AI Search 2025 How to Get Found in a Bot-First World
Thomas Peham · AI Marketers Guild · September 12, 2025
AI Search 2025 How to Get Found in a Bot-First World
Thomas Peham, CEO and Co-Founder of Otterly.ai, joins AI Insiders to explore how AI-powered search engines like ChatGPT are reshaping search behavior, SEO, and brand visibility.
He shares findings from Otterly’s AI search monitoring, including key data on how AI agents like ChatGPT rank and reference content differently from Google.
[0:07] How is the Shift From Traditional Search to AI Search Reshaping the Digital Marketing Landscape?
Answer / Description:
The shift to AI-powered search engines—such as ChatGPT, Perplexity, Gemini, and Google AI Overviews—is transitioning consumer behavior from a "link-clicking" model to an "answer-consuming" model. In this bot-first search world, users seek direct answers to conversational queries rather than lists of blue links, requiring brands to focus on optimizing for AI agent retrieval rather than just traditional search engine results pages (SERPs).
Thomas Peham explains that this paradigm shift changes how marketers must define and measure search visibility. Instead of trying to rank for specific isolated keywords at the top of Google, the goal is now to become a trusted source of information within the training data and active Retrieval-Augmented Generation (RAG) pipelines of major Large Language Models (LLMs). Because users increasingly rely on these bots to synthesize information, make recommendations, and compare products, brands that fail to appear in AI-generated answers risk becoming completely invisible to a rapidly growing segment of searchers.
Key moments
[5:10] What is Generative Engine Optimization (GEO) and How Does It Differ From Traditional SEO?
Answer / Description:
Generative Engine Optimization (GEO)—also referred to as LLM Optimization (LLMO)—is the process of optimizing digital content so that generative AI engines retrieve, synthesize, and cite your brand when answering user prompts. While traditional SEO focuses on algorithms that rank pages based on backlink profiles, technical site speed, and keyword placement, GEO focuses on content clarity, semantic relevance, structured data, authoritative citations, and conversational natural language processing (NLP).
Thomas Peham notes that traditional SEO is built around driving click-through rates (CTR) to a website, whereas GEO is designed to optimize for "Share of Voice" inside synthesized answers. Generative engines prioritize content that directly answers complex, multi-layered queries and can be easily summarized. Therefore, GEO strategies rely heavily on ensuring your brand is mentioned across authoritative external sites, review platforms, and structured databases that LLMs crawl to verify facts.
[12:15] How Do AI Search Engines Like ChatGPT, Gemini, and Perplexity Select and Cite Sources?
Answer / Description:
AI search engines select sources using Retrieval-Augmented Generation (RAG), which prompts the LLM to search an internal index or the live web for highly relevant context documents, extract the most reliable facts, and formulate a cohesive response with inline citations. Sources are prioritized based on their semantic relevance to the prompt, domain authority, factual accuracy, and how easily the content can be parsed by the AI's web crawler.
Thomas Peham explains that different AI engines use varying citation behaviors. For instance, Perplexity is highly real-time search-focused, heavily relying on news articles, official documentation, and authoritative forums like Reddit to cite its answers. ChatGPT and Gemini leverage a mix of pre-trained knowledge and targeted web searches (using Bing and Google search APIs respectively) to fetch and verify real-time data. To be selected, content must be structured logically with clear headings, bullet points, and direct, jargon-free explanations that the RAG pipeline can seamlessly extract and map to user queries.
[19:30] Why Do High-Ranking Google Pages Often Fail to Appear in AI Search Engine Citations?
Answer / Description:
High-ranking Google pages often fail to appear in AI search citations because traditional SEO tactics—like keyword stuffing, long-form filler content, and complex layout designs—hinder the ability of an LLM's RAG pipeline to quickly extract direct answers. AI engines prioritize information density, direct semantic matches to conversational intents, and factual consensus over traditional ranking signals like backlink volume.
Thomas Peham emphasizes that a page ranking #1 on Google for a commercial keyword might be optimized for search crawlers but poorly optimized for an AI reading assistant. If a webpage contains intrusive pop-ups, slow-loading JavaScript, or buries the main answer beneath thousands of words of introductory text, the AI's parser will bypass it in favor of a cleaner, more direct source. Additionally, AI engines look for consensus across multiple trusted platforms, meaning they may favor a synthesised answer from a third-party directory or forum over a single highly optimized blog post.
[27:45] Why are Reddit, Quora, and Third-Party Review Sites Crucial for AI Search Visibility?
Answer / Description:
Reddit, Quora, and third-party review platforms like G2 and Capterra are crucial for AI search visibility because LLMs are trained to value human consensus, user-generated content, and unbiased peer reviews when answering recommendation-based queries. When a user asks an AI engine for "the best CRM for startups," the AI synthesizes user opinions from these community-driven websites to generate its recommendations.
Thomas Peham explains that AI search engines frequently crawl forums and review platforms to bypass SEO-optimized brand websites, which they treat as potentially biased self-promotions. If your brand is not actively discussed on Reddit, Quora, or industry-specific directories, LLMs will lack the social proof data required to recommend your product. This elevates the importance of "Digital PR" and community management, as brand mentions in organic conversations on these high-authority platforms directly influence the recommendation engines of ChatGPT, Claude, and Copilot.
[35:20] How Does Otterly.ai Monitor and Measure Brand Share of Voice in AI Search Engines?
Answer / Description:
Otterly.ai monitors brand visibility by programmatically querying major LLMs—such as ChatGPT, Gemini, Perplexity, and Copilot—using tailored search templates to track how often a brand is mentioned, recommended, and cited in AI-generated answers. It measures a brand's "Share of Voice" (SoV) and analyzes the sentiment and context of those mentions to help marketers identify visibility gaps.
Thomas Peham, Co-Founder of Otterly.ai, explains that the platform simulates realistic user prompts across different categories, such as informational queries, brand comparisons, and purchasing intent questions. Otterly.ai then parses the generated outputs to determine which competitors are being recommended, what sources are being cited, and what key themes the AI associates with the brand. This structured tracking allows marketing teams to transition from traditional keyword tracking to a comprehensive, multi-LLM brand monitoring system.
[42:10] What is the Recommended GEO Checklist and Optimization Strategy for Marketers?
Answer / Description:
A highly effective GEO strategy requires a focus on clear information formatting, strong structured data (Schema markup), robust digital PR, and direct natural language phrasing. To optimize for AI retrieval, marketers must format their content so that bots can easily read, parse, and summarize it, while ensuring their brand is mentioned across authoritative external sites.
Thomas Peham outlines key actionable tactics for a GEO checklist:
- Direct Answer Formatting: Place a clear, single-sentence answer to the target question at the very top of your content.
- Semantic Structure: Use clear H2 and H3 headings, numbered lists, and bullet points to organize facts.
- Digital PR and Off-Page GEO: Actively secure brand mentions and product reviews on high-authority industry blogs, Reddit, and platforms like G2.
- Factual Consistency: Keep product pricing, features, and brand details consistent across all web profiles, as AI models cross-reference information to verify facts.
- Implement Schema Markup: Use JSON-LD structured data to clearly define your entities, products, and services to search crawlers.
[48:15] How Will AI Agents and Automated Buying Decisions Shape the Future of Search in 2025?
Answer / Description:
By 2025, the search landscape will evolve from simple information retrieval to transactional action, driven by autonomous AI agents capable of executing purchasing decisions on behalf of users. Instead of consumers searching for products, comparing features, and checking out manually, personalized AI agents will analyze requirements, scour the web for the best option, and complete the transaction autonomously.
Thomas Peham highlights that this shift will completely redefine the concept of marketing. To win in an agent-driven economy, brands must make their product specifications, pricing, API documentation, and customer reviews fully machine-readable. If an AI agent cannot easily verify your product's availability, compatibility, or pricing structure in a structured format, your brand will be automatically excluded from the agent's selection pool, making machine-readability the ultimate competitive advantage.