers, participate authentically in communities like Reddit, and monitor AI citations. YXR’s Reddit marketing expertise can help brands navigate this complex landscape and turn conversations into citations. By investing in AI SEO now, you’ll be prepared for a search world where your brand’s reputation and influence matter more than its rank.

---

Beyond Today: Trends Shaping AI SEO

The AI search landscape is still developing. Several trends suggest where AI SEO may head in the coming years:

Generative Engine Optimization (GEO)

Foundation’s content team coined the term Generative Engine Optimization (GEO) to describe the process of optimising your presence for AI‑powered answer engines, not just search engines. GEO involves monitoring where your brand is cited, building content that anticipates related questions and investing in the communities that AI systems draw upon. It requires tracking AI Overviews, Reddit posts, Q&A sites and other generative results (Foundation). Tools like STAT, SeerSignals and AI Monitor are emerging to help marketers measure their share of voice in AI answers. Brands that succeed with GEO will be those that adapt their content strategies to emerging metrics such as share of AI citations, sentiment within AI responses and mention velocity (G2).

The role of machine translation and global reach

AI models operate across languages. Reddit’s machine‑translation programmes, for example, led to an influx of users from France, Spain, Germany, Brazil and Italy (Foundation). As more non‑English content becomes indexed, brands must consider multilingual strategies. A positive mention on a French subreddit may eventually influence English AI answers, because LLMs incorporate translated content. Building brand mentions in multiple languages and cultures will be essential for global reach.

Hallucinations and trust signals

One challenge with AI search is the risk of “hallucinations” – AI systems generating incorrect or fabricated information. AI Overviews attempt to mitigate this by citing authoritative sources, but the prevalence of hallucinations has made trustworthiness and E‑E‑A‑T signals more critical than ever. Ensuring that your content is factually accurate, cited and updated reduces the risk of being misrepresented. Partnering with authoritative organisations or industry bodies can also bolster trust.

Emerging platforms and channels

While Reddit and G2 dominate AI citations today, other communities are gaining influence. Discord servers, niche forums and community Q&A platforms provide rich conversations that AI models may soon tap. The Pew Research Center, for instance, found that government (.gov) websites and educational (.edu) domains see higher proportional representation in AI answers than in classic SERPs (G2). Keeping an eye on where your audience congregates online will help you maintain presence across evolving channels.

Ethical considerations

As marketers strive to earn mentions in AI outputs, it is important to do so ethically. Avoid astroturfing (creating fake positive reviews or scripted discussions) because such tactics undermine trust and can backfire if exposed. Respect the rules of each subreddit and community, and prioritise transparency when posting as a brand. The AI era rewards authenticity; misleading tactics may lead to long‑term damage.

---

Practical Checklist for AI SEO

To conclude, here is a practical checklist to help you kick‑start your AI SEO initiatives:

• Audit your mentions: Use tools to identify where your brand is already mentioned across Reddit, G2, review sites, blogs and news. Analyse sentiment and context.
• Enhance your content: Update existing articles to answer not only the main query but adjacent questions. Include data, comparisons and user perspectives. Use structured formatting, schema markup and concise paragraphs to make content AI‑friendly (G2).
• Engage in target communities: Identify subreddits, Q&A boards and social platforms where your audience gathers. Participate authentically, provide value and build relationships. Encourage satisfied customers to share their experiences.
• Leverage review sites wisely: Maintain profiles on G2, Capterra and other review platforms. Request honest reviews, respond to feedback and highlight user satisfaction. Use badges or widgets from these sites to reinforce trust signals on your own website.
• Monitor AI search features: Track when AI Overviews appear for your keywords, how often your brand is cited and what content gets sourced. Adjust your strategy based on these insights.
• Collaborate with a specialist: Consider partnering with a Reddit marketing agency like YXR to navigate platform nuances, manage community engagement and monitor AI citations.

By following this checklist and staying aware of broader trends, you position your brand not only for traditional SEO success but also for visibility in the AI‑driven search experiences that are redefining how users discover and evaluate solutions.