SEO + GEO · Generative engines

GEO · Generative Engine Optimization for ChatGPT, Perplexity, Gemini and AI Overviews

GEO (Generative Engine Optimization) methodology so your brand gets cited when ChatGPT, Perplexity, Gemini, Claude or Google AI Overviews answer your buyers. Less noise, more levers: enriched schema, llms.txt, citability and factual consistency.

  • +3x citations in generative engines within 90 days
  • 0 placeholders: citable content from the first paragraph
  • llms.txt + schema @graph deployed on every key page
  • Real measurement: brand mentions, share of AI voice and referral traffic
Request GEO audit
See methodology

GEO (Generative Engine Optimization) is the discipline that optimizes websites, structured data and brand signals so that generative engines (ChatGPT, Perplexity, Gemini, Claude, Bing Copilot and Google AI Overviews) cite your content when answering your customers. Unlike classic SEO —focused on ranking URLs in a list of 10 blue links— GEO works to make your brand appear as a source cited inside the generated answer. The main lever is citability: short, verifiable passages with concrete data that LLMs can extract without rewriting.

GEO vs classic SEO

Same goal, different physics

Classic SEO optimizes for a ranking. GEO optimizes so an LLM chooses to cite you inside its answer. The signals, the content format and the measurement all change.

DimensionClassic SEOGEO
GoalSERP position (10 blue links)Verbatim citation inside the LLM’s answer
Unit of successOrganic clickBrand mention + verifiable citation
Winning formatURL + meta + H140-80 word passage with citable data
Key signalsBacklinks, CTR, dwell timeFactual consistency, schema, llms.txt, sameAs
CrawlersGooglebot, BingbotGPTBot, PerplexityBot, Google-Extended, ClaudeBot, OAI-SearchBot
MeasurementClicks, impressions, average position (GSC)Share of AI voice, citations/month, LLM referral traffic
RiskRanking dropHallucination, wrong citation, attribution to a competitor

Operational takeaway: it’s not GEO instead of SEO, it’s GEO on top of SEO. The technical foundation (crawlability, Core Web Vitals, useful content) is taken for granted; the new levers live above it.

Prioritized signals

The 8 levers that move citations in generative engines

Stack prioritized by observed impact in GEO audits on B2B accounts. Execution order: top first.

  1. 01

    Content citability

    Rewrite key passages in 40-80 word blocks with data, source and clear entity. Always answer-first in the opening paragraph: “X is…”. LLMs extract what is citable, not what is flowery.

  2. 02

    Verifiable E-E-A-T

    Human authorship with a real bio, public LinkedIn, credentials and sameAs. At cronuts.digital the byline belongs to Albert Puig (Co-founder · CPO) and the CEO Borja Planells, with Person schema linked to LinkedIn and Crunchbase.

  3. 03

    Enriched @graph schema

    Full JSON-LD: Organization, Person, WebPage, Article, Service, FAQPage, BreadcrumbList and speakable. Schema acts as ground truth for the LLM when the text is ambiguous.

  4. 04

    llms.txt deployed

    Machine-readable map of the site designed for LLMs: index of canonical URLs, semantic description, editorial priority. It complements robots.txt and sitemap.xml, it does not replace them.

  5. 05

    Consistent brand mentions

    Mentions in industry press, directories (Crunchbase, G2, LinkedIn company), podcasts and guest posts. Generative engines learn the brand↔category association by frequency and context.

  6. 06

    Cross-site factual consistency

    Same NAP, same services, same claims on web, LinkedIn, Crunchbase, GMB and press releases. A single contradiction lowers the probability of citation.

  7. 07

    Speakable + FAQPage schema

    Explicitly mark which passages are direct answers to a question. AI Overviews and voice-first answers prioritize content with SpeakableSpecification.

  8. 08

    Controlled access for AI crawlers

    Explicitly allow GPTBot, PerplexityBot, Google-Extended, ClaudeBot and OAI-SearchBot in robots.txt. Blocking by omission = not existing in generative.

Platforms covered

6 generative engines where we work your visibility

Each engine has its own crawler, its own citation criteria and a different answer surface. The strategy is tuned engine by engine.

  • ChatGPT

    ChatGPT web and search

    Crawler GPTBot + OAI-SearchBot. Prioritizes sources with clear authorship, Article schema and cross-site consistency. Cites with a direct link in search mode.

  • Perplexity

    Perplexity

    Crawler PerplexityBot. Answer engine with mandatory citation on every sentence. Rewards sources with concrete data, explicit dates and FAQ structure.

  • Gemini

    Google Gemini

    Crawler Google-Extended. Leverages the Google index + E-E-A-T signals. Solid classic SEO is a precondition, not a bonus.

  • Claude

    Anthropic Claude

    Crawler ClaudeBot. More conservative and citable answers. Values complete schema and the absence of factual contradictions.

  • AI Overviews

    Google AI Overviews

    Generative answer inside the SERP. Powered by the classic index + Gemini. Passages with speakable and FAQPage raise the probability of appearing in the summary.

  • Copilot

    Microsoft Bing Copilot

    Powered by the Bing index + GPT-4/5. Crawler Bingbot. Citation card with a direct link. Relevant in B2B enterprise and regulated sectors.

cronuts methodology

5 phases to go from invisible to cited in generative engines

Sequence proven on B2B accounts in professional services and SaaS. Each phase has an auditable deliverable.

  1. 01

    Citability and AI visibility audit

    We measure your current share of AI voice: how many times ChatGPT, Perplexity, Gemini and Claude cite you against your competitors, on your cluster of critical prompts. Numerical baseline before moving anything.

  2. 02

    Schema @graph + speakable

    We deploy enriched JSON-LD on key pages: Organization, Service, FAQPage, Article, BreadcrumbList and speakable. Validated with Rich Results and the Schema.org validator.

  3. 03

    llms.txt + AI crawler control

    We publish /llms.txt with the editorial map of the site and update robots.txt with explicit access for GPTBot, PerplexityBot, Google-Extended, ClaudeBot and OAI-SearchBot.

  4. 04

    Content rewrites with citability

    We rewrite the key passages (hero, TL;DR, definitions, FAQs) with answer-first pattern, verifiable data and tagged entities. Each pillar page passes the SEO+GEO top-1 checklist (18 points).

  5. 05

    Brand mentions and continuous measurement

    Plan of mentions in industry press, directories and podcasts. Monthly dashboard with share of AI voice, citations/month and referral traffic from generative engines. Monthly iteration.

Tech stack

Tools we use to audit and deploy GEO

  • AI audit

    Batch prompts

    Systematic tests on ChatGPT, Perplexity, Gemini and Claude over the client’s cluster of critical prompts.

  • Schema

    JSON-LD + Rich Results

    Generation, validation and deployment of @graph schema via ACF/WPML or its equivalent in the client’s CMS.

  • Crawlers

    Logs + robots.txt

    Log analysis to detect visits from GPTBot/PerplexityBot/Google-Extended and tune access.

  • Brand

    Mention tracking

    Scraping of mentions in press and directories. Entity matching and consistency check on NAP/claims.

  • Search Console

    GSC + AI Overviews

    Tracking of impressions and clicks from AI Overviews inside Search Console, cross-referenced with GA4.

  • Reporting

    Monthly dashboard

    Notion/Looker deliverable with share of AI voice, citations/month, brand mentions and LLM referral traffic.

Does your brand show up when ChatGPT answers your buyers?

If the answer is “I don’t know,” then it doesn’t show up. We measure it for you in an initial GEO audit and deliver the 90-day plan to go from invisible to cited.

Email info@cronuts.digital
Book a diagnosis