ENPTESTry it free
Product

A complete diagnosis of the site, the Google Business Profile and Bing, with proof for every number.

Performance and searchCore Web VitalsTechnical SEOStructured dataSemantic SEO
AI and trustAI readinessAI and search botsE-E-A-T and trustPrivacy and security
Local presenceGoogle Business ProfileBing analysis
Follow-upSearch Console and GA4Monitoring and alertsReportsGoogle rankingsKeyword researchSEO and AI goals
Who it's for

Agencies, consultants and companies that need to show what is wrong with a site and follow the fix.

SegmentsAgencies and consultanciesServices and educationE-commerceSaaS and startupsHealthcare and clinicsReal estate and local
Start hereSample reportFree diagnosisPlans and pricing
Help

Straight answers, a real report to explore and real people on the other side.

LearnFrequently asked questionsSample reportAll featuresFree toolsAboutCookie policy
Talk to usContact usWhatsAppEmailSign in to the app
Try it free
Semantic and entity SEO

Show search engines and AI what your site is about and who is behind it

Semantic SEO is the practice of organizing a site by topics and entities, not only keywords, so that search engines and AI understand what it covers and who is behind it. Path Ranker groups your pages by subject, points out the pillar page of each group and checks what your markup declares about the company, the authors and the content. It also checks whether each passage makes sense on its own, which is how search engines and AI assistants read.

Try it freeSee pricing

How it works

01

Reads what the crawl brought back

It uses the page text, the links placed in the body copy and the JSON-LD markup the analysis already read. It works without any API key.

02

Groups by subject and checks the markup

Content pages are compared by vocabulary (or by meaning, with your own OpenAI or Gemini key) and grouped. Then come the 16 entity signals and the 6 context checks.

03

Returns the score and what to do

A 0 to 100 score, the average of topical authority, entity readiness and context engineering. Each finding carries an example taken from your site and the fix.

What it checks

Subject groups

Content pages are grouped by subject. The screen shows each group, the words that define it and how many pages stand alone, with no other page on the same topic.

A well-linked pillar page

In each group of 3 or more pages, the pillar is the page the others link to most. A group is well linked when the pillar links to at least 30% of the others and at least 30% link back; menus and footers do not count.

Related pages with no link

Pairs of very similar pages that do not link to each other in either direction become internal link opportunities, with both URLs side by side.

Who the organization is

It checks the organization @id, the profiles in sameAs, a reference source (Wikidata, Wikipedia, the company LinkedIn page, the Google Business Profile link), the Wikidata item and details such as logo, address, contact, legal name and tax ID.

Authors, subjects and expertise

It checks whether article authors are declared as Person with a page or profile, whether pages say what they are about (about and mentions) and whether the organization lists its areas of expertise (knowsAbout).

What Google sees in the text

On the Pro and Enterprise plans, Google reading of the text shows how much weight the brand has on the home page and which subjects Google finds in the text that the markup does not declare.

One description, one spelling

It compares the meta description, og:description and the organization description in the markup, and counts the spellings of the brand name in titles and text ("Acme360" and "Acme 360" count as different spellings).

Passages that stand on their own

It flags generic headings ("Benefits", "Learn more"), sections that open with "this" or "also", blocks over 450 words with no heading and articles with no in-text link to a service, product or contact page.

What it does not do

  • Without your own OpenAI or Gemini key, pages are compared by their words, not by meaning: two pages on the same theme with very different vocabulary can end up in separate groups.
  • Grouping by subject needs at least 8 content pages. On smaller sites, the score comes from entity readiness and context engineering only.
  • The pillar page assessment only appears when the crawl reaches the whole site. If it stops at the page limit of your plan, the pillar may be on a page it did not read, and that part is left out of the score.
  • It does not measure links from other sites or whether the content is true, and the context checks work on words: they do not recognize synonyms.

Which plans include it

Starter, Pro and Enterprise, in the full analysis or in the "Semantics and structured data" group, with the Semantic SEO menu in the dashboard. Comparing by meaning uses your own OpenAI or Gemini key, saved under Integrations, on any paid plan. Google reading of the text runs in up to 20 analyses a month on Pro and 60 on Enterprise. The Free plan does not include this area.

See pricing

Frequently asked questions

  • No. Everything works without one, and pages are compared by vocabulary. With an OpenAI or Gemini key under Integrations, the comparison is by meaning; the cost goes to your account with the provider, and a page that has not changed is not sent again for 30 days.
  • The title, the first H1 and the start of the text (up to 2,000 characters) of each page compared, at most 150 pages. If the key fails, the analysis goes on by vocabulary and the report shows the error code.
  • The page that gathers a subject and that the other pages on the same subject point to. Path Ranker picks the most linked page of each group and shows whether it and the others link to each other in the text.
  • Structured data checks whether your markup is valid against the schema.org vocabulary. This page asks something else: whether the markup says who the company is, who writes and what each page is about, and how the pages connect by subject.
  • Partly. With fewer than 8 content pages, grouping by subject does not apply and the score comes from entity readiness and context engineering.

Other features

See how search engines and AI understand your site

Run an analysis and get the subject groups, the pillar pages and what is missing from your markup, with an example and a fix for each point.

Try it free