Search Marketing FAQ

Podcasts that are transcribed and indexed (via show notes pages or third-party transcription) contribute to AI training data and retrieval. Niche podcasts with strong show notes earn citations on category queries. Appear as a guest on industry podcasts; each appearance generates a discoverable transcript. Most podcasts contribute via show notes pages rather than direct audio retrieval.

in AI Citation Sources

Three layers. Tier-1: Wikipedia, major news outlets, industry publications. Tier-2: Reddit, Quora, YouTube. Tier-3: niche forums, expert directories, podcast appearances. Each tier compounds AI engine citation. Build all three over 6-12 months. Quality matters more than quantity - one Wikipedia mention outweighs 100 low-quality directory listings.

in AI Citation Sources

Digital PR earns links and brand mentions from authoritative publications. Each mention compounds AI citation likelihood. Strategies: pitch original research, expert commentary via HARO, exclusive interviews, controversial-but-defensible takes that earn discussion. Modern digital PR is the highest-leverage AI authority builder beyond direct content publication.

in AI Citation Sources

Critical. Original surveys, proprietary data, first-hand testing, and exclusive interviews earn citations because AI engines prefer primary sources over derivative summaries. One primary-research piece often earns 50-100 third-party citations over its lifetime as other publishers cite your data. Investment in primary research compounds across AI engines and traditional SEO.

in AI Citation Sources

Survey your customer base on category questions. Mine internal data for industry-relevant insights. Run benchmark studies on competitors' products. Test products and publish results. Conduct expert interviews. The pattern: produce data others want to cite. Publish with clear methodology, visualizations, and downloadable raw data. Each primary research piece is a citation magnet.

in AI Citation Sources

Five elements. Clear methodology section explaining how data was collected. Sample size and date range. Charts and tables for AI extraction. Plain-text summary of key findings. Downloadable raw data (CSV or appendix). AI engines and human writers prefer data-driven content over opinion. Each citation grows category authority.

in AI Citation Sources

Academic papers (peer-reviewed studies, university research) are cited heavily by AI engines for technical and medical queries. Government sources (CDC, FDA, FTC, IRS) dominate regulatory and YMYL queries. Cite these in your own content to inherit their authority. Most brands underuse academic and government citation - it lifts E-E-A-T and AI citation likelihood significantly.

in AI Citation Sources

Hierarchy varies by query but typical weighting: peer-reviewed academic (highest for medical/scientific), government sources (highest for regulatory/YMYL), Wikipedia (high for general factual), major news (high for current events), industry publications (high for category authority), Reddit (high for personal experience), YouTube (high for visual/how-to), Quora (medium for professional Q&A), brand sites (variable based on E-E-A-T).

in AI Citation Sources

First-party data (your customer surveys, usage analytics, sales data) gives you authoritative claims no competitor can make. Publish data-backed reports periodically. AI engines cite first-party data because it's primary-source and unique. Brands with strong data publishing programs (Spotify Wrapped, Salesforce State of [X]) earn outsized AI citation share by becoming the source for category statistics.

in AI Citation Sources

Expert bylines with verified credentials lift E-E-A-T and AI citation likelihood. Each article should have a named author with linked bio, credentials, and Person schema. AI engines cite content from credentialed authors more often than anonymous or generic content. Build your expert team's byline portfolio over time - 10-20 articles per expert in 18 months.

in AI Citation Sources

Five elements. Visible About page with company history, leadership, mission. Author pages with credentials for each content contributor. Editorial guidelines documented. Citations to primary sources within content. Third-party recognition (industry awards, certifications, press mentions). Consistent NAP and brand identity across the web. E-E-A-T compounds; build slowly over 12-24 months.

in AI Citation Sources

llms.txt is the emerging convention (llmstxt.org spec) pointing AI engines to a structured site summary in markdown at /llms.txt. ai.txt is a separate proposal less widely adopted, often focused on training-data permissions. llms.txt has more momentum in 2026; ai.txt usage is sparse. Most sites should publish llms.txt and skip ai.txt.

in llms.txt & AI Bot Management

llms.txt summarizes a site's structure, key URLs, and content in markdown format that AI engines can ingest efficiently. Without llms.txt, AI engines crawl HTML and parse structure imperfectly. With it, the site explicitly signals which pages contain authoritative content. Improves citation accuracy for AI engines that respect the spec.

in llms.txt & AI Bot Management

No direct ranking impact. llms.txt is an AI-engine convention, not a Google ranking signal. Indirect benefit: better AI engine citation could drive more brand awareness and branded search, which compounds organic visibility. Treat llms.txt as a GEO tactic, not an SEO one. Adoption is still early.

in llms.txt & AI Bot Management

No. llms.txt is a hint to AI engines, not a directive. Adoption varies: some engines respect it (early adopters), most ignore it (still emerging). Publishing llms.txt is low-cost and may help over time as adoption grows. Pair with traditional content quality and E-E-A-T for actual citation impact.

in llms.txt & AI Bot Management

Place at the site root: yoursite.com/llms.txt. Same pattern as robots.txt and sitemap.xml. Some sites also publish llms-full.txt (full content dump) alongside llms.txt (summary). Both should be at the root level for AI engines to discover via convention.

in llms.txt & AI Bot Management

Markdown format per the llmstxt.org spec. Structure: H1 with site name, brief description, then sections (H2 headers) listing key URLs with brief descriptions. Example: '# Capconvert' followed by 'Search marketing agency...' then '## Services' with bulleted links. Keep concise; aim for under 5KB total file size.

in llms.txt & AI Bot Management

Five elements. H1 site name. Brief description (1-3 sentences). Key URLs grouped by topic (services, products, blog, about). Brief description per linked URL. Optional: contact info, ownership, primary categories. Skip: every page (use sitemap.xml for completeness), pricing details that change frequently, time-sensitive content.

in llms.txt & AI Bot Management

No. llms.txt summarizes the site - it's a curated landing index, not a complete URL map. List highest-value pages (key services, top blog posts, About, Contact). Use sitemap.xml for comprehensive URL listing. llms.txt is the AI-engine analog of a homepage navigation menu, not a full content inventory.

in llms.txt & AI Bot Management

Update when site structure changes significantly: new top-level pages, major product launches, brand-name changes, restructured navigation. Monthly review is reasonable for active sites. Quarterly for stable sites. llms.txt doesn't need real-time freshness - it's a long-lived summary, not a feed. Most updates should be additive.

in llms.txt & AI Bot Management

robots.txt controls bot access (which URLs to crawl/avoid). llms.txt provides an AI-friendly site summary (what content exists). They serve different purposes and should coexist. robots.txt is required and respected by most reputable bots. llms.txt is optional and recognized by a smaller set of AI engines. Publish both.

in llms.txt & AI Bot Management

No. llms.txt is descriptive (what content exists), not directive (what bots can access). Use robots.txt to allow or deny specific AI bot user agents. llms.txt complements robots.txt without replacing it. Common stack: robots.txt for access control + llms.txt for content summary + sitemap.xml for URL completeness.

in llms.txt & AI Bot Management

Add to robots.txt: 'User-agent: GPTBot' on one line, 'Disallow: /' on the next. This blocks OpenAI's training-data crawler from accessing any URL. Note: blocking GPTBot does NOT prevent ChatGPT from citing your content via its search feature (OAI-SearchBot is separate). Most sites should leave GPTBot allowed.

in llms.txt & AI Bot Management

In robots.txt, explicitly allow GPTBot ('User-agent: GPTBot' then 'Allow: /') and disallow specific other bots ('User-agent: Bytespider' then 'Disallow: /'). Order matters - put specific rules before general fallbacks. Test with Google's robots.txt tester or curl simulating each user agent.

in llms.txt & AI Bot Management