The semantic web has been talked about forever but until recently it was all talk and not much action. Adoption rates of RDFa, microdata and other structured data types has been slow due to there being little added benefit to adding them to pages.
Search engines, and in particular Google, have begun using the data in rich snippets. In the past search engines have managed to drive adoption of similar technology. Robots meta tags and robots.txt are two examples of search engines affecting adoption of technology due to their position in the content discovery paradigm.
What is a Schema?
Schemas are HTML tags webmasters use to markup their pages with machine readable language describing the content on the page. The markup can be read and understood by major search providers and other crawlers or programs.
Webmasters use the schema.org vocabulary to add structured data about their pages content. Schema provides the vocabulary and microdata format (open data format standards), to add an extra layer of information to your HTML content consisting of a set of tags introduced with HTML5.
The long term goal of this is to support a wider range of formats. The initial focus is on microdata.
This guide will help get you up to speed with microdata and schema.org, so that you can start adding markup to your web pages. Google provides a rich snippets testing tool, which you can use to test your markup and identify any errors.
One important consideration is who are the controllers of schema.org? A consortium of the major search engines – Google, Microsoft and Yahoo.
Think about this… name five Internet destinations around in 1995 that are still thriving. I can only think of 2 – AOL and Yahoo. AOL is not even the same company it was in 2005, never mind 1995. Yahoo is on the block and could end up owned by people only interested in a very small part of the net.
All participants are public companies. What if the board decides schema.org needs to be a profit center and they decide to license it for a fee? Is there anything to stop this? While it isn’t within the scope of this article to answer that it is important to at least consider possible future developments.
One other thing to consider is almost every search engine to date has gone to broken business heaven. Are search engines really viable to act as “curators” if not them then who should be running this “ORG”? It is one argument for RDFa but that is perhaps better suited to another post.
As search engines begin using this machine readable structured data by adding them as rich snippets to results the adoption rates will increase as there is a definite positive impact to click-through rate (CTR). This is more about additions of stars (reviews), images (videos) and other features of Rich Snippets which make them stand out in the SERP.
This could seem like a ranking benefit but there are many other less costly ways to achieve similar CTR increases without the added cost of the extra markup.
In my opinion, as a pure ranking signal, the SEO Value is 0. Microdata signals are simply too easy to spam and therefore a very noisy signal that any self-respecting search engine wouldn’t put much trust in as a ranking factor.
Microdata Doesn’t Replace a Sitemap
You would think that this data would make the sitemaps redundant. Google uses the sitemap to assign attribution so if your videos are embedded in sites or on a third party platform you still need your valid Google Video Sitemap. Also keep in mind that the best way to get your thumbnail choice is to enter it in your sitemap.
Implementing Video Microdata
Google advises webmasters to add markup for videos. Below is a general video on why you should be adding microdata:
You begin adding microformats to the HTML by first identifying the position you want the information and adding the ItemScope element to specify the HTML block (div) is about an item and you add the Itemtype attribute after it to identify the type it is. here is an excerpt from the schema.org site explaining the different Item Types
The broadest item type is Thing, which has four properties: name, description, url, and image. More specific types share properties with broader types. For example, a Place is a more specific type of Thing, and a LocalBusiness is a more specific type of Place. More specific items inherit the properties of their parent. (Actually, a LocalBusiness is a more specific type of Place and a more specific type of Organization, so it inherits properties from both parent types.
See the Full Type Hierarchy
The scope and type are in the first tag and in this case video looks like this:
< div itemscope itemtype=”http://schema.org/VideoObject”>
Properties of an item use the
itemprop attribute and you add these to the element enclosing the information the property pertains to:
< h2>Video: < span itemprop=”name”>A Title here < /span> < /h2>
Use < span>... < /span> tags to associate the
itemprop attributes to the appropriate text on the page. When you are unable to markup the information (for example an image) use the meta method like this:
< meta itemprop="thumbnailURL" content="thumbnail.jpg" />
It is possible that an itemprop could be an itemtype itself for instance if you were including reviews or people involved in the video like a director or producer and in this case you embed it in the Video data like this:
< div itemprop="director" itemscope itemtype="http://schema.org/Person"> < span>Director: < span itemprop=”name”>A Director < /span> (born August 16, 1954) < /span>
In all cases your opening div tag must have a closing tag as illustrated above. Many of the types on schema.org have “expected types” which are text and URL. This also allows for a the embed type (shown above) which can include child types of the “expected type”.
Pages that are a collection of items should have each item marked up separately and use the url itemprop to link to page like this:
< div itemscope itemtype=”http://schema.org/Person”>
< a href=”name1.html” itemprop=”url”>Name 1 < /a>
< div itemscope itemtype=”http://schema.org/Person”>
< a href=”
name2.html” itemprop=”url”>Name 2 < /a>
Implementation Best Practices
- Don’t attach attributes to hidden content.
- Don’t input values that are meant to deceive anyone or anything (e.g., search engines).
- Test, test and test again and know the value you are getting from this task!