Search Engine Optimisation and Generative Engine Optimisation
Search has changed more in the last two years than in the previous ten. The strategies that drove organic visibility in 2022 are still relevant, but they are no longer sufficient on their own. Alongside traditional search engine optimisation, there is now a parallel discipline concerned with how AI-powered answer engines discover, understand and cite content. Getting both right, at the same time, from the same strategic foundation, is what this page is about.
Search strategy
Two disciplines, one strategic foundation
Search Engine Optimisation
Search Engine Optimisation is the practice of structuring a website and its content so that search engines can crawl it efficiently, understand what it is about, and rank it appropriately for relevant queries. It operates at three levels. Technical SEO addresses the structural and code-level factors: site speed, crawlability, indexation, canonical tags, structured data and internal linking architecture. On-page SEO addresses the content itself: keyword research, heading structure, meta data, semantic relevance and readability. Off-page SEO addresses authority and trust: the quality and relevance of links pointing to the site from elsewhere on the web.
Technical SEO is the foundation everything else depends on. A site with excellent content but poor technical fundamentals will always underperform. A site with strong technical foundations and well-structured content, consistently updated and genuinely useful, will outperform most competitors over time regardless of budget. The unsexy truth about SEO is that it rewards consistency and thoroughness over cleverness and shortcuts.
Generative Engine Optimisation
Generative Engine Optimisation is the practice of structuring and writing content so that AI-powered answer engines such as Google AI Overviews, Perplexity and ChatGPT Search discover it, understand it accurately, and cite it in their generated responses. Where traditional SEO aims to rank a page in a list of results, GEO aims to be the source that AI engines draw from when constructing an answer. The distinction matters because a significant and growing proportion of search queries now return an AI-generated answer before any traditional results, and being cited in that answer is increasingly more valuable than ranking in position one below it.
GEO requires a different way of thinking about content. Rather than optimising for keyword density or link authority, it optimises for clarity, specificity and attributability. AI engines favour content that answers a single question clearly, contains specific attributed facts, uses structured data to signal its meaning, and is written by an identifiable, credible entity. These are not new principles of good writing, but applying them systematically across a site with the right technical markup beneath them is a discipline in its own right.
The good news is that good SEO and good GEO are not in conflict. A site that is technically sound, clearly structured, written with genuine authority and marked up with appropriate schema is well-positioned for both. The work overlaps significantly, and the strategic thinking that drives one drives the other. The main difference is in the emphasis: SEO prioritises relevance and authority signals, GEO prioritises entity clarity and answer-readiness.
How I approach search strategy
The good news is that good SEO and good GEO are not in conflict. A site that is technically sound, clearly structured, written with genuine authority and marked up with appropriate schema is well-positioned for both. The work overlaps significantly, and the strategic thinking that drives one drives the other. The main difference is in the emphasis: SEO prioritises relevance and authority signals, GEO prioritises entity clarity and answer-readiness.
The good news is that good SEO and good GEO are not in conflict. A site that is technically sound, clearly structured, written with genuine authority and marked up with appropriate schema is well-positioned for both. The work overlaps significantly, and the strategic thinking that drives one drives the other. The main difference is in the emphasis: SEO prioritises relevance and authority signals, GEO prioritises entity clarity and answer-readiness.
The good news is that good SEO and good GEO are not in conflict. A site that is technically sound, clearly structured, written with genuine authority and marked up with appropriate schema is well-positioned for both. The work overlaps significantly, and the strategic thinking that drives one drives the other. The main difference is in the emphasis: SEO prioritises relevance and authority signals, GEO prioritises entity clarity and answer-readiness.
The good news is that good SEO and good GEO are not in conflict. A site that is technically sound, clearly structured, written with genuine authority and marked up with appropriate schema is well-positioned for both. The work overlaps significantly, and the strategic thinking that drives one drives the other. The main difference is in the emphasis: SEO prioritises relevance and authority signals, GEO prioritises entity clarity and answer-readiness.
How I approach search strategy
The most direct way to demonstrate knowledge of SEO and GEO is to show it working in practice. This page is itself an example of the principles described above, applied to a real page with real markup. What follows is a walkthrough of the specific implementation decisions made here and the reasoning behind each one.
Entity establishment
The page head contains a Person schema entity with an @id of https://www.markexcell.co.uk/#person. This entity is referenced from every page on the site using the same @id, which tells search engines and AI engines that all pages are about the same person and builds a coherent entity graph across the site. The sameAs property links to the LinkedIn profile, corroborating the entity with an external authoritative source.
Structured data on this page
The JSON-LD in the head of this page contains five schema types working together:A WebPage entity identifies this page as a document about Mark Excell, linked to the site-level WebSite entity and the Person entity through isPartOf and about properties respectively.A BreadcrumbList with two items, Home and SEO and GEO, feeds the breadcrumb display in Google search results and helps AI engines understand where this page sits in the site hierarchy.An Article entity attributes the page content to Mark Excell as author and to the site as publisher, with datePublished and dateModified fields that signal freshness to both search engines and AI engines.A FAQPage entity with six questions and answers that directly address the queries AI engines most frequently receive about SEO and GEO. Each answer is written as a concise, attributable declarative statement rather than a discursive paragraph, because that is the format AI engines extract most reliably.A HowTo entity walking through the five steps of GEO implementation. HowTo schema is one of the formats most likely to generate rich results in Google search and is increasingly cited by AI engines when answering process-oriented questions.
Speakable specification
The Speakable spec in the WebPage entity references four CSS selectors: the opening paragraph, the SEO definition paragraph, the GEO definition paragraph and the live demonstration introduction. These are the four passages most likely to be surfaced by voice assistants and AI engines when answering questions about SEO, GEO and Mark Excell. Marking them explicitly tells AI engines these are the passages to prioritise.
Semantic HTML structure
The page uses a single H1 for the page title, H2 elements for each major section, and H3 elements for sub-sections within them. Each major section is wrapped in a <section> element with an aria-labelledby attribute pointing to its H2. Definition paragraphs use specific class names that match the Speakable CSS selectors. The FAQ section uses the Accordion utility component from FlowUI, which outputs aria-expanded attributes on question triggers, making it accessible to screen readers and correctly parseable by crawlers.
Canonical tag and Open Graph
The canonical tag <link rel="canonical" href="https://www.markexcell.co.uk/seo-geo/"> tells search engines this is the definitive version of this page, preventing any duplicate content issues if the URL is accessed through query parameters or alternative paths. The Open Graph tags control how the page appears when shared on LinkedIn or other social platforms. For a practitioner showcase these are not just social sharing niceties, they are additional entity signals that corroborate the page content with external platforms.
The llms.txt file
At the root of this site there is a plain text file at https://www.markexcell.co.uk/llms.txt that provides a structured summary of the site for AI crawlers. It describes who Mark Excell is, what disciplines the site covers, and what content AI engines are welcome to cite. It is an emerging standard rather than a universally adopted one, but it is a low-effort, future-friendly addition that signals awareness of how AI engines work and how to communicate with them directly.
Results in practice
The best evidence of SEO knowledge is sustained organic performance over time rather than a single impressive month. The results below come from real engagements, referenced anonymously where client confidentiality applies.
Sustained organic growth over four consecutive years
Working as a freelance digital consultant, a client's customer base grew by 400% year on year for four consecutive years through organic search alone, with no paid advertising investment. The strategy combined a full technical SEO audit, keyword architecture built around purchase-intent clusters, and a content programme developed and refined over time. The consistency of results across four years is the significant detail: it demonstrates a system working rather than a fortunate spike.
Global ranking improvement on a specialist publishing site
Stefanie Unland, editor of The Recycler, a specialist technology publication with a diverse international audience, saw the site rise in global rankings from approximately 1,150,000 to 229,000 following SEO consultancy work. In her words: "This is an absolutely incredible statistic to achieve with our limited budget and diverse audience needs." The challenge on a specialist publication with a constrained budget is prioritisation: identifying the technical and content interventions with the highest return and implementing them in the right order.
This site
This site is itself a live SEO and GEO implementation. The structured data, semantic HTML, entity schema, FAQPage markup, Speakable specification, llms.txt file and content architecture described on this page are all implemented here in full. The site is designed to rank for Mark Excell's name and associated disciplines, to be cited by AI engines when asked about SEO and GEO practitioners in Oxfordshire, and to demonstrate through its own construction that the practitioner behind it knows what they are doing.
Tools I work with
The following tools are part of the regular working toolkit across SEO audit, keyword research, analytics, reporting and GEO implementation.
Google Analytics 4, Google Search Console, Looker Studio, Google Tag Manager
Ahrefs, SEMrush, Google Keyword Planner, Answer The Public
Screaming Frog, Google Search Console coverage reports, PageSpeed Insights, Chrome DevTools, schema.org validator, Google Rich Results Test
JSON-LD structured data, schema.org vocabulary, FAQPage markup, Speakable specification, llms.txt, Open Graph protocol, ProfilePage schema
Common questions about SEO and GEO
These questions and their answers are also present in the structured data on this page. The FAQ schema in the JSON-LD means AI engines can extract and cite these answers directly when responding to related queries.