What is an entity in SEO?

An entity is a distinct, identifiable thing such as a company, person, product or place. Search engines and language models store facts about entities and the relationships between them, rather than only matching keywords.

Why does ChatGPT recommend competitors and not my client?

When ChatGPT answers from its model's knowledge, it recommends brands that appear often and consistently in what it was trained on and in what it retrieves. If competitors are discussed more on trusted sites, they are more likely to be named.

Does a Wikipedia page help AI visibility?

Wikipedia is widely used as a reference source, so a legitimate, notable entry can help. Most businesses do not meet Wikipedia's notability rules, and creating promotional entries breaks its policies. Focus on sources appropriate for the client's size and category.

How long does entity SEO take?

Fixing inconsistencies on owned profiles takes weeks. Building enough third-party presence to change how AI assistants describe a brand usually takes months, because it depends on other sites publishing and on models being updated.

Entity SEO for LLMs: How to Make AI Engines Know Your Client's Brand

Ask an AI assistant to recommend an accountant in Leeds, a CRM for a ten-person team or a running shoe for flat feet, and it will name a few brands. Some of those brands will have weaker websites than your client. They are named anyway because the assistant has a clearer, more consistent picture of who they are.

That picture is what entity SEO builds. This article explains what entities are, why they matter so much for AI assistants, and a practical plan agencies can run for any client.

What is an entity, and why does it matter for AI?

An entity is a distinct thing that can be identified and described: a company, a person, a product, a place. Search engines have organised information around entities for years. Google's Knowledge Graph stores facts such as a company's name, founders, headquarters and products, and the links between them.

Language models work in a related way. They do not store a neat database, but they learn associations between a brand and the facts and topics it is mentioned alongside. When a model has seen a brand described clearly and consistently in many places, it is more likely to mention it, and to describe it correctly.

That matters because many AI answers name brands without citing a specific page. For those answers, the quality of the client's website is only part of the story. The rest is what the wider web says.

How AI assistants decide which brands to mention

There are two paths to being named in an AI answer.

Retrieval. The engine searches, finds a page that mentions the brand, and cites it. Strong pages and third-party coverage that ranks well both help here. This is covered in how ChatGPT and Perplexity choose sources.
Model knowledge. The engine answers from what its model already knows. Here, the brand is named because it was described often and consistently in the material the model learned from.

Entity SEO strengthens both paths. Clear, consistent facts make retrieved pages easier to use, and wide, consistent coverage builds the associations a model relies on.

The entity SEO plan

1. Write the canonical facts down

Start with a single document of the facts that define the client: legal and trading name, what they do in one sentence, products and services, locations served, founding year, key people, contact details and official profile URLs. Agree the wording with the client. Every other step uses this document as the reference.

2. Make the about page the best source of those facts

The about page should state the canonical facts in plain, quotable sentences. "Nordic Roast is a specialty coffee roaster in Oslo, founded in 2016, that sells single-origin beans online across Norway and Sweden" is far more useful to an engine than a mission statement. Add the people behind the business with real credentials, and link to official profiles.

3. Mark up the organisation properly

Use Organization or LocalBusiness structured data with the name, logo, URL, contact details and sameAs links to official profiles. This does not get a page cited by AI on its own (see does schema markup get you cited by AI?), but it helps search engines connect the site to the rest of the brand's footprint.

4. Fix inconsistencies everywhere the client controls

Audit every profile the client owns: Google Business Profile, social accounts, marketplaces, app stores, directories and partner listings. Old names, outdated services, wrong addresses and conflicting descriptions all weaken the picture. Align each one with the canonical facts.

5. Build presence on the sources engines read for the category

Every category has a set of sites that AI answers cite repeatedly: review platforms, industry directories, comparison sites, trade publications and community forums. Find them by running prompts for the client's category and noting which third-party domains appear. Then earn presence there through reviews, listings, contributed expertise and coverage.

This is the highest-value step for most clients. Many brands are absent from AI recommendations not because their site is weak, but because the sites engines trust for that category never mention them.

6. Connect people to the brand

For many categories, particularly professional services, health and finance, the people behind a business carry as much weight as the business itself. Author pages with credentials, consistent bios across the web, and genuine expert contributions help engines connect expertise to the brand.

7. Keep facts current

When prices, services, locations or leadership change, update the site and the main profiles at the same time. Conflicting information across sources is a common reason assistants describe a brand wrongly.

How to measure entity strength

Entity strength is hard to measure directly, but a few signals give a reliable read.

Signal	How to check
Unprompted mentions	Ask assistants open category questions and note whether the brand is named
Description accuracy	Ask "what is [brand]?" and compare the answer with the canonical facts
Knowledge panel	Search the brand name and check whether Google shows a panel with correct facts
Third-party coverage	Count mentions on the category's most-cited third-party sites
Consistency	Audit owned profiles against the canonical facts

Run these checks at the start of an engagement and every quarter. For the ongoing prompt tracking, see AI share of voice.

Common entity problems agencies find

Two names in use. A trading name on the site and a legal name everywhere else, with no link between them.
A rebrand that never finished. The old brand still appears on directories and marketplaces.
Vague about pages. Values and mission, but no plain statement of what the business is and does.
No people. Expertise-heavy businesses with no named experts anywhere.
Absent from category sources. A strong site, but no presence on the review and comparison sites engines cite.

What this looks like for different clients

Local businesses. Consistent name, address and phone details, a complete Google Business Profile, local directory listings and genuine reviews matter most.

SaaS and e-commerce. Presence on review platforms, comparison articles, integration directories and community discussions carries the most weight.

Professional services. Named experts with credentials, contributions to industry publications and professional directory listings build the strongest signals.

A worked example: auditing one brand

Here is what an entity audit looks like in practice for a fictional client, a physiotherapy clinic group with four locations.

Canonical facts. The clinic trades as "Northside Physio" but is registered as "Northside Physiotherapy Ltd". It offers sports injury treatment, post-surgery rehabilitation and pilates. It has four clinics and nine physiotherapists.

Owned profiles. The audit finds one clinic still listed under an old name on two directories, a closed location still showing on a map listing, and inconsistent opening hours between the website and the Google Business Profiles. None of the physiotherapists have profile pages on the site.

AI check. Asked "what is Northside Physio?", one assistant describes it correctly but lists only two locations. Asked for "physiotherapists for running injuries" in the client's city, assistants name three competitors, each with strong presence on review sites and a local sports club partnership page.

Plan. Fix the directory listings and hours within two weeks. Add physiotherapist profile pages with qualifications. Rewrite the about page with the canonical facts in plain sentences. Build review volume on the two platforms that assistants cited, and pursue a partnership mention with local running clubs, which is where competitors earned their presence.

Re-check after a quarter. Descriptions should list all four clinics correctly, and the brand should start appearing in at least some category answers.

The pattern is typical. The fastest wins are usually inconsistencies the client controls. The biggest wins are usually presence on the sources engines already trust.

How Rankwyn helps

Rankwyn's AEO Citation Radar records how AI assistants describe and mention each client across a fixed prompt set, including which third-party domains they cite for the category, so you can see where entity work is needed and whether it is working.

Key takeaways

AI assistants name brands they can describe with confidence, and that confidence comes from consistent facts across many trusted sources, not from any single page.
Start every engagement by writing down the canonical facts and agreeing the wording with the client, then use that document as the reference for every profile and page.
The about page should state those facts in plain, quotable sentences, with named people and credentials where expertise matters.
The quickest wins are usually inconsistencies the client already controls: old names, closed locations, conflicting hours and outdated service lists.
The biggest wins are usually presence on the third-party sites that AI answers already cite for the client's category, which you can find by running category prompts.
Re-check descriptions and unprompted mentions every quarter, because model knowledge changes slowly.

The bottom line

AI assistants recommend brands they can describe with confidence. That confidence comes from consistent facts on the client's own site and profiles and from presence on the sources engines trust for the category. Write the facts down, make the about page the best source of them, fix every inconsistency, and build presence where the category's AI answers already look. For the full strategy, see our AEO and GEO guide.