AI tools are making decisions about who you are. When someone asks ChatGPT or Claude about a service in your sector, those tools pull together whatever information they have about your organisation and generate a response. The quality of that response depends largely on the quality of information available to them.
This is where entity mapping comes in.
In the world of AI and search, an “entity” is a named thing: a person, place, organisation, or concept. AI tools build their understanding of entities from a range of sources, including your website, structured data, knowledge bases like Wikidata, and content from across the web.
Entity mapping is the practice of creating a structured context document that defines your organisation as an entity. It sets out who you are, what you do, who you serve, where you operate, and how you are distinct from other organisations that may share a similar name or operate in a similar space.
The goal is to give AI tools something clear and authoritative to draw on, rather than leaving them to piece together an accurate picture from whatever they can find.
AI search tools do not always get it right. They can confuse organisations with similar names, surface outdated information, or simply have very little to work with if an organisation has a limited online footprint. For not-for-profits, this is a real issue. The sector is full of organisations doing important work that is not always updated or well-documented in the places AI tools look.
If someone asks an AI tool about a service your organisation provides and the response is vague, inaccurate, or refers to a completely different organisation, that is a missed opportunity and potentially a source of confusion for the people you are trying to reach.
An entity mapping document typically includes a structured summary of your organisation’s key attributes: your legal name and any trading names, your purpose, your geographic area of operation, your primary services or programs, and the relationships between your organisation and others in your sector.
It sits alongside other AI readiness work such as schema markup on your website, an llms.txt file, and ensuring your organisation is correctly represented in knowledge bases. Together, these give AI tools a more complete and accurate picture of who you are.
The easy way is to outsource the work and you can contact us if you’d like us to take care of it for you.
Otherwise the EntityMap site at entitymap.org includes a practical implementation guide, and there are two paths depending on how hands-on you want to be. An AI-assisted prompt approach that requires no account, and a free generator at waikay.io/entitymap that crawls your site and builds the files automatically.
What you end up with is two files: an entitymap.json and an entitymap.html, published at the root of your website. You then add a line to your robots.txt, a link tag to your site header, and a link in your footer so AI crawlers can find it. The entitymap.org site provides a pre-publish checklist to make sure everything is in order before you go live.
It is not a five-minute job, but it is not out of reach for a small team either, particularly with a person on your team that knows how to update your website properly.
The practice of entity mapping has been evolving across the SEO and AI search community for some time, but it is now moving toward something more formal.
A new open standard called EntityMap was published for public consultation on June 1, 2026, with a formal launch scheduled for July 1, 2026. It is published at entitymap.org under a CC BY 4.0 licence with no vendor lock-in and no proprietary software requirement. The standard has been endorsed by R.V. Guha, one of the founders of schema.org, which is the structured data framework that underpins how search engines currently read and interpret web content.
EntityMap is not a replacement for existing standards like schema.org or sitemap.xml. It fills a gap they were not designed to address. Where schema.org describes what is on an individual page, EntityMap describes what an organisation is across its entire web presence: what it does, who it serves, how its key areas of work relate to each other, and where the evidence for each claim lives. That structured picture is what AI retrieval systems need to represent an organisation accurately rather than piecing something together from fragments.
The standard is brand new and adoption is in its early stages. What it signals, though, is that the sector is taking the problem seriously and moving toward agreed frameworks.
If your organisation has a common name, operates across a complex service landscape, or simply wants to make sure AI tools represent you accurately as this technology becomes more central to how people find services, entity mapping is something worth understanding now, not later.
We are keeping a close eye on how this space develops. If you want to talk through what it might mean for your organisation, get in touch.
