Posted on 20th February 2017 by Jazsmine.Stroulger in SEO


Abstract
: Literature and publications often refer to Structured Data as a means to increase awareness of a Brand (or website) in the SERPs. This article introduces the use of Structured Data and JSON-LD, suggesting an integrated approach to modelling content using Schema.org to reduce uncertainty and equivocality, favouring machine learning by the Search Engines. This article offers insights on both benefits and limits of this approach and presents examples of websites where this procedure has been used intensely and how their organic traffic profile (from the Google Search Console) has changed.

Introduction

The web offers a wealth of data on just about anything you can immagine. A common and acceptable definition of data is:

“an array of values, or qualitative/quantitative variables in raw or unorganized form such as alphabets, numbers, or symbols that refer to, or represent, conditions, ideas, or objects”

But data is helpful only when we are capable of processing it, providing meaningful value added insights, derived from an analytical process of rationalisation and interpretation: for example a table of average temperatures over a number of years is simply seasonal data based on statistics. It becomes information when a climatologist interprets the data and provides insights on the global warming phenomenon.

The Web is Dumb

With billions of pages online, where Anyone can say Anything about Anything, the task of understanding and correlating all this data is challenging to say the least.

Structured data can be used by the search engines to increase their understanding of our content by disambiguating (i.e. reducing Uncertianty and Equivocality).

Search engines consume structured data when available not only to enhance listings with rich snippets, but also to enhance the User Experience with more accurate search results while using fewer resources to understand the context and extract structured data from HTML pages.

Structured Data is Enabling

Well structured data is a key factor which drives automated processing: the more regular and well defined, the easier their reuse.

HTML is ideal to format and manage content but not data, because it lacks the capability to link data which is distributed across the web: a standard mechanism is necessary in order to define the existence and specify the meaning of connections between data – Linked Data.

Linked Data was discussed by Tim Berners-Lee in this article where a set of best practices for publishing and interlinking data were introduced to make the web grow  and have become known as the Linked Data Principles:

  1. Use URIs as names for things
  2. Use HTTP URIs so that people can look up those names
  3. When someone looks up a URI, provide useful information, using the standards (RDF, SPARQL)
  4. Include links to other URIs. so that they can discover more things

The first principle introduces the conceptual activity of linking not only to reference documents and digital content, but also everyday things and abstract concepts such as places, people, relationships, services, professions, products, creative works and more in general anything we deal with and are required to describe.

The second principle is self explanatory and the third covers more technical aspects which go beyond the scope of this article. The fourth principle advocates the use of hyperlinks to create a network of linked data, based on the assumption that Anyone can say Anything about Anything and everyone should have a common understanding.

The web we know today is a web of documents linked to one another in a random and uncontrolled way: any association to things is totally up to the user who can understand exactly what the content is about because of one fundamental cognitive capability humans are capable of: providing context.

Structured data is the answer to this necessity.

Structured Data Implementation

Structured data is now implemented with Schema.org

Schema.org is:

“… a collaborative, community activity with a mission to create, maintain, and promote schemas for structured data on the Internet. In addition to people from the sponsoring companies (Google, Microsoft, Yahoo and Yandex), there is substantial participation by the larger Web community.”

A detailed description of the schema.org vocabulary is beyond the scope of this article. A great starting place to familiarize with Schema.org is this getting started page which explains the process of structured data implementation.

Inline Implementation

Structured data implementation can be inline by adding additional tags to the HTML of web  pages. Microdata, a set of tags introduced with HTML5, provide a limited solution in this respect, however this solution, apparently simple to implement, presents a number of complications as the amount of structured data increases.

Simple things like access to a template can become a corporate nightmare: an additional <div> required to embed items can break the page layout.

Another aspect which complicates the inline approach to structured data implementation is the necessity to use a growing number of meta tags (to define Types) which drive the definition of things. Meta tags inline will create undesired spaces and disrupt the page layout. This can be solved with a display:none CSS rule. Again, this solution may create unforeseen problems in the page layout and could lead to problems if for some reason the website were reviewed by Search Engine Staff and such implementation found to be in contrast with relevant guidelines, causing discussions and problems if the structured data were to be removed.

These and perhaps other reasons induced all major search engines to consider an alternative and less intrusive ways to implement structured data.

JSON-LD

JSON-LD offers the advantage of a seamless integration of structured data, avoiding the problems described in the previous paragraph: major benefits are no interference with page layout, ease of implementation at any point of the page.

For example, a JSON-LD file can be easily implemented in wordpress via the per page add to head plugin for wordpress, or an ad hoc software development.

Example of a JSON-LD File for a Blog Post

The following is a simplified example of a JSON-LD file for a Blog Post:

<script type=”application/ld+json”>

{“@context”: “https://schema.org/”,”@id”: “http://www.example.com/blog/blogposturl”,”@type”: “BlogPosting”,”inLanguage”: “en-US”,”dateCreated”: “2017-01-18″,”datePublished”: “2017-01-19″,”dateModified”: “2017-01-19″,”headline”: “Headeline”,”description”: “Description”,”keywords”: “Your Keywords”,”mainEntityOfPage”: “URL of Entity”,”about”:

[{“@type”: “Thing”,”url”: “URL “,”name”: “NAME”,”description”: “Description”}],”author”:

{“@type”: “Person”,”name”: “Author’s name”},”publisher”:

{“@type”: “Organization”,”name”: “Name of Organization”,”url”: “http://www.example.com/”,”location”:

{“@type”: “PostalAddress”,”addressCountry”: “USA”,”addressRegion”: “STATE”,”addressLocality”:

[“Location 1″,”Location 2″]},”logo”:

{“@type”: “ImageObject”,”url”: “http://www.example.com/logo.jpg”,”width”: “223”,”height”: “93”}},”image”:

{“@type”: “ImageObject”,”url”: “http://www.example.com/wp-content/image-name.jpg”,”height”: “114”,”width”: “72”,”description”: “Description of image”},”articleBody”: “Blogpost Text Here”}

</script>

You can test this file – copy and paste here to see the outcome.

The Bigger picture with Structured Data

Nowadays, most literature online will focus on the benefits of structured data for a limited number of types (Hotels, Recepies, Events). A comprehensive and strategic approach to structured data can offer much more than an enhanced presence in the SERPs.

Long term implementation on a number of websites in different niches has shown a substantial change in the quality of organic traffic.

Legal

Structured data has been implemented systematically on a legal website which had been heavily penalized with the implementation of Penguin end of 2012. The website had lost 90% of organic traffic. After an extensive (and painful) backlink cleanup, a content curation strategy was adopted: 4 to 5 blogposts per month (every month) have been implemented using structured data.

The screenshot from Google Analytics shows the significant increase in traffic over the past 36 months. During this time there was no link building campaign.

What is interesting to point out and understand is that Structured data, at this point in time, may not provide a competitive advantage, but it will influence the quality of your traffic: you’ll get better traffic not more traffic.

This factor alone should encourage SEOs to take a closer look at using JSON-LD to implement structured data.

Final Considerations

Structured data is far more than achieving sexy snippets in the SERPs which should be viewed as an additional advantage.

How you should proceed in using Structured Data

  1. Identify the most relevant Properties and Types which best describe your business
  2. Create a model of your website and remember the devil is in the details
  3. Implement, test and verify constantly: Schemas are being constantly updated – what validates today may not tomorrow
  4. Monitor the evolution of your organic visibility, take note of the new keywords you should see in your Webmaster Tools and Search Console – are they useful and representative? If not you will need to review your structured data model

Sign up to the mailing list for more info about the event