Schema Forge

How it works

How Schema Forge reads a page and builds JSON-LD

The steps from URL to markup: fetching the page safely, detecting existing schema, building nine schema types and flagging every gap.

From URL to markup

Short answer

Schema Forge fetches the public HTML of a page, reads what it already says, and builds JSON-LD from those facts. Anything it cannot find is left out and listed as a task, never invented.

  1. Fetch the page

    You give it a public URL. The server downloads the HTML with a 15-second limit and a 2 MB cap, follows up to five redirects, and refuses any address that points to a private network.

  2. Read what is already there

    It parses existing JSON-LD (including @graph blocks), microdata and RDFa types, then reads titles, headings, meta tags, tel: and mailto: links, the address element, breadcrumb trails, recipe and video elements, and question and answer patterns.

    Markup the page already has is checked against the properties Google asks for.

  3. Build nine schema types

    FAQPage, Product, LocalBusiness, Article, Organization, BreadcrumbList, Event, Recipe and VideoObject, each scored by how strong the page's signals are. Types the page already has are marked "Already on page", and everything that is ready can be copied as one combined block.

  4. Flag every gap

    Each block comes with a checklist: no price found, no author, address not split, no opening hours. Each gap has a field, and the markup updates as you type.

  5. Publish and test

    Copy the script tag into the head or the end of the body, then check it with Google's Rich Results Test and the Schema Markup Validator.

What it will and will not do

It will

  • Use only facts the page states
  • Tell you what is missing
  • Check markup you already have
  • Keep the page out of storage

It will not

  • Invent prices, ratings, reviews or hours
  • Read content added by JavaScript, unless rendering is on
  • Detect accordions built from generic div elements
  • Guarantee a rich result

Questions

Does Schema Forge run JavaScript on the page?

By default no. It reads the HTML the server returns, so a page that builds its text in the browser may come back with little or nothing. If the server has JavaScript rendering enabled, a checkbox next to the address box loads the page in a headless browser first.

What happens to private or internal URLs?

They are refused. Addresses that resolve to private, loopback or reserved ranges are blocked, including after redirects.

How is the match strength decided?

Each type has its own signals. A page with three or more question and answer pairs, or with a phone number and an address, scores high. A single weak signal scores low.

Try it on a real page

It takes a few seconds and needs no account.

Open the generator