Generative engine optimization (GEO) is the discipline that emerged in 2023–2025 as generative AI engines began functioning as a primary discovery surface. Where classical SEO optimizes for the ranked list of links, GEO optimizes for inclusion in the synthesized answer itself — the citation appearing inline in a ChatGPT response, the source link in a Perplexity answer, the brand name surfaced in a Google AI Overview. The core levers of GEO overlap with SEO but weight differently: **Structured directness.** Generative engines preferentially cite sources whose answer to the underlying query is clearly stated in the first 100–300 words. Long preambles before the answer reduce citation likelihood significantly. **Named expertise.** Author bios with verifiable credentials (named lawyers, doctors, engineers, certified practitioners) outperform anonymous content. The author entity matters to the LLM's confidence in citing the source. **Original data and analysis.** Aggregated content is rarely cited; the original source it aggregated from often is. Publishing original Canadian-context survey data, client-portfolio analysis, or proprietary calculations creates citation magnetism. **Schema markup.** Article, FAQPage, HowTo, Person, Organization, and Service schema help the LLM parse entities cleanly and cite confidently. **Crawler access.** Generative engines respect robots.txt directives. Sites blocking GPTBot, ClaudeBot, PerplexityBot, or Google-Extended will not be cited by the corresponding engines. Audit and unblock the crawlers you want citations from. **Existing citation graph.** LLMs partly seed their citation behaviour from the citation graph of the open web — Wikipedia, government sites, major media. Earning citations from established sources compounds GEO visibility. GEO is not a replacement for SEO; it is an extension. The two disciplines share most foundations (technical health, content quality, authority signals) but diverge on a handful of structural choices that materially affect AI citation. Canadian businesses building modern SEO programs in 2026 should treat both surfaces as primary.