WordPress Schema Markup: Which Types Earn Rich Results in 2026
Which schema types still earn rich results in 2026, which have been cut back, and how to add valid JSON-LD to WordPress without writing any code.
In 2026 the schema types that still earn a visible rich result in Google are Product, Recipe, Event, Course, Breadcrumb and Local Business. Article rarely produces a visible feature any more but keeps you eligible for Top Stories; FAQ has been narrowed to a small set of authoritative sites; HowTo has largely been retired as a rich result; Organization has no per-page rich result of its own. All ten still do a second job that has nothing to do with Google's decoration: they tell answer engines such as ChatGPT and Perplexity exactly what your page is, who wrote it and what it claims, which is why the practical advice is to add every type that genuinely describes your content and to expect a rich result from only some of them.
Disclosure: RankWave AI and Hydrogen SEO are both built by Sheesh Labs. Where this guide uses Hydrogen SEO's documentation, its schema tools or its plugin benchmark, read it as a builder's view rather than an independent one. Hydrogen SEO is Beta software (v0.0.20 at the time of writing).
What is schema markup and why JSON-LD?
Schema markup is structured data: a block of machine-readable text on a page that states, in a shared vocabulary from schema.org, what the page contains. This page is an Article by this author published on this date. This page is a Product at this price, in stock. This business is at this address and opens at these hours. Visitors never see it. Crawlers, search engines and AI systems read it so they do not have to infer those facts from your prose.
JSON-LD is the format to use. It is a single <script type="application/ld+json"> block, usually in the <head>, that is independent of your visible HTML. Google recommends it, it is the format every mainstream WordPress SEO plugin emits, and because it sits apart from your markup it survives theme changes and block edits that would break the older inline microdata approach. A plugin that renders it on the server, with no front-end JavaScript, gives every crawler the same block whether or not that crawler executes scripts. Hydrogen SEO's schema markup overview explains the server-side approach in more detail.
One limit to hold onto throughout: schema describes content, it does not improve it. Marking up a thin page gives you an accurately described thin page.
Which schema types still earn rich results in 2026?
The honest status of each type, drawn from Hydrogen SEO's guide to which schema types actually work and its schema types reference. "Visible" means Google still shows an enhanced listing for the type on ordinary sites; eligibility is never a guarantee, because Google decides per query what to display.
| Type | Visible rich result in 2026? | Who needs it |
|---|---|---|
| Product | Yes: price, availability and rating snippets | Anyone selling anything; one product per page |
| Recipe | Yes: recipe cards and carousels with image, time and rating | Food publishers; one recipe per page |
| Event | Yes: event listings with dates and location | Webinars, conferences, workshops, meetups |
| Course | Yes: course listings | Course and training providers |
| Breadcrumb | Yes: readable trail in place of the raw URL | Every site; enable site-wide |
| Local Business | Yes: knowledge panel and local details | Any business with a physical location or service area |
| Article / BlogPosting | Rarely a visible feature; Top Stories eligibility | Every blog and news site, on every post |
| Organization | No per-page rich result; identifies the publisher behind the site | Every site, once, site-wide |
| FAQ (FAQPage) | Narrowed since 2023 to authoritative government and health sites | Support and product pages with genuine visible Q&A |
| HowTo | Largely retired on most surfaces | Step-by-step tutorials, for machine readability only |
Two of those rows are the ones people get wrong. Recipe is one of the most visually rewarded types left in search, and publishers who fit it should mark up completely rather than partially. FAQ is the reverse: it was over-applied for years because it produced an eye-catching dropdown, and the dropdown is now rare. The markup is not dead, as the next section explains, but adding it for SERP real estate is a plan from 2021.
Which types are worth adding for AI answers even without a rich result?
Four of the ten earn their place without any visible reward from Google: Article, FAQ, HowTo and Organization. The reason is the same for all of them. An answer engine assembling a response has to decide what a page is, who stands behind it, when it was published and whether its facts are structured enough to lift safely. Prose forces that system to infer; JSON-LD states it. No AI provider publishes how heavily it weighs structured data, and anyone who tells you otherwise is guessing. What is known about retrieval systems is that explicit beats inferred, and schema is the web's standard way of being explicit. That is why it sits alongside content structure and llms.txt as a pillar of answer engine optimisation.
- Article states authorship and freshness, the two signals both search engines and assistants use to decide whether to trust and cite you. Add it to every post.
- FAQ packages each answer as a discrete, self-contained unit, which is exactly the shape an answer engine extracts. Lead each answer with the direct answer; machines usually take the opening sentence.
- HowTo does the same for ordered steps. One imperative action per step.
- Organization ties the whole site to a named publisher with a logo and a URL, so a citation has somewhere unambiguous to point.
Add these where the content genuinely fits and treat them as machine-comprehension infrastructure rather than as a lottery ticket. The wider set of habits, including how to structure the visible content these types describe, is in our 25-point WordPress SEO checklist.
How do WordPress SEO plugins differ in schema output?
Considerably, and in a direction most people would not predict. The Hydrogen SEO team installed seven SEO plugins one at a time on the same WordPress 7.1 site, left every setting at its default, and counted the JSON-LD each printed on a standard post. The numbers below come from the Hydrogen SEO seven-plugin benchmark, re-run on 8 September 2026. Sheesh Labs builds Hydrogen SEO, so weigh that when reading the row it wins.
| Plugin (version) | JSON-LD types on a standard post | JSON-LD bytes |
|---|---|---|
| Yoast SEO 28.4 | 12 | 3,976 |
| Slim SEO 4.10.1 | 10 | 3,828 |
| Hydrogen SEO 0.0.18 | 10 | 2,071 |
| The SEO Framework 5.1.4 | 9 | 1,881 |
| All in One SEO 5.0.1.1 | 8 | 4,544 |
| Rank Math 1.0.278 | 6 | 2,665 |
| SEOPress 10.2 (free) | 0 | 0 |
Three readings of that table. Yoast prints the richest graph in the field, twelve types, and if maximum coverage is the goal it delivers. Hydrogen SEO printed nearly the same coverage, ten types, in roughly half the bytes, which the benchmark scored as the most efficient schema; Slim SEO reached the same ten types at nearly Yoast's size. All in One SEO used the most bytes for the fewest types among the plugins that printed any. And SEOPress free printed no JSON-LD at all at defaults, which is not a defect so much as a reminder that "installed an SEO plugin" and "has schema" are not the same statement. Rank Math's front-end output is gated behind registration or a Skip button, so its six types are what you get after that step.
Bytes matter less than correctness, but they are not nothing: every byte of <head> is downloaded before the first pixel renders. Our broader comparison of the same seven plugins, including cost, sitemaps and AI features, is in best WordPress SEO plugins in 2026, and our sister site has a two-plugin deep dive in Hydrogen SEO vs Yoast SEO.
How do you add schema to WordPress without code?
You have three realistic routes: hand-written JSON-LD in your theme, which does not survive a theme change or scale beyond a few pages; a generic snippet plugin, which is better but still manual per page; or an SEO plugin that generates schema from structured fields. The third is the sane default. Using Hydrogen SEO as the worked example, because its documentation is the one we can vouch for, the workflow has three layers.
Plugin defaults, set once
The Schema screen in the plugin menu sets behaviour site-wide: which type applies to which post type, and the default field values. Every field accepts smart tags, so you type @ and insert %title% for the headline, %author% for the author and %date% and %modified% for the dates. One Article configuration then produces correct, page-specific JSON-LD on every post. Enable Breadcrumb here too; it runs alongside any content type.
Per post, for the exceptions
The SEO Settings sheet in the block editor handles the pages that need something different: a pricing page carrying Product, a tutorial carrying HowTo, a support page carrying FAQ. You choose the type, fill its fields, and save. The FAQ and HowTo fields want real questions and imperative steps; the FAQ and HowTo schema guide covers what makes a pair worth marking up. The markup is rendered server-side with the page, so no front-end script is involved.
AI-assisted, with a human review
Two accelerators exist for people who would rather not fill fields by hand. AI-assisted schema drafts markup from a plain-language description of a business or page, and the plugin's Hydrogen AI agent takes a job such as "add FAQ schema to this page", reads the visible questions and answers, structures them and asks before writing anything. Both need a free hydrogenseo.com account (1,000 AI credits on signup); every non-AI schema feature works with no account and no key. Whichever route you take, you review the output before it ships. Nobody should let a model invent a price. What the plugin does beyond schema is on our Hydrogen SEO page.
How do you validate schema markup?
Invalid schema is worse than none: it disqualifies the page from rich results and tells every machine reading it that the site is careless. Validation takes about two minutes a page. Hydrogen SEO's rich results validation guide lays out the loop; the short version follows.
- Rich Results Test at
search.google.com/test/rich-results. Enter the live URL, not pasted code, so you validate what Google actually fetches. It lists each detected type with a status and shows which field is affected. - Schema.org validator at
validator.schema.org. This checks something different: whether the markup is well-formed against the vocabulary, regardless of Google's rich-result programmes. Google's tool only reports types Google decorates; this one sees everything, which is what matters for answer engines. - Errors before warnings. An error means a required field is missing or malformed and the item is ineligible until fixed. A warning means an optional field is absent; add it if the data exists and never invent a value to silence it.
- Purge, then re-test. If you run a page cache or CDN, clear that URL first or the validator will fetch the old HTML. One fix, one re-test.
- Search Console monthly. The Enhancements reports show valid, warning and error counts per type across the whole site, and the Validate Fix button asks Google to re-crawl after a bulk change.
What are the common schema mistakes?
- Two plugins, two graphs. Running two SEO plugins, or an SEO plugin plus a theme that prints its own structured data, produces duplicate and often conflicting JSON-LD for the same page. Pick one primary plugin; if you are moving between plugins, import first and deactivate second.
- FAQ on pages that have no FAQ. Site-wide FAQ markup on pages without visible question-and-answer content violates Google's guidelines and fails validation against the visible page. Mark up only what a visitor can read.
- Marking up what is not on the page. The general form of the previous mistake. Schema must mirror the visible content; describing invisible content can earn a manual action.
- A structured price that drifts from the visible one. The most common Product validation error. Keep the two in sync, ideally by pulling both from the same source.
- Live Event markup on a past event. Expired events with event schema are a steady source of Search Console warnings. Update or remove the markup when the date passes.
- Chasing FAQ and HowTo dropdowns. Adding those types for the visual treatment, then judging them a failure when none appears. Add them for machine readability or not at all.
- Treating a pass as a result. The Rich Results Test confirms eligibility. Google still decides, per query, whether to show anything, and for FAQ and HowTo it now mostly does not.
Where should you start?
For a typical WordPress site the order is: Article on every post and Breadcrumb site-wide, because they are the universal baseline; then your business type, Product or Local Business, filled completely; then FAQ and HowTo where the content genuinely has questions or steps, with honest expectations; then the specialist types, Recipe, Event or Course, if they apply; then validate every template you touched and fold a Search Console check into your monthly routine. That is an afternoon for most sites, and none of it needs code, a paid plan or an account. What it needs is the discipline to mark up only what is true and visible, which is the one rule every answer engine and every rich-result programme agrees on.
Frequently asked questions
Does schema markup improve WordPress rankings?
Not directly. Google has said structured data is not a ranking factor on its own. What it does is make a page eligible for rich results such as prices, recipe cards and event dates, and it states your page's facts in a form that answer engines can lift and cite. Better presentation tends to lift click-through, and that is where any traffic gain comes from.
Is FAQ schema still worth adding in 2026?
Yes, with modest expectations. Google narrowed FAQ rich results in 2023 to a small set of authoritative government and health sites, so most pages will not get the expandable treatment. The markup still gives answer engines such as ChatGPT and Perplexity a clean question-and-answer unit to quote. Add it only where the questions and answers are genuinely visible on the page.
Which WordPress SEO plugin outputs the most schema?
On a standard post at default settings in the Hydrogen SEO benchmark (re-run 8 September 2026), Yoast SEO 28.4 printed the richest graph at 12 JSON-LD types in 3,976 bytes. Hydrogen SEO 0.0.18 printed 10 types in 2,071 bytes, the most efficient. SEOPress 10.2 printed no JSON-LD by default. Sheesh Labs, which builds RankWave AI, also builds Hydrogen SEO.
How do I validate schema markup on a WordPress page?
Run the live URL through Google's Rich Results Test at search.google.com/test/rich-results for eligibility, and through the Schema.org validator at validator.schema.org for well-formedness. Fix errors first, because they block eligibility; warnings only flag optional fields. Purge any page cache before re-testing, and watch Search Console's Enhancements reports for site-wide regressions.
Can a WordPress page carry more than one schema type?
Yes. Breadcrumb plus one content type, such as Article or Product, is the standard pairing, and a contact page can carry Local Business plus FAQ where both genuinely fit. What you must avoid is two SEO plugins each printing their own graph for the same page, which produces duplicate or conflicting markup that validators flag and search engines have to guess between.