Structured Data That Matches the Page
Search engines and AI assistants read web pages differently from people. Alongside the visible text, many pages include structured data: a block of machine-readable information that says, in effect, "this page is an article, written by this person, published on this date, by this organization, and here are the questions it answers."
Structured data can help a page be understood and presented well. It can also be abused, and search engines have strict rules about that. Our approach is simple to state: structured data must describe what a visitor can actually see on the page. Nothing more, nothing invented.
What we generate for articles
For blog posts in our platform, we generate structured data that describes:
- The article itself: its headline, description, main image and the dates it was published and last changed
- Its type: a general blog post, or a technical article for step-by-step guides
- The author: the person named in the byline, with a link to their author page
- The publisher: the organization behind the site, with its logo
- Its topic: the category it belongs to and the tags on the post
- Its language
Every one of those facts comes from the post itself or from the site's own settings. The author is the real byline. The dates are the real dates. The image is the image visitors see.
FAQ data only for visible questions
Many of our articles end with a frequently asked questions section. When they do, we also describe those questions and answers as FAQ structured data, drawn directly from the visible section. If a question is not on the page, it is not in the data.
We are careful about expectations here too. Search engines show FAQ rich results only for certain kinds of sites, so we never promise that FAQ data will create special search listings. Its value is in clearly labeling questions and answers for any system that reads the page, including AI assistants that look for direct answers.
Fixing a gap in our own platform
Our blog editor has long let authors choose a structured data type and enter FAQ entries for a post. While preparing this blog, we found that the public page was not using either: every post was described as a basic blog post, and the FAQ entries went nowhere. A setting that does nothing is worse than no setting, because it misleads the people who fill it in.
So we fixed it. Posts now use the type the author chose, include author and publisher details and add FAQ data when the author listed questions. It was a small change with a large effect across every site on our platform, and a good reminder of our Zero Dead Code rule.
What we never generate
Some kinds of structured data are tempting because they can make search results more eye-catching. We do not generate them unless they are true and supported:
- Reviews and ratings without real, consented reviews from real customers
- Prices or availability that differ from what the page shows
- Authors or credentials that do not match the visible byline
- Dates changed to make a page look fresher without a real update
- Hidden claims that appear only in the data, not to visitors
Search engines treat misleading structured data as spam, and they are right to. It deceives the people searching.
Organization details belong to the owner
The publisher information on every page comes from the site's own search settings: the organization's name, logo, social profiles and type. Organizations fill those in once, and every page benefits. We only fill in details that the organization has confirmed, such as its official social profiles. We never guess an address, phone number or rating.
Structured data is not a ranking trick
It is worth saying plainly: structured data does not force a page to rank, and it does not make AI assistants cite a page. It describes the page accurately so that systems can understand it. The page still has to be genuinely helpful. You can read more about what we do and do not do for AI search in what we do for AI search visibility.
Checking it
After a release that changes structured data, we inspect real pages and validate the output: types, required fields, matching dates and matching FAQ text. The goal is boring correctness, every time.
Frequently asked questions
What is structured data?
Structured data is machine-readable information added to a web page that describes its content, such as the article's author, publish date and questions it answers.
Does FAQ structured data guarantee special search results?
No. Search engines show FAQ rich results only for certain sites. FAQ data still labels questions and answers clearly for systems that read the page.
Can I add review stars to my pages with structured data?
Only when the reviews are real, collected from customers and shown on the page. Fake or self-written review data violates search engine policies.
Related Posts
What We Do for AI Search Visibility, and What We Do Not
Everyone wants to be cited by AI assistants. Here is what genuinely helps, from crawlable pages to truthful structured data, and the tricks we refuse to use.
Why Every Import Starts With an Automatic Backup
Imports change a lot of data at once, so we never start one without a verified recovery point. Here is how the automatic backup step works and why it is not optional.
AI Drafts Are Not Finished Work
AI can produce a full website draft or article in minutes. Here is why we treat every AI result as a draft until a person reviews it and chooses to publish.
Designing AI Features That Refuse to Invent Facts
AI models are confident even when they are wrong. Here is how we design AI features to work from approved sources, admit gaps and hand off to people instead of improvising.
Audit Trails: Recording Who Changed What
When something changes unexpectedly, the first question is who changed it and when. Here is how audit trails answer that question without becoming a privacy problem.
Comments (0)
No comments yet. Be the first to share your thoughts!