Finding entities
To find the entities that matter for a topic, combine six methods: look up candidates in Google’s Knowledge Graph Search API and in Wikidata, run your and competitors’ text through Google Cloud’s Natural Language API, inspect search results and knowledge panels, read competitors’ structured data, and list the related entities that keep appearing together. Then record each entity with its identifier in an entity map.
Which methods should you use?
| Method | Best for | Cost | Guide |
|---|---|---|---|
| Knowledge Graph Search API | Does Google have this entity? What ID and type? | API key | Guide to With the Knowledge Graph API |
| Wikidata | IDs, properties, official links, relationships | Free | Guide to With Wikidata |
| Natural Language API | Which entities a text mentions and how central they are | Paid beyond free tier | Guide to With Google’s Natural Language API |
| SERP and knowledge panels | How Google presents an entity to searchers | Free | Guide to In SERPs and knowledge panels |
| Competitor schema | Which entities competitors declare in JSON-LD | Free | Guide to In competitors’ schema |
| Related entities | What a complete page on a topic must mention | Free | Guide to Related entities and co-occurrence |
What should you look for?
- Your own entities: organization, people, products, locations. Does each exist in Google’s Knowledge Graph and Wikidata, and are the facts right?
- Topic entities: the concepts, standards, tools and places a topic involves.
- Ambiguous names: other entities with the same or similar names (see disambiguation).
- Attributes: the properties engines associate with each entity type, such as founder, headquarters or service area.
What is a practical workflow?
- List seed topics and your own entities.
- Look each up in the Knowledge Graph Search API and Wikidata; record IDs.
- Search Google for each and note panels, “People also ask” questions and related searches (SERP inspection).
- Run the top-ranking pages and your page through the Natural Language API; compare entity lists.
- Extract competitors’ JSON-LD (competitor schema).
- Group everything in an entity map and assign entities to pages.
This workflow is practitioner practice; no search engine publishes an official entity-research method. Practitioner practice
What are the pitfalls?
- Cloud Natural Language output is not Google Search’s view of a page; it is a separate product.[1]
- Knowledge Graph API scores are not ranking metrics.[2]
- Copying competitors’ entity lists produces sameness; add entities from your own experience.
Finding entities with the Natural Language API
Google Cloud’s Natural Language API extracts entities from text. Send a document to the analyzeEntities method and you get each entity’s name, type (PERSON, ORGANIZATION, LOCATION and others), mentions and, in the v1 API, a salience score from 0 to 1 showing how central it is, plus a Knowledge Graph MID and Wikipedia URL when one is known. It is a useful proxy, not Google Search’s own analysis.
What does analyzeEntities return?
Per Google Cloud’s documentation:[1][3][4] Officially documented
- name and type, such as PERSON, ORGANIZATION, LOCATION, EVENT, WORK_OF_ART, CONSUMER_GOOD, or OTHER.
- metadata, which can include
mid(a Knowledge Graph machine ID such as/m/012z8_) andwikipedia_url. - salience (v1): a score in the 0 to 1.0 range giving “the importance or centrality of that entity to the entire document text.”
- mentions: each place the entity appears, marked PROPER or COMMON.
The v2 reference describes a probability on mentions rather than salience, so use the v1 endpoint if you need salience.[4]
How do you run it?
- Enable the Cloud Natural Language API in a Google Cloud project and set up credentials.[1]
- Send the page’s main text (not the navigation) to the endpoint:
POST https://language.googleapis.com/v1/documents:analyzeEntities
{"document": {"type": "PLAIN_TEXT", "content": "Local Blitz is a marketing agency in Encinitas, California..."},
"encodingType": "UTF8"}
- Sort entities by salience and note which have a
mid. - Repeat for the top-ranking pages on the same query and compare.
How do you read the results?
| Observation | Possible action |
|---|---|
| Your main entity is not the most salient | Name and define it earlier; make it the subject of sentences |
Entity has no mid | The API could not link it to a known entity; add context such as location or type |
| Competitors mention entities you do not | Check whether they belong in a complete answer (related entities) |
| Wrong entity linked (same name) | Add disambiguating attributes in the text |
These actions are practitioner interpretations. Practitioner practice Chasing a salience number is not a goal in itself; see entity salience.
What are the limits?
- It is a Google Cloud product; Google has not said Search uses the same model.
- Usage beyond the free allowance is billed; check current pricing.
- Results vary with the text you send, so strip boilerplate first.
Finding entities in SERPs and knowledge panels
Search results show how Google interprets an entity. Search the name (and variations), then record whether a knowledge panel appears and what it says, which related entities and questions are shown, what sources Google cites, and whether AI Overviews describe the entity correctly. Repeat in a private window and from relevant locations.
What should you record?
| SERP feature | What it tells you |
|---|---|
| Knowledge panel | Google has an entity; check type, description, facts, image and source links[5] |
| Business Profile | Local entity details: name, category, address, reviews[6] |
| “People also ask” | Questions searchers link to the entity; candidates for H2s and FAQs |
| Related searches | Associated entities and attributes |
| Carousels and lists | Entities Google groups together (for example people in a company) |
| Top organic results | Which sources Google trusts for the entity: profiles, directories, press |
| AI Overviews | How AI answers summarise the entity and which pages they cite[7] |
How do you run a SERP entity check?
- Search the exact name, then the name plus a qualifier (city, industry, role).
- Use a private window, since results can be personalised by history and location.[8]
- Screenshot the panel and note every fact and its source link.
- List the domains ranking for your brand name; these are the sources engines associate with you.
- Note any same-named entities appearing instead of yours.
- Repeat monthly and log changes (measuring entity SEO).
This routine is practitioner practice. Practitioner practice
How do you read a knowledge panel for entity clues?
- Subtitle (such as “Marketing company”): the type Google assigns.
- Description source: often Wikipedia or the official site; tells you which source is trusted.
- Profiles row: social accounts Google has linked; compare with your sameAs list.
- “People also search for”: entities Google considers related.
Can you check AI answers the same way?
Yes. Ask Google AI Mode, ChatGPT search and others who or what the entity is, and note errors and cited sources. See entities in AI search.
Finding entities in competitors’ schema
Competitors’ structured data shows which entities they explicitly declare: their Organization or LocalBusiness details, authors, sameAs profiles, services, and about or mentions properties. Extract it with Google’s Rich Results Test, the Schema Markup Validator or your browser’s view-source, then compare types, properties and entity links with your own markup.
How do you extract competitor schema?
- Schema Markup Validator (validator.schema.org): paste a URL to see all schema.org data, including types Google does not use for rich results.[9]
- Rich Results Test: shows which Google rich result types are detected and any errors.[10][11]
- View source: search for
application/ld+jsonto read the raw JSON-LD.
All three are Officially documented tools. JSON-LD added by JavaScript may only appear in the rendered view.[12]
What should you record?
| Look at | Why |
|---|---|
@type | How they classify themselves (for example a specific LocalBusiness subtype) |
@id | Whether they build a connected entity graph |
sameAs | Which profiles and databases they link as identity references |
author, founder, employee | Which people they connect to the brand |
about, mentions | Topic entities they declare, sometimes with Wikidata or Wikipedia links |
areaServed, knowsAbout | Service areas and claimed expertise |
How do you use what you find?
- Choose the most specific type that is true for you (see LocalBusiness).
- Spot profile types you lack, such as industry directories or Wikidata.
- Find topic entities worth covering in content (related entities).
Only mark up what is true and visible on your own pages; Google’s guidelines prohibit misleading or irrelevant markup.[8] Officially documented Competitors’ markup can be wrong, so check claims against Google’s documentation.[13]
A worked example from Google’s documentation
Google Cloud’s entity-analysis guide describes the method as inspecting text “for known entities (proper nouns such as public figures, landmarks, etc.)” and returning information about them.[3] Its sample sentence is: “President Trump will speak from the White House, located at 1600 Pennsylvania Ave NW, Washington, DC, on October 7.” In the guide’s sample output:[3]
- the person is returned with type
PERSON, ametadataobject holding a Knowledge Graphmidand awikipedia_url, asaliencevalue (about 0.79 in the sample) and its mentions, each markedPROPERorCOMMON; - the date is returned as type
DATEwithmonthanddaymetadata; - the street address is returned as type
ADDRESSwith parsed metadata:street_number,street_name,locality,broad_region,narrow_regionandcountry.
The sample also shows that linking is uneven: the street “Pennsylvania Ave NW” (type LOCATION) comes back with a mid but no wikipedia_url, while the White House and Washington, D.C. have both.[3] For your own audits (inference): a mid or wikipedia_url means the system linked the name to something it knows; its absence means the name was typed but not linked. Practitioner practice Google says each document needs its own request, so audit page by page.[3]
Frequently asked questions
What is the easiest free way to find entities?
Search Google and Wikidata for the topic, note the knowledge panel, related searches and Wikidata properties, and read the top pages' structured data.
Do I need paid tools?
No. The Knowledge Graph Search API needs an API key and the Natural Language API has usage costs beyond its free allowance, but Wikidata, SERP inspection and schema extraction are free.
Does Google Search use the Natural Language API?
Google has not said so. Treat the API as an approximation of how a machine reads your text, not as Google Search's view.
What is a good salience score?
There is no target. Salience is relative within one document; what matters is that the main entity is clearly the subject.
Why is there no salience in my results?
The v2 API reference replaces salience with mention probability. Use the v1 endpoint if you need salience.
Why do different people see different panels?
Google tailors results using factors such as location, language and search history, so results vary.
What if a competitor's panel shows for my brand name?
That signals a disambiguation problem. Strengthen your entity home, use consistent naming and add distinguishing facts; see entity disambiguation.
Is it OK to copy a competitor's schema?
Copy the structure, never the facts. Your markup must describe your own visible content accurately.
Why does the Rich Results Test show less than the validator?
The Rich Results Test only reports types Google supports for rich results; the Schema Markup Validator shows all schema.org data.
See also
References
Pages accessed September 29, 2026 unless a date is given. See all sources and our editorial policy.
- ^ "Natural Language API basics". Google Cloud.
- ^ "Google Knowledge Graph Search API". Google for Developers.
- ^ "Analyzing entities". Google Cloud.
- ^ "Entity (REST reference, v1)". Google Cloud.
- ^ "About knowledge panels". Google Knowledge Panel Help.
- ^ "Tips to improve your local ranking on Google". Google Business Profile Help.
- ^ "AI features and your website". Google Search Central. Last updated December 10, 2025.
- ^ "General structured data guidelines". Google Search Central.
- ^ "Schema Markup Validator". Schema.org.
- ^ "Rich Results Test". Google Search.
- ^ "Rich Results Test (help)". Search Console Help.
- ^ "Generate structured data with JavaScript". Google Search Central.
- ^ "Introduction to structured data markup in Google Search". Google Search Central.