List the important names on a page
Find likely people, organizations, products, and other named things in pasted page text so you can review them in one place.
Entity Coverage Analyzer
Free, no signup. Find likely entities in your page text, then reconcile selected names against Google’s Knowledge Graph and Wikidata. Local extraction works when the AI path is unavailable.
Runs entirely in your browser — nothing you paste is uploaded or stored. The optional AI extraction sends your pasted text to the bounded model endpoint. Reconciliation sends only a selected entity name to Google KG and Wikidata; local extraction stays in your browser. Anonymous run-level outcome counters may be used for aggregate research; URLs, domains, IPs, and identifiers are never included, and no statistic is released below 100 runs.
Sample report Deterministic local example
For “Cloudflare operates Workers AI. Cloudflare provides services to Acme Analytics.”, local extraction returns:
Cloudflare · 2 mentions · heuristic
Workers AI · 1 mention · heuristic
Acme Analytics · 1 mention · heuristic
No sameAs links appear in the JSON-LD until a user performs a lookup and accepts an identifier. The example intentionally stops before reconciliation because a real match depends on the live service and human identity review.
How to use it
- Paste the visible text or HTML from one page.
- Choose Extract entities for the local capitalized-name heuristic, or the optional model-assisted extraction.
- Review the proposed names and select Reconcile IDs only for entities you intend to check against Google KG and Wikidata.
- Inspect ambiguous results before using the generated
aboutandmentionsJSON-LD.
Site passport Local context for this saved site
Local data
Saved targets, named lists, and recent check summaries remain only in this browser.
Rate this tool
Entity JSON-LD
What you get
- heuristic — proposed by the local capitalized-name pattern.
- observed — parsed from the point-in-time model-assisted response, not independently verified.
- linked — a lookup returned a stable identifier; confirm the identity manually.
- ambiguous — more than one candidate was returned and the first must not be assumed correct.
How it works
The local path counts capitalized sequences of up to five terms, removes a small stop list, and sorts by mentions. The optional AI path asks the bounded endpoint for candidate names. Google KG and Wikidata lookups are separate and user-triggered. Generated WebPage JSON-LD includes only linked entities: the first under about, subsequent matches under mentions, and the reconciled KG, Wikidata, Wikipedia, and claimed social URLs under sameAs.
Features
- Local extraction that works without an API or model.
- Optional model-assisted candidate extraction with visible provenance.
- Per-entity, on-demand Google KG and Wikidata reconciliation.
- Richer
sameAsJSON-LD restricted to reconciled entities. - Explicit ambiguity and unavailable-result states.
Limitations
The heuristic misses lowercase brands and common-noun concepts, and can mistake headings or sentence openings for entities. Mention count is not topical importance. Model extraction can vary. The first Knowledge Graph result is not necessarily the intended entity, and the generated ordering is not a semantic judgment. This tool does not compare the page against a required topic model or competitors.
Compare existing entity schema
Paste an existing Person or Organization node to find missing name, URL, description, and identity links after reconciliation.
Frequently asked questions
What is entity coverage?
Here it means the named people, organizations, products, and concepts that appear in the supplied page text. The tool inventories likely names; it does not decide which entities"Entity & identity schema" is a practitioner label — not an official Google or schema.org category, and not exhaustive of identity-capable schema.org types — for the three main types that declare who or what is behind a site: Organization, LocalBusiness, and Person. a topic must include.
What is the difference between extraction and reconciliation?
Extraction proposes names found in text. Reconciliation connects a selected name to a stable knowledge-graph identifier. A name can be extracted correctly but reconciled to the wrong real-world entity, so every match needs review.
Does a Knowledge Graph miss mean an entity does not exist?
No. A miss can reflect ambiguity, coverage, quota, or service availability. The tool reports it as unmatched or not evaluated, never as proof of nonexistence.
Can I publish the generated JSON-LD as-is?
Review it first. Only reconciled entities are included, but the top match can still be wrong. Confirm each identity and ensure the entity is genuinely about or mentioned by the page.
Does adding more entities improve AI visibility?
Not automatically. Clear, accurate coverage can reduce ambiguity, but stuffing names or schema that the visible page does not support creates misleading markup rather than authority.
Feature requests for Entity Coverage Analyzer
Upvote what you want most. New ideas can be submitted from the floating Feedback menu; requests appear here once approved, and the most-wanted rise to the top.
You won't be emailed about that request anymore.
Loading…
➕ Request a feature
New requests are reviewed before they appear here.
Where this tool helps
Common use cases
Check whether a key name is clear and repeated
See how often each likely entity appears, while remembering that a higher count does not automatically make it more important.
Connect a name to a stable identity
Look up one selected name in Google Knowledge Graph and Wikidata, then manually confirm that the returned person, company, or product is the intended one.
Draft reviewable entity JSON-LD
Create about and mentions markup only from reconciled entities, with identity links that can be checked before anything is published.
Find gaps in existing entity schema
Compare an existing Person or Organization node with a reconciled result and hand supported missing fields to the Schema Generator.
Watch the full workflow
Entity Coverage Analyzer walkthrough
Read the transcript
Entity Coverage Analyzer
An entity is simply a specific thing with a name, such as a person, company, product, or place. This walkthrough shows how to find likely entities in a page, check one identity, and create schema you can review. We’ll use fictional store copy and repeatable example lookup data.
Step 1
Entity coverage means the named things that appear in a page. For example, an article might mention a company, one of its products, and an expert. This tool puts likely names into a list. It does not judge whether the list is complete or whether adding more names would improve rankings.
Step 2
Use it when you want to inventory the names on one page, check whether a key name is clear, connect a name to a stable identity, draft entity schema, or compare that draft with existing schema. It is a page-review tool, not a competitor research tool or a guarantee of search visibility.
Step 3
Paste the visible text or H-T-M-L from a single page. Our example is fictional store copy. It mentions Cloudflare, Summit Trail Supply, a product called Northwind Alpine Pack, a hiking club, and an editor. Keeping the example fictional makes the result safe and repeatable.
Step 4
“Extract entities” uses a simple rule in your browser. It looks for capitalized names and does not send the page to a model. “Try AI extraction” sends the text to the optional model endpoint and may vary. For a first review, the local button is the clearest and most private starting point.
Step 5
Choose “Extract entities.” The tool immediately creates a candidate list. This is extraction: it found text that looks like a name. It has not yet proved what any name refers to, and it has not contacted Google Knowledge Graph or Wikidata.
Step 6
Each row shows a proposed name, how many times it appears, and the word “heuristic.” Heuristic means a useful rule of thumb, not a verified fact. A larger mention count only means the name appears more often. It does not prove that the entity is more important or that the page is better optimized.
Step 7
The local rule can miss brands written in lowercase and ideas expressed as common nouns. It can also mistake a heading or the first words of a sentence for an entity. Treat this list as a fast review queue. Remove false positives mentally, and notice important names that the rule missed.
Step 8
Reconciliation is a separate step. It asks whether a selected name matches a stable record in Google Knowledge Graph or Wikidata. This matters because the same name can describe different people or companies. Only choose “Reconcile IDs” for a name you are prepared to review.
Step 9
We’ll reconcile Cloudflare. The captured response is fixed example data so the demo never depends on a live service. The status says a Wikidata identifier was applied. In real work, open the result and confirm that its name, description, and context match the entity discussed on your page.
Step 10
The tool now drafts WebPage J-S-O-N L-D. The first linked entity goes under “about,” and later linked entities go under “mentions.” Same-as holds the stable identity link. Unreconciled names are left out. That is intentionally cautious, but you must still confirm that the page is truly about or mentions each linked entity.
Step 11
You can copy the J-S-O-N L-D, download it as a file, or copy a share link that restores the pasted text. These are handoff tools. They do not publish anything to the website, and the shared link contains the input in the address, so review it before sending sensitive text.
Step 12
The comparison panel accepts an existing Person or Organization schema node. This example has a spaced version of the name and is missing its canonical address, description, and identity link. Choose “Find entity gaps” to compare it with the reconciled result.
Step 13
The result becomes a small fixing list: use the reconciled name, add the canonical entity address, add a concise page-supported description, and add the reconciled same-as link. “Open in Schema Generator” carries those supported fields into the next tool for review.
Step 14
The main features are private local extraction, optional model-assisted extraction, one-name-at-a-time reconciliation, cautious about and mentions markup, clear unavailable and ambiguous states, export controls, shareable input, and a handoff to the Schema Generator.
Step 15
This tool cannot tell you which entities a topic must cover, whether a first lookup result is correct, or whether adding schema will improve rankings or A-I visibility. It does not compare competitors. Use it to make names and identity evidence easier to review, then apply human editorial and technical judgment.
Use the result as a review list
Use the extracted list to ask whether the page names its important people, organizations, products, and places clearly. Reconcile only the names you need, confirm every match yourself, and publish schema only when it accurately reflects the visible page. The tool creates evidence for a review; it does not decide what your page should say.
Common issues & how to fix them
- Warning Likely entity has not been reconciled Fix: Review candidate identities and attach the correct stable Knowledge Graph ID before generating entity references.
- Warning Important expected entity is absent Fix: Add accurate coverage of the expected entity where it materially helps answer the page’s topic.
- Warning Entity candidate maps to multiple identities Fix: Disambiguate the name with type and context, then select the verified entity ID rather than the first match.
- Information Entity coverage is sparse Fix: Add missing topic-critical entities and relationships based on the page’s actual subject, not raw mention counts.