SEO Schema

[Schema markup](/schema-markup/) is a block of code — usually JSON-LD in a <script> tag — that tells search engines exactly what your page contains. Think of it as a label for a machine. Your page still looks normal to a visitor, but Google gets a cheat sheet: this headline is an h1, this number is a price, this paragraph is a review. I use JSON-LD because it sits in the head or body without cluttering the visible HTML, unlike Microdata or RDFa.

The vocabulary comes from [Schema](/schema/).org, a shared standard that covers everything from recipes to events. The most common types for SEO are Article, Product, FAQPage, LocalBusiness, and BreadcrumbList. But here is the catch: schema does not guarantee a rich result. Google decides whether to show stars, prices, or expandable questions based on quality signals, not just the markup.

I once added Product schema to a site with 500 pages and saw zero change in two months. Then I focused on the top 20 products, rewrote the descriptions, and added review schema only after actual reviews appeared. CTR on those pages jumped by about 30%. The markup only worked because the underlying content earned the visibility.

Schema matters most when your page competes in a crowded SERP where rich results can make the difference between a click and a scroll. If you run a local restaurant, LocalBusiness schema with opening hours and address can push you above plain listings. For an e-commerce site, Product schema with price and availability gives shoppers a reason to choose your result.

But there is a threshold. I have tested FAQ schema on thin FAQ pages — five questions with one-line answers — and got nothing. Google wants substantive content. The same applies to Article schema: if the article is 200 words of fluff, the markup won't save it.

I always start with a [technical SEO audit](/seo-audit/) to identify which pages already have quality content and decent signals. Then I apply schema only to those. For WordPress sites, plugins like Yoast or Rank Math handle JSON-LD cleanly, but I prefer manual control because plugins sometimes apply markup to pages where it doesn't belong — like adding Article schema to a contact page. That is a common cause of over-markup warnings in Search Console.

Testing is non-negotiable. I run every URL through Google's Rich Results Test before and after deployment. If the test shows errors, I fix them before going live. A broken schema is worse than no schema because it wastes crawl budget and can trigger manual actions.

I used to believe more schema equalled better SEO. Early in my career, I added the same Article markup to every page on a clients site — including the login page. Google ignored the entire set. I then learned that over-markup dilutes the signal; a page about a product should not claim to be an article. Now I match the schema type exactly to the primary content of the page.

Another mistake: marking up content that was not on the page. I added Review schema to a product page that had zero user reviews. I thought it would encourage Google to pull from elsewhere. Instead, the Rich Results Test flagged missing review body. Google's guidelines are clear: the marked-up content must be visible to users. I now only add schema for things I can see on the screen.

I also overlooked the connection between schema and other technical foundations. A page with slow load times and broken [canonical tags](/canonical-tags/) will not benefit from schema. I wasted weeks on markup before fixing [duplicate content](/duplicate-content/) issues that were draining the pages authority. Now I ensure the page is technically sound first, then add schema as the last layer.