Entity clarity and schema

Why models get entities wrong

A language model builds its picture of a company from many fragments — your site, directories, news, social profiles, and other people's pages. When those fragments disagree, or when another organisation shares your name, the model resolves the ambiguity by guessing, and it guesses in favour of whichever entity has more corroborating text.

This is why a new company with a generic or shared name can be invisible no matter how good its content is. The content is being filed under someone else.

The four fixes, in order

First, settle on one description sentence and use it verbatim everywhere. Not a variation per channel — the same words.

Second, publish Organization schema on your home page with name, url, description, logo and a sameAs array linking every verified profile you control. The sameAs array is what lets a model tie the fragments together with confidence.

Third, add Person schema for the people who write your content, with their own sameAs links. Author identity is a meaningful signal for both search and generative retrieval.

Fourth, if a name collision exists, say so in plain text on your about page. An explicit disambiguation sentence is read and reused.

Schema that is worth the effort

Organization and Person, always. Article on content pages, with author linked to your Person entity. BreadcrumbList where your structure is deep. FAQPage where you have genuine question-answer pairs. Product or SoftwareApplication if you sell one.

Schema that is rarely worth it: markup added to a page with no substantive content, and markup that describes something the page does not actually contain. Both are ignored at best.

Checking it landed

Validate the markup, then do the real test: ask several models to describe your company in one sentence, on separate days. Consistent and correct answers mean the entity is settled. Answers that vary in industry, location or customer mean it is not, and more content will not help until it is.

What we measured on real domains

Everything below comes from ThoughtForge runs against live domains, not from a study we read. Each run scores three things out of 100: identity (can an engine say who you are without contradicting itself), durability (does your brand survive a multi-turn conversation), and retrieval (can your pages be chunked into something quotable).

Across every domain we have scanned, one number is consistently the worst, and it is not the one people optimise for. Durability sits at or near zero almost everywhere. Brands that get named in turn one disappear by turn three, because nothing on the site gives the model a reason to keep holding onto them.

• thoughtforge.app, July: identity 16, durability 16, retrieval 49, overall 27.
• thoughtforge.app, September, after entity and schema work: identity 52, durability 0, retrieval 48, overall 33.
• A services domain (megap•••••••••••.com.au): identity 28, durability 0, retrieval 47, overall 25.
• An industry association (propertyasso•••••••.com.au): identity 4, durability 0, retrieval 50, overall 18.
• A consultancy (obse•••••••••.com): identity 40 in July, 52 in September; overall 30 to 35.

What actually moved the numbers

Identity is the score that responds fastest. On thoughtforge.app it went from 16 to 52 in about six weeks, and the work was unglamorous: one Organization schema block that agreed with itself, one description of the product used in the same words on the site, in the schema and on every third-party listing, and crawler access opened to GPTBot, ClaudeBot and PerplexityBot.

Retrieval barely moved — 49 to 48 — because retrieval is about page structure, not metadata, and we had not restructured the pages yet. That is the honest shape of this work: metadata fixes buy you identity, and only rewriting the pages buys you retrieval.

Durability moved backwards, from 16 to 0. Multi-turn persistence is the hardest of the three and the one almost nobody is working on. If your brand only needs to be named once, ignore it. If you are in a considered purchase where the buyer asks the assistant four or five follow-ups, it is the whole game.

The consistency audit

List every place your organisation is described: homepage, about page, Organization schema, LinkedIn, Crunchbase, directories, app stores, press mentions. Put the descriptions side by side. Contradictions between them are the single biggest driver of a low identity score we have measured — one association we scanned resolved to different founding dates and addresses depending on the source, and scored 4 out of 100 on identity.

Fix by choosing one canonical description and propagating it, rather than by writing a better new one and leaving the old ones in place.

Schema that earns its keep

Organization with name, url, description, logo, sameAs and a real address does more than any other block. WebSite adds the search relationship. Article and BreadcrumbList belong on content pages.

Validate it, then read it aloud. If the schema says something the page does not, you have created a new contradiction rather than resolving one.

Frequently asked questions

Does schema directly improve AI citations?

Indirectly. It makes a model more confident about facts it can already read on the page. It does not create facts, and it does not rescue thin content.

How many sameAs links should I include?

Every profile you actually control and keep current — typically LinkedIn, X, Crunchbase, GitHub and any industry directory. Dead or unmaintained profiles do more harm than good.

What if a bigger company shares my name?

Disambiguate explicitly in text, be relentlessly consistent in your own description, and build corroboration on third-party pages. It is slow, but it is the only thing that works.

Related guides

• Why AI doesn't mention your brand — Six concrete reasons ChatGPT and Perplexity skip your brand, how to tell which one applies to you, and the fix for each.
• FAQ schema in 2026 — Google retired most FAQ rich results, but FAQPage schema still helps answer engines. When to use it, how to write it, and when to skip it.
• AI search visibility — AI search visibility is how often answer engines mention and cite your brand. How to measure it properly, what moves it, and what to ignore.