Learn
How AI search works, in plain words
These guides explain how AI search works in plain words, with numbers, dates and the engines named. Each opens with a direct answer, cites the engine's own documentation for every claim about how it works, carries a published date and an author, and says what cannot be guaranteed. They are the notes we work from, published.
01
Start here: what GEO is and how it differs
Two guides define the field. The first explains Generative Engine Optimization: where the term came from, how an engine builds an answer, what the work involves and how long it takes. The second sets GEO beside AEO and SEO, with a table, a decision list for which to do first, and the timelines compared.
What is Generative Engine Optimization (GEO)?
Generative Engine Optimization (GEO) is the work of becoming a source that ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews cite. Where the term comes from, how engines build answers, what the work involves and how long it takes.Published 17 October 2026 · 6-minute readGEO vs AEO vs SEO: what each one is for
SEO ranks a page in a list of links. AEO wins the direct-answer slot inside Google. GEO earns a citation inside an answer written by ChatGPT, Gemini, Perplexity or Claude. A comparison table, a decision list for which to do first, and the timelines.Published 17 October 2026 · 6-minute read
02
How the engines choose their sources
Each engine selects sources differently, so we wrote one guide per engine where the mechanics are documented. ChatGPT search retrieves through third-party providers with Bing as its default, so Bing indexing comes first. Google AI Overviews are built from Google's own rankings, so SEO comes first. Both guides separate what is documented from what we infer.
How ChatGPT chooses which sources to cite
ChatGPT search retrieves pages through third-party search providers, with Bing as its default, then picks passages that answer the question. What OpenAI documents, what it doesn't, the three crawlers, and a checklist for getting indexed and quoted.Published 17 October 2026 · 6-minute readHow to get cited in Google AI Overviews
Google AI Overviews are built from Google's normal index and rankings: Google says they only show information backed by top web results. What Google has documented, three steps that follow from it, what doesn't help, and a realistic timeline.Published 17 October 2026 · 6-minute read
03
Playbooks: what to do and what to pay
Two guides for people deciding what to do and what to pay. The SaaS playbook walks through five steps, from the prompt map to the report a sales team can use. The pricing guide lists what Bangalore agencies publish, checked on 17 October 2026, what drives the price, and how to compare two quotes line by line.
GEO for SaaS companies: a practical playbook
How a SaaS company gets named when buyers ask ChatGPT, Perplexity or Google for the best tool, a comparison or a how-to. Prompt map, retrievability for docs and pricing, pages written to be quoted, third-party listings, and what to report to sales.Published 17 October 2026 · 6-minute readWhat GEO costs in India
Published GEO prices in India start around ₹50,000 a month; most agencies quote on request. What Bangalore agencies publish, what drives the price, what a plan should include, retainer versus sprint versus audit, and how to compare two quotes.Published 17 October 2026 · 6-minute read
04
Technical: the files and markup engines read
Two guides for the people who ship the fixes. The llms.txt guide covers the specification, who reads the file today, and our own file annotated line by line. The schema guide covers the five types that matter for a service business, what Google says schema does and does not do, and why we generate every block from one data file.
llms.txt: what it is and how to write one
llms.txt is a Markdown file at /llms.txt that tells AI systems what a site is and which pages matter, proposed by Jeremy Howard in September 2024. The spec, who reads it today, Citemark's own file annotated, and six steps to write yours.Published 17 October 2026 · 5-minute readSchema markup for AI search: what actually matters
Which schema.org types help AI search and which don't: Organization as ProfessionalService, Service, Article, BreadcrumbList and FAQPage after Google's May 2026 withdrawal. Why to generate schema from one source of truth, and how to check it.Published 17 October 2026 · 6-minute read
Method
How these guides are written
Every guide follows the rules this site follows. The H1 is a question or a claim; the first paragraph answers it in 40–60 words; every H2 does the same. Claims about how an engine works link to that engine's documentation, checked on the published date. Timelines are the ones we publish to clients. Nothing is guaranteed.
- A published date and an author line on every guide. Today the author is Shoaib Mustaque.
- A sources section with the URL of every document a claim rests on.
- The updated date changes only when the content does, never on a rebuild.
- Written by the people who do the work, in the voice we use with clients.
Read enough? See what the engines say about you
The free audit runs your category's prompts on ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews and shows who each engine cites today, with the three fixes first.