Tool · Google Knowledge Graph

Are you in Google’s Knowledge Graph?

“Google knows me” is not enough. What matters is whether it knows you as the same entity that Wikidata describes.

How do you check if you are in Google’s Knowledge Graph? Type your name, your brand or a kgmid from a knowledge panel. You will see what the Google Knowledge Graph Search API returns and whether Google, Wikidata and Wikipedia describe the same entity.

The tool at a glance

10 entities

from Google’s Knowledge Graph per check, with type, description, kgmid and a link to the knowledge panel.

3 sources

Google, Wikidata and Wikipedia. I compare identifiers, not just names.

0 guessing

a Wikidata item is linked to a Google result only by identifier. Similar names are shown separately.

24 hours

is how long the server remembers a Google answer, so the same question is not asked over and over.

How it works

What the tool checks in Google and Wikidata

First it asks the Google Knowledge Graph Search API, the official API for finding entities in the Knowledge Graph. For each entity Google returns an identifier (kgmid, e.g. /g/11md2bpb8_), schema.org types, a short description, a score and sometimes an excerpt from Wikipedia.

Then it looks for that kgmid in Wikidata, in P2671 (Google Knowledge Graph ID) and P646 (Freebase ID). If Google gives a Wikipedia article, the tool also checks which Wikidata item that article belongs to and which Google identifier that item has.

Finally it compares both leads, which gives one of three statuses. What the Knowledge Graph is and where Google gets its data, I explain in my post on the Knowledge Graph and Knowledge Panel.

Result

Three statuses and what they mean

Confirmed by identifier
A Wikidata item has the same kgmid. Google and Wikidata describe the same entity.
Mine: /g/11md2bpb8_ = Q133444548
Mismatch
The Wikipedia article given by Google belongs to an item with a different kgmid. These are two different entities.
“Kraków Główny”: bus stop /g/11x_97jms7, station /m/02rn23x
Not linked
No Wikidata item has this kgmid. That is common for smaller brands.
If it is your entity, add the kgmid to its item.

Similar names from Wikidata appear only on request, in a separate box marked “to check”. The tool never links entities by name, because with common names that easily attaches someone else’s item.

Important

An empty result does not mean Google does not know you

The API returns only part of the Knowledge Graph. A knowledge panel can show up in search even if the API does not return it, and an entity in the API does not guarantee a panel. Google itself says the API is not suitable for production-critical use.

So with an empty result the tool says it carefully: Google did not return this entity in the API. A check by identifier is more precise than by name. You can find the kgmid in a knowledge panel: click “Share”, open the copied g.co/kgs/… link and look for the kgmid= parameter in the address bar. Paste the whole address into the tool.

Privacy

What happens to the name you type

The podrez.pl server sends the name or kgmid to Google, so Google does not see your IP address. The server keeps Google’s answer for 24 hours under a hash of the query, without your IP. I keep no list of the names people type.

Requests to Wikidata and Wikipedia go straight from your browser. Wikimedia sees your IP address and the identifiers: kgmid, QID and article title. It gets the name you typed only if you click “Show similar names in Wikidata”.

A link to a result keeps the name after the # sign, the part of the address that browsers do not send to the server. When you open such a link, the tool removes it from the address bar.

What I measure. Google Analytics only learns that someone used the tool, the type of result (e.g. “confirmed” or “not linked”) and which button was clicked, e.g. “Open the knowledge panel in Google”. No name, no identifier, and only in line with your cookie settings.

Before you fix anything

What to do with the result

Confirmed: check whether Google’s short description fits who you are. It is Google’s own text. You can only influence it indirectly, with consistent descriptions in the sources Google uses.
Mismatch: work out which entity is yours. Fix sameAs on your site and the identifier in the Wikidata item, but do not force someone else’s kgmid onto your record.
Not linked: if it is definitely your entity, add the kgmid to its Wikidata item: /g/… as P2671, /m/… as P646. How to edit an item, I show in my Wikidata step-by-step guide.
Empty result: start with the entity on your own site (schema with sameAs) and a Wikidata item.

Tools and lists

More tools and lists

Free, no sign-up. Each one answers one specific question.

Tool
Entity Observatory
See a page’s structured data as a relation map and add the Wikidata record next to it.
Tool
Is your site in Common Crawl?
See how many of your pages made it into the Common Crawl archive and whether anything blocks CCBot.
Tool
AI bots in logs
Drop in a server log and see which AI bots actually visit and whether your server turns any of them away.
Search
List of bots and user agents
A searchable list of bots: who runs them, what they do and how to handle them in robots.txt.
Table
AI knowledge cutoff dates
How far the knowledge of ChatGPT, Claude, Gemini and Grok reaches. Cutoff dates in one place.

The Knowledge Graph is one thing, bot access to your site is another. Check it in High Five, AI! by Maciej Chmurkowski.

On the Knowledge Graph, knowledge panels and Wikidata

Post
Knowledge Graph and Knowledge Panel
How Google decides who you are.
Post
Wikidata step by step
How to get into the database that Google and language models rely on.
Service
AI search visibility
From evidence to action: how I work on brand visibility in AI answers.

What next

Google doesn’t know you, or knows you wrong?

A mismatched entity, no Wikidata item, a description that doesn’t fit you, or two entities with the same name. This is the kind of thing I sort out in entity SEO work.

Want me to take a look?

Send a short brief: your name or brand and what the tool showed. I’ll tell you what is worth fixing first.

Send a short brief AI search visibility

Frequently asked questions

How do I check if I am in Google’s Knowledge Graph?

Type your name or brand in the field at the top of the page. The tool asks the Google Knowledge Graph Search API and shows up to 10 entities with their kgmid. If yours is among them, Google has you in its Knowledge Graph. A check by the kgmid from your knowledge panel is more precise.

What is a kgmid?

It is an entity identifier in Google’s Knowledge Graph. It looks like /m/… (older entities, from the Freebase era) or /g/… (newer ones). You can see it in a Google search address, in the kgmid parameter, e.g. after opening the “Share” link of a knowledge panel.

Does an empty result mean I have no knowledge panel?

No. The API returns only part of the Knowledge Graph. A knowledge panel can appear even if the API does not return the entity, and an entity in the API does not guarantee a panel.

How do I find my knowledge panel?

Type your name in the field at the top of the page. Next to the best-matching entity, click “Open the knowledge panel in Google”. The link searches by kgmid, so Google shows that entity’s panel if it has one. You can also look for it yourself: search Google for your name or brand with a qualifier such as your job or city. When a panel appears, click “Share” and open the copied link. The address bar shows the kgmid, which you can paste into the tool.

What does the score mean?

It is the API’s resultScore, a measure of how well an entity matches the query. It is comparable only within a single search, so the tool shows it as a percentage of the top result.

Why doesn’t the tool match Wikidata by name?

Because names repeat. For smaller brands, guessing by name easily attaches someone else’s item. The tool links results by identifier only and shows similar names separately, marked “to check”.

Where does a mismatch come from?

Google sometimes creates a separate /g/… entity instead of using an existing one, e.g. for a bus stop with the same name as a railway station, and takes the description from an article about something else. A mismatch can also come from a wrong identifier in the Wikidata item.

How often can I check?

10 times a minute and 300 times a day from one IP address. The tool also has an overall limit of Google queries. When it runs out, the tool tells you when to try again.

Do you store the names I check?

I keep no list of names. The server keeps Google’s answer for 24 hours under a hash of the query and counts queries per hour, without their content. Google Analytics only gets the type of result and which button was clicked.

Made by Ewelina Podrez-Siama, in SEO since 2009. Latest version of the tool: 1.0.4, 3 November 2026.

Data: Google Knowledge Graph Search API, Wikidata (CC0) and Wikipedia (CC BY-SA). The tool is not affiliated with Google or the Wikimedia Foundation.