VALIDATE WITH OUR FREE SCHEMA MARKUP VALIDATION TOOL

Free Schema Markup Generator

Search engines and AI models do not read your page the way a person does. Structured data is how you tell them, in plain terms, what the page is.

Schema markup generator

Build valid JSON-LD structured data for 15 schema types, check it against what Google actually asks for, then paste it into your site. Runs entirely in your browser — nothing is uploaded, stored or counted.

Pick a schema type

15 types
schema.jsonld 
{}

Open Google’s Rich Results Test →

Result preview

An approximation, not a promise. Google decides page by page whether to draw a rich result at all.

Import markup you already have

Paste an existing JSON-LD block — or the whole page source, and we’ll dig the ld+json out of it. The form switches to the type it finds and fills in what it can.

How it works

The generator assembles JSON-LD structured data for fifteen schema types, covering content pages, shop pages, business listings and the handful of site-wide types that sit underneath everything else. JSON-LD is the format Google has recommended for years: it lives in its own script tag rather than being threaded through your visible markup, so you can redesign a page without snapping your structured data in half.

01

Say what the page is

Fifteen types, sorted into content, commerce, business and everything else. Choose the one that describes the page you have, not the one you wish you had.

02

Fill in what you have

Required fields sit at the top, recommended below them, optional last. Every field wants information that is genuinely on the page. Invented data is worse than a blank.

03

Watch the readiness strip

It tallies filled fields against the spec as you type, and each gap is a link that jumps you to the input responsible. No hunting through a long form.

04

Paste it into the page

Copy the block with the script wrapper switched on, drop it into the page, then put the live URL through Google’s Rich Results Test to confirm it reads.

Nothing leaves your browser

The whole generator is HTML and JavaScript running on this page. There is no account, no crawl allowance, no daily cap and no server round trip, so nothing you type is transmitted or stored anywhere. Close the tab and it is gone.

Bring markup you already have

Paste an existing JSON-LD block, or an entire page source, into the import panel. The form switches to the type it finds and fills in everything it recognises. It is the quickest way to see what your theme or SEO plugin is actually emitting.

Where FAQ and HowTo stand

Google retired the FAQ accordion for ordinary sites in 2023 and dropped HowTo results entirely. FAQ is still here, with a warning attached, because assistants and AI Overviews continue to read it. HowTo is not, because there is nothing left for it to earn.

What each editor section means

The editor has a few moving parts, and each one is doing a specific job. Here is what you are looking at.

Schema type picker

The first and most consequential choice. Content covers Article, Course, FAQ, Recipe and Video. Commerce covers Product, Review and Software. Business covers Job Posting, Local Business, Organization and Person. Everything else sits under Other: Event, Breadcrumbs and Website. Switching type clears the form, so pick before you type.

Sub-type selector

Two types can be narrowed. An Article can be declared as a NewsArticle, a BlogPosting or a TechArticle. A Local Business can be a Restaurant, a Hotel, a Dentist, a legal or financial practice, and so on. The fields stay the same; the label you hand to the search engine gets more precise.

Required fields, red edge

The minimum Google asks for before a page is even considered for the rich result attached to that type. Miss one and the Rich Results Test reports the item as ineligible rather than imperfect. These always sit at the top of the form.

Recommended fields, gold edge

Optional by the letter of the spec, and the reason one listing carries a photograph, a price and a star rating while another is a plain blue link. If you have the data, put it in. This is where most of the visible payoff lives.

Optional fields

Accepted by Schema.org, neither required nor requested by Google. Worth adding when the data is real and adds something a reader would care about: a genuine review count, a product identifier, a social profile that unambiguously belongs to you. Padding this section achieves nothing.

Readiness strip

The running tally above the form: how many required and recommended fields are filled, out of how many exist for the type you chose. Anything outstanding is listed underneath as a link, and clicking it scrolls to the field and focuses it.

JSON-LD panel

The finished markup, rewritten on every keystroke. Leave the script wrapper switched on and you get a block you can paste straight into a page. Switch it off and you get bare JSON, which is what most plugin and CMS fields want.

Result preview

A sketch of the listing your markup is reaching for, drawn differently per type: stars and a price for products, a date block for events, tags for jobs, a thumbnail for recipes and video. Treat it as a sanity check on your own data, not a rendering of what Google will do.

Import panel

The reverse trip. Drop in a JSON-LD block or a whole page source and the tool finds the structured data, works out the type, and loads what it can back into the form. Useful for auditing an existing page, or for moving a page from one schema type to another without retyping everything.

Frequently asked questions

Schema markup is a small block of machine-readable data that states, unambiguously, what a page is: an article, a product, a job opening, a business with an address and a phone number. Search engines and AI models use it to decide how to present you. On its own it is not a ranking signal, and anyone telling you otherwise is selling something. What it does is make you eligible for the treatments that change behaviour: star ratings, prices, event dates, image carousels, citations inside AI answers. The ranking stays where it was; the click-through rate often does not.

Anywhere in the document, though the head is the conventional home. Copy the block with the script wrapper switched on. In Avada or any builder with a code element, drop it into the page directly. If your SEO plugin has a field for custom schema, paste the bare JSON with the wrapper switched off. For a site-wide type such as Organization or Website, add it once through your theme header rather than page by page. Whichever route you take, put the published URL through the Rich Results Test afterwards, because a caching layer or a security plugin will occasionally strip a script tag it does not recognise.

All three express the same Schema.org vocabulary. Microdata and RDFa do it by sprinkling attributes through your existing HTML, which means your structured data and your layout are welded together and a redesign can quietly destroy both. JSON-LD sits in its own script tag, untouched by anything the theme does. Google has recommended it for years, it is what this tool produces, and there is no practical argument left for the other two on a new build.

Match the type to what the page genuinely is. Blog posts and news take Article or one of its narrower forms. Shop pages take Product, with a price, stock status and real ratings if you have them. Cooking content takes Recipe. Anything with a date and a venue takes Event. A physical premises takes Local Business. About pages and author boxes take Organization or Person. Question-and-answer pages take FAQ. Navigation trails take Breadcrumbs, and your home page takes Website. One page can carry more than one block — an article with breadcrumbs above it is a normal, valid arrangement.

No. Valid markup makes you eligible, and eligibility is where Google’s discretion begins. It decides per page, per query, per device. Pages that meet the spec exactly, sit on a site with some established trust, and answer an obvious intent do best. Pages whose markup describes something the visitor cannot actually see on the page do worst, and repeated mismatches can earn a manual action. Fix what the readiness strip flags, confirm with the Rich Results Test, then give it time.

Yes, for different reasons than in 2022. Google withdrew the FAQ accordion from ordinary search results in August 2023, keeping it for a small set of government and health authorities. The markup itself never stopped being read. AI Overviews and voice assistants lean on it heavily, because a clean question-and-answer structure is exactly the shape they want to quote. That is why FAQ is still in this generator, with a banner attached so you know what you are and are not buying. HowTo was dropped the following month with nothing left behind, so it is not offered here.

Often not, and the honest answer is to check what they are already emitting before you add anything. Both plugins output a reasonable site-wide graph and cover the common post types. Where they fall short is the page that does not fit their assumptions: a job posting, a course, a product outside WooCommerce, a local landing page for a second location, an event. Paste your live page source into the import panel and you will see exactly what is there. Add a block for what is missing; do not duplicate what is not.

Deliberately not. Crawling on your behalf would mean a server, a queue, an account and a daily allowance, and that is the point at which a free tool starts asking you to upgrade. This one runs entirely in your browser instead. The import panel gets you most of the way there: open your page, view source, paste it in, and the tool pulls out any structured data it finds and loads it into the form. Nothing is uploaded, nothing is logged, and there is no limit on how many times you do it.

Want the markup handled properly across the whole site?

Generating one good block is straightforward. Keeping structured data correct across hundreds of pages, through theme updates and plugin changes, is infrastructure work. That is what we do.