Structured data describes, it does not decorate
Schema markup is a block of JSON-LD telling search engines what a page's content means — that this string is a price, that one is an author, this number is a rating out of five.
It does not change how the page looks and it is not a ranking factor. What it can do is make the page eligible for a richer search result: star ratings, prices, event dates, FAQ accordions, breadcrumbs. Eligible, not entitled — search engines decide whether to show them, and frequently do not.
The [msm_tool slug="schema-markup-generator"] builds the common types and omits any node that would carry only a @type and no information, because an empty node is worse than no node.
The markup must match the visible page
This is the rule with actual consequences.
Marking up a rating that does not appear on the page, an offer that is not really available, or an author who did not write it is a policy violation. The penalty is a manual action that removes rich results across the whole site, and it is not quick to reverse.
So: no marking up reviews that were not left. No FAQ schema for questions not shown to visitors. No price in the markup that differs from the price on the page. If it is in the structured data, a human should be able to find it on the page.
JSON-LD is the format to use
Three formats exist — JSON-LD, Microdata and RDFa. Google recommends JSON-LD and it is the right choice for practical reasons: it sits in a single script block rather than being woven through the markup, so it can be added, changed or removed without touching the HTML.
It also survives a redesign. Microdata scattered across your templates does not.
Which type
- Article — blog posts and news. Wants a headline, author, and dates published and modified.
- Product — wants price, currency and availability. The type most likely to produce a visible rich result, and the one most closely policed.
- LocalBusiness — address, opening hours, phone. Genuinely useful for anywhere with a physical location.
- FAQPage — questions and answers that appear on the page. Rich result eligibility for this has been narrowed considerably; it remains useful for machine readability.
- Organization — name, logo, social profiles. Usually on the home page only.
- BreadcrumbList — low effort, and it changes how the URL line displays in results.
One type per page, matching what the page actually is. A product page is a Product, not a Product and an Article and an FAQ.
Required and recommended
Each type has required properties and recommended ones. Missing a required property means no rich result at all; missing recommended ones means a less complete one.
Dates are the usual stumbling block: they must be ISO 8601, and datePublished and dateModified should be honest. Backdating or falsely updating a modified date to look fresh is a well-known tactic and a well-known thing to be caught doing.
Validate before shipping
Two checks, and they answer different questions:
- The Rich Results Test tells you whether Google will consider the page eligible, and for what.
- The Schema Markup Validator tells you whether the markup is valid schema.org, regardless of what any search engine does with it.
Markup can be valid and ineligible. Both checks are worth running.
What to expect
Adding structured data does not guarantee a rich result, does not improve ranking on its own, and may show nothing for months before appearing. Its reliable benefit is that machines can read your page accurately — which matters more each year as search results, assistants and aggregators consume pages rather than display them.
Add it because the description is true and useful, not because it is expected to move a number.