Making your business clear to machines, not just to people.

An entity, in search terms, is a distinct thing such as an organisation, person or place that can be told apart from others with similar names. For a business, entity understanding is whether a system can work out who you are, what you do and how you relate to your topic. Google's Organization structured data documentation addresses part of this: adding the markup to your home page can help Google better understand your organisation's administrative details and disambiguate it in search results, and some properties can influence which logo appears in results and in your knowledge panel. It also advises using the same name and alternate name that you use for your site name. The page lists properties such as alternate name, legal name and contact point, and says there are no required properties: you add the recommended ones that apply to your organisation.

What is not documented is how ChatGPT, Perplexity or Gemini build their picture of a brand, or whether they read this markup. Google's AI features guide does not present entity work as a special AI tactic either; it points back to the ordinary foundations. So entity work is best seen as removing ambiguity: the same name, description and facts, stated clearly and in the same way wherever your brand appears.

Cultured Digital covers entity and brand understanding in this part of AI & Search Visibility. Entity understanding is also part of our structured data and entities work in technical SEO, and topic and entity strategy is covered under Shape the site.

Structured data is a standardised format for giving explicit clues about the meaning of a page. Google says it uses structured data to understand a page and to gather information about the web, such as the people, books or companies in the markup. Two rules matter for consistency. Google's general structured data guidelines say not to mark up content that is not visible to readers, and list markup that does not represent the main content of the page as a reason a feature may not appear. Its AI features guide says to make sure structured data matches the visible text on the page and to check that Merchant Center and Business Profile information is up to date.

There are also limits on what to expect. Google says there is no special schema.org structured data that you need to add to appear in AI features, and that structured data enables a feature to be present but does not guarantee it. As far as we found, the other vendors do not document how they use markup. Consistent facts are about clarity: if details such as an address or a description differ between your site, your markup and your profiles, a system has to choose between them. Google also says not to create blank or empty pages just to hold structured data, and lists errors the Rich Results Test cannot catch among the reasons a feature may not appear, so validation is necessary but not proof.

Cultured Digital covers structured data and consistent facts as part of machine understanding; schema design and validation is described under structured data and entities. Our own home page shows illustrative markup, labelled as illustrative, for a ProfessionalService with a name, founding date and area served.

Machine reading starts with access. Google's AI features guide lists the basics: crawling allowed in robots.txt and by any CDN or hosting infrastructure, content findable through internal links, and important content available as text. It also says you do not need new machine-readable files, AI text files or markup to appear. Google's link guidance says it can generally only crawl links that are a elements with an href attribute.

JavaScript adds a further question. Google documents that it processes JavaScript web apps in crawling, rendering and indexing phases (JavaScript SEO basics). We found no equivalent documentation from OpenAI or Perplexity on whether their crawlers run JavaScript, so we treat content that exists only after scripts run as a risk to check, not a known fact. Both vendors do publish crawler details: OpenAI recommends allowing OAI-SearchBot in robots.txt and its published IP ranges, and Perplexity says you may need to allow its bots through a web application firewall.

Shortcuts are a risk. Google's spam policies define cloaking as presenting different content to users and search engines to manipulate rankings, so serving machines a different version of a page from the one people see needs care. The AI features guide also lists supporting text with high-quality images and videos where applicable, and a great page experience for users. None of this is specific to AI; it is the same list as for Google Search overall, which is the point.

Cultured Digital covers sites optimised for machine reading: technical access, rendering and readable structure. See Rendering, and SEO-first websites for builds.