Behind the Scenes of Structured Data

Nov. 14, 2012 | by Kevin Ellen

What is structured data? How does it work with

I strongly believe that structured data is so much more than just the technique behind the famous (or even infamous) rich snippets. In this post I’ll take an in depth look into how it can be used.

ItemType: BlogPosting
ItemProp Name: Behind the Scenes of Structured Data
ItemProp Author: Kevin Ellen
ItemProp Description: An introduction to Structured Data and how the mechanics work. This article should help people understand Structured Data in a better fashion, than just blindly using this to gain Rich Snippets for the Google Search Engine Results.

ItemProp ArticleBody: Structured data is a term not often used and actually is more known under the name of ‘’ or ‘rich snippets’. I have to say that this is not entirely correct, as is a vocabulary (a type of) structured data and ‘rich snippets’ are actually enhanced results in the SERP’s (Search Engine Results Page).

Structured data in a nutshell is actually a way of bringing structure to the data on your webpage and site. This helps the search engines create semantics between pages and better understand the content on a page. If you take a product, like an oven for example, you want the search engines to understand that your webpage is about a product not just terms. Bringing structured data to your website is a good way to educate the search engines that your page is about ovens… And you know what? They can even reward you for helping them, but more on that later.

How does it work?

Structured data exists of silos (itemTypes) and each silo has properties. Each property is defined with the term ‘itemprop’. So let’s illustrate this to keep things simple. We have a book with the name ‘A Book’s Name’ and this book has the short description ‘This is a description of the book mentioned in the page. The book is about…’. And it goes on, as the book has 255 pages and it is on offer £7.99. This information, relatively unstructured so far, will result in the following structure:

  • ItemType: Book
  • ItemProp: Name – Value: A Book’s Name
  • ItemProp: Description – Value: This is a description of the book mentioned in the page. The book is about…
  • Itemprop: numberOfPages – Value: 255
  • ItemProp: Offers – Value: ItemType: Offer
  • ItemProp: Price: £7.99

This is just a simple example of how to structure data, which is straightforward in most cases. However sometimes it gets a bit more technical, i.e. using meta data or using the <link> attribute.

A quick example of HTML code with structured data

Using, one of the vocabularies of structured data, is relatively easy and has little impact on the actual HTML code and data. Most of the items sit inside an element, for text this being most often a paragraph (<p>) or a span (<span>).

<div itemscope itemtype="">
<span itemprop="name">A Book's Name</span>
<span itemprop="description">This is a description of the book mentioned in the page. The book is about...</span>
<span itemprop="numberOfPages">255</span>
<div itemprop="offers" itemscope itemtype="">
<span itemprop="price">£7.99</span>
<link itemprop="availability" href="" />

Rich snippets – Google rewarding structured data users

Rich snippets is a term used to define additional information in the SERPs. These results stand out more and provide more information, and a number of studies have pointed towards improved click through rates (CTR) from results that feature rich snippets. At the time of writing, Google supports various rich snippets (various schemas), such as products (prices and availability shown and with software applications even the operating system it’s compatible with), reviews (star rating shown), a combination of these two and mark-up for recipes. Below are some examples of how they could potentially look:

Rich Snippet Example of Price, Rating and Pre-Order
Product with Rating, Reviews, Price and Availability
Rich Snippet with Example of Price, Rating and Operation System
Software Application with Reviews, Price and Operation System

Different vocabularies

There are different vocabularies for structured data, with being the best known. is a vocabulary invented by the larger search engines (Yahoo!, Google, Bing and Yandex). is now the most accepted; most used vocabulary and has the highest chances of being picked up and displayed in SERPs.

Another format, depreciated for most schemas, is Google has been using this vocabulary for years, but is slowly leaning more towards; however it might be years before data-vocabulary is completely depreciated. An example where data-vocabulary still excels compared to would be breadcrumbs.

Recent update

Google recently announced a new collaboration between and GoodRelations. GoodRelations is a structured data vocabulary which is aimed at eCommerce. This step from shows Google’s intention to understand the semantics between products, items, companies and services better. A few example questions the vocabulary asks for are:

  • What are the dimensions of a certain product?
  • What services does this company offer?
  • What is the logo of this brand?
  • Or could you tell me more about this football stadium (sports location!)

In my eyes, structured data is very important for anyone who would like to create semantic relations between products and keywords. It can help your site be understood better by the search engines and therefore make it easier to find.

Who should be using structured data?

At the moment, Structured Data can be very beneficial for websites containing the following items:

  • Reviews
  • Products (including software applications)
  • Music
  • Events
  • Recipes
  • Authors

But of course, that is only from a CTR point of view, as these are the schema’s being understood well enough for Google to base rich snippets on. However structured data can be useful for any website, company or organisation. With the new collaboration between and GoodRelations the vocabulary of schema has greatly increased.

By structuring data on your pages properly, the chance of Google showing rich snippets in the SEPRs is improved. This in turn will have a beneficial influence on the CTR.

Frequently asked questions about structured data

ItemProp AggregateRating: ItemType AggregateRating

ItemProp RatingValue:4

Itemprop comment: ItemType Comment
ItemProp AggregateRating: ItemType AggregateRating
Rated this post 5 Stars
Itemprop name:

Can I use RDFa and

ItemProp CommenText Yes, you can use RDFa alongside There is almost no need for this anymore, as RDFa is mostly depreciated… However, for some items it can still be very useful.

Itemprop comment: ItemType Comment
ItemProp AggregateRating: ItemType AggregateRating
Rated this post 3 Stars
Itemprop name:

Can you confirm 100% that Google will show Rich Sippets?

ItemProp CommenText No, Google won’t always do what you tell it do, but you can increase the chances of something happening. By providing a clear structure to your site architecture and data -and by being genuine, you can increase the chances of Google creating Rich Snippets.

Itemprop comment: ItemType Comment
Itemprop name:

Can I use data only shown in the source code for structured data?

ItemProp CommenTextYes, you can use in-source <meta> tags to use data which is only positioned in the source. The right implementation would be: <meta itemprop=”{itemprop”} content=”{value}” />.

Itemprop comment: ItemType Comment
ItemProp AggregateRating: ItemType AggregateRating
Rated this post 4 Stars
Itemprop name:

Kevin! What are all these mark-ups you have disturbed the post with?!

ItemProp CommenText I have marked the post up following a potential tree structure for a blog post. The result of this? Something similar to this:

example blog post score


I have given a fictitious score of 4/5 stars to show that even a blog-post could potentially have a rating with reviews. After-all, it is always nice to have some more information in the results page, is it not?

Be Sociable, Share!

    Comment (1)

    • relocation

      Great blog post. ItMay 5, 2013 02:47 pm

    Please note: the opinions expressed in this post represent the views of the individual, not necessarily those of iCrossing.

    Post a comment