Search engines

Google describes its work in three stages. Crawlers download the pages they discover, mostly by following links from pages they already know. Indexing analyses the text, images and video of each page and stores the result in a large database, the index. Serving picks the indexed pages that best match a query and puts them in order. Searchers only see the third stage, yet a page that fails at either of the first two never reaches it. Search engine optimisation (SEO)SEOSearch engine optimisation: the work of making pages easy for search engines to crawl, index and choose as a result for relevant queries.In the glossary is the work of making a page easy to crawl, to understand and to choose. Google states that it accepts no payment to crawl a site more often or to rank it higher, and that it guarantees none of the three stages.

For many businesses, search is where a customer's need first becomes visible: someone types a service, a product or a company name and decides from a short list of results. A page missing from the index takes no part in that comparison, however good the offer behind it. The same index also feeds AI answers. Google's AI Overviews and AI Mode draw on pages that are indexed and eligible for a snippet, so classic SEO is the groundwork for AEOAEOAnswer engine optimisation: the work of making pages easy for AI assistants to find, read and cite. It builds on SEO, because assistants mostly answer from a live web search.In the glossary as well.

Most of the signals a search engine reads are set on the page itself. Each page needs a title that names its subject, since Google builds the clickable headline of a result mostly from it. The meta description is a suggestion: Google writes snippets primarily from the page content and uses the description when it sums up the page better. Headings that follow the content, words in the address that mean something to people and alt textAlt textA short text alternative for an image, written in its `alt` attribute. Screen readers read it aloud and search engines use it; purely decorative images get an empty one.In the glossary on images help crawlers and readers alike. A canonical addressCanonical URLThe preferred address of a page when several URLs show the same content. The element `rel="canonical"` tells search engines which one to index and display.In the glossary tells the search engine which of several near-identical URLs to show, hreflanghreflangAn annotation that links the language versions of a page to each other, so that search engines show each searcher the version in their language.In the glossary links each page to its translations, and a sitemapSitemapAn XML file that lists the addresses of a website that should be indexed. It helps search engines find new and changed pages sooner.In the glossary lists the pages worth indexing.

A common reason a new site stays invisible is a leftover block from the test phase, usually a noindexnoindexA rule in the HTML or in an HTTP header that keeps a page out of search results. It stops indexing, which a robots.txt rule cannot.In the glossary rule. The robots.txtrobots.txtA text file at the root of a website that tells crawlers, one by one by name, which paths they may load. It controls crawling, not whether a page appears in search results.In the glossary file controls crawling, not indexing: to keep a page out of the results, Google advises noindex or a password. Structured dataStructured dataMachine-readable facts about a page, usually schema.org vocabulary in JSON-LD. They can make a page eligible for rich results and must match what the page visibly says.In the glossary can make a page eligible for rich resultsRich resultsSearch results with extra elements such as images, ratings or prices, built from structured data. Google decides when to show them and supports only certain types; FAQ rich results ended on 7 May 2026.In the glossary, but only for the types Google supports. Since 7 May 2026, Google no longer shows FAQ rich results. Changes take time: Google notes that some show within hours and others only after several months, so results are best judged over weeks in Search Console before titles are rewritten again.