How structured data works, where it comes from, what its origin is, and how it affects SEO and rich results
Search engines have to make a great effort to understand the content of a web page, despite updates in both Artificial Intelligence and algorithms designed to better understand language.
For this reason, Google, Bing, Yandex, and other search engines, joined the Schema.org project, among others, so that websites could have a standard and common format to structure the most important data on a page.
Structured data, therefore, is used to provide relevant data information to search engines.
Google usually takes this data into account when generating Rich Results. However, adding structured data, even when it is correctly implemented, does not guarantee an appearance in rich results, nor does it guarantee that those results will match 100% of what is presented in the structured data itself.
Structured Data helps us appear in rich results because it helps search engines understand our content, but it does not guarantee anything.
This may lead us to ask, “So what is it for then?” Structured Data is just another tool to help search engines understand our content, and this can help us appear in Rich Results, even if it does not guarantee it.
In any case, before thinking about implementing structured data, we must take into account much more basic, and in fact more important, aspects for understanding our website. Such as the correct use of HTML with tags like nav, ul, li, or the famous Headings that we are so passionate about in SEO (H1, H2, H3).
HTML is the foundation of a website’s structure and is what allows Google to differentiate what is a navigation bar, a heading, and a list. Google uses HTML to provide rich results. What Google does with Rich Results is show the user information (which it believes it has understood from the page it extracted it from) that it considers can satisfy the search intent as quickly as possible.
When structured data first appeared, there was a period of chaos and competition around implementation. Today there is greater consensus, and it is recommended to use JSON whenever possible. However, it can still be useful to know other types of implementations to understand alternatives, or to recognize which structured data formats no longer make sense today.
It is a data interchange format that has become a strong alternative to other markup formats such as XML. But moving away from technicalities that go beyond what really concerns SEO, JSON has become the most recommended structured data format by search engines, especially since Bing and Google read it. This is because JSON was not previously read by Bing, and there were three structured data alternatives that had to be embedded in HTML, requiring changes to the HTML structure to implement them. Thanks to this format, that is no longer necessary at all.
Its main advantage is that it does not need to be embedded within the HTML content. This also makes it much easier to express “child” elements or elements nested within the main entity.
Both are HTML specifications that must be included directly in the code to mark structured data. Both are quite similar; however, although Microdata is open community–based, RDFa is more comprehensive. That said, it would not be surprising if these formats were destined to disappear, since they essentially do the same as JSON but in a less complete and more problematic way to implement.
Data Vocabulary was a structured data format promoted by Google that was discontinued in 2020 (not without prior warnings) and that well-known CMSs such as Prestashop continued using until not long ago. The website itself warns of its discontinuation. Therefore, today it does not work and there is no point in using it.
This is usually the biggest question in the SEO sector. Beyond checking what the competition is doing (by the way, the competition does not always do things correctly), the ideal and easiest approach is to check whether the main element of your website appears in Google’s Search Gallery.
It is a very comprehensive guide that explains the main characteristics Google takes into account and how to implement them correctly.
Another recommendation is to go to Schema.org and check whether there is something specifically tailored to your business. For example, if it is the website of a veterinary clinic, it is better to specify that rather than using the generic MedicalOrganization.
On the other hand, it is recommended to add structured data that is relevant to each individual page (although some can be extended across the entire website).
You can nest and include different individual elements as long as they are relevant and are displayed on the page itself.
It is common to see practices such as:
All of these practices can not only prevent structured data from appearing in the SERPs, but can also lead to manual action penalties, as they are considered Black Hat practices.
Other practices, which are not considered black hat but should still be avoided, include adding elements to structured data that are not very relevant to the page itself or providing outdated information.
Although there is the Codelab, which is like a learning “website” provided by Google to experiment with structured data, those who are not used to working with code may find it more intimidating than following a step-by-step approach directly.
What I recommend for anyone who has never worked with structured data before is to take an example from the Search Gallery and start experimenting with the Schema Markup Validator to check that everything is working correctly. For this, it is important to understand how JSON really works, and I recommend learning a bit about the syntax (how braces, commas, and quotation marks work) and testing possible errors with a JSON validator.
It is normal to feel lost with this type of implementation, and it is far too extensive to explain or assimilate all the concepts in a single post. For this reason, in the Technical SEO Master’s program, we explain all these practices step by step and provide ongoing mentoring so that anyone can carry out this type of implementation (and any SEO-related implementation) with maximum confidence and professionalism.
There are also tools, plugins, and extensions that allow people without technical knowledge to implement structured data, but they usually generate very generic structured data and do not always adapt well to all projects.
To validate the structured data of a website, it is best to use both markup validation tools: the one provided by Schema and the one provided by Google.
One of the great advantages of structured data, especially for large blogs or e-commerce sites, is that it can be automated through programming and become very effective.
This allows you to maintain the entire desired structured data structure and adapt it using fields or conditionals.
On my own website, you can see that all articles include structured data. By using variables and custom fields with a predefined structure, I essentially use this simple code (obviously, the variables are declared elsewhere and the fields are pulled from the database):<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "NewsArticle",
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "<?php echo $url_sin_string;?>"
},
"headline": "<?php the_title(); ?>",
"image": [
"<?php the_field('open_graph'); ?>"
],
"datePublished": "<?php echo get_the_date('c'); ?>",
"dateModified": "<?php echo get_the_modified_date('c'); ?>",
"author": {
"@type": "Person",
"name": "<?php if ( get_field( 'author_json' ) ){the_field( 'author_json' );} else{echo "Carlos Sánchez";}?>",
"url": "<?php if ( get_field( 'url_author_json' ) ){the_field( 'url_author_json' );} else{echo "https://sanchezdonate.com/sobre-mi/";}?>"
},
"publisher": {
"@type": "Person",
"name": "Carlos Sanchez"
},
"description": "<?php the_field("metadescription"); ?>"
}
</script>
Try, as much as possible, to avoid having the JSON loaded dynamically via CSR. It can be read and work, but it may cause issues.
Although there is fierce competition to appear in structured data and occupy as much space as possible in the SERPs, it may not always be beneficial to appear there. When Google displays this information directly, and sometimes without citing the source, it can lead to the opposite of what was expected: fewer people may end up visiting our website.
This has happened with websites that provided information such as time, calculations, lotteries, and even song lyrics. A well-known example is GENIUS, which decided to use different types of quotation marks to check whether Google was copying its content verbatim (which it was). This type of practice can make these kinds of websites unprofitable.
While Rich Results can lead to an increase in website traffic, they can also have the opposite effect, making it unnecessary for users to leave Google to see the information they need from the website that hosts the content.
PD: GENIUS took Google to court, but lost because GENIUS does not own the intellectual property of the songs it transcribes and translates.
I currently offer advanced SEO training in Spanish. Would you like me to create an English version? Let me know!
Tell me you're interested