Preparing Your Website for Voice Search and the Semantic Web
Preparing for SEO in 2019 and beyond
Shortly after the world’s first website was launched in 1991 many more began crowding the internet. Search engines filled the demand for structure and accessibility, making it easier to find information by sorting websites based on keywords and backend optimization.
In the beginning, many search results were filled with spammy and irrelevant content. Search Giants began implementing rules and regulations to satisfy user demands and provide a level playing field to earn honest rankings.
With more intelligent algorithms, search engines have drastically improved over the years. To rank well, organizations must follow SEO Best Practices.
Today, aside from regular making their sites look attractive to visitors Marketers and Web Developers are challenged with Rich Search Results, influenced by Structured Data, Semantic Search, Schema and AMP.
Semantic Search
The Semantic Web, refers to a web of data that is machine-readable. A web in which computers can transform structured and unstructured data into an intuitive and responsive database. It is a method of integrating content from around the web. According to the W3C, “The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries”. Therefore, Semantic Search can understand a searcher’s intent, extract answers from around the web and deliver more personalized results.
Semantic Search is the reason why we see News Carousels, and Knowledge Graphs in Google Search, and why your Google Home is able to answer ‘What’s the closest Pizza Place open right now ?’. If you’re unfamiliar with what Rich Search results look like, check out the image below.
The Semantic Web works off of Structured Data, (and some unstructured data as well). Which is why, every modern webmaster should be including Structured Data Markup in their content.
Structured Data
Webpages used to be published purely in html, which enabled web authors to make websites interesting and easy for humans to read.
As humans, we quickly understand the stylistic patterns of publishers — for example, a line of text at the top of the page in size 18, underlined and bold font, is probably a ‘Title’ or ‘Headline’, while a smaller line of text in grey, italicized font that says ‘Written by John Doe’, signifies the ‘Author’ of the article who is also a ‘Person’. While this is easy for humans to understand, it is extremely difficult for search engines.
Historically, and still today, web applications including search engines, price comparison websites (i.e. booking.com), reservation engines (i.e. Ticketmaster), and others wishing to access and display data from the web would be required to build their own custom extractors to convert HTML into Structured Data. These scrapers were often fragile, difficult to build and prone to errors.
Structured Data — is a solution that works by labelling pieces of data within content. For example, if you publish a news article you can use Structured Data to label pieces information, such as: content type, article headline, author, publisher, and more.
For Search Engines (and other applications), Structured Data is easier to understand than plain html web pages.
Today, Structured Data is used by a variety of tools and applications including major search engines, voice assistants like Google Home and Alexa, social medias like Pinterest, Email, and much more. As we can see from data collected by w3techs, the implementation of Structured Data Markup is increasing.
In addition to increased adoption by website builders, the actual Schema Vocabulary is growing everyday, with more and more contributors. It is expanding areas in topics ranging from automobile descriptions to specific product details, and the way that Search Engines, Voice Assistants, and other applications are using this data is growing as well. For example, Google’s Speakable (Beta) which reads marked up sections of news articles to users on Google Voice-activated assistants.
Although plain HTML pages (sans structured data markup), will get indexed and ranked for SERPs. Structured data is really the future of search, the Semantic Web, and it is the only way to get indexed in certain applications or tools.
Schema
Today, the leading resource for Structured Data is Schema.org, the centralized home on the web for the Schema project. The Schema Project began as a colaboration between major search engines Google, Yahoo, Bing! and Yandex (Russian search engine) in 2011, with the purpose of resolving the issue of every search engine trying to decode websites, comprehend, organize and display data. The Schema Project’s mission today is to standardize Structured Data Markup.
Although the Schema project began as a closed group, it has incrementally opened up, moving discussions to W3C public forums and moving decision making to the open, with a steering committee who includes members from sponsor companies, academia and the W3C.
There are different ways of incorporating structured data markup, called Syntaxes. The most popular are Microdata, JSON-LD, RDFa and microformats. The two common vocabularies which can be used with these syntaxes are Schema.org, or Microformats.org. Since 2017, Google recommends JSON-LD whenever possible.
We can think of Schema like a language, and by standardizing Structured Data Markup, it means developers and webmasters only have to incorporate one set of code (Schema) to be understood by all of the tools and applications reading it.
Considering the substantial amount of efforts and collaboration put in to the Schema Project so far, we imagine that it will continue to grow and become an integral part of the future of web development, content creation and generally publishing anything on the internet.
Structured Data and SEO
One of the common myths is that Google can figure out your website without you doing any work, but even though Google is getting smarter at figuring things out and understanding content — you should use the correct markup to benefit from special features exclusive to markedup contents.
Although Google claims that structured data markup is not a ranking signal, it is very possible that it will improve your SEO. For example, Google delivers Rich Search Results, also known as SERP position 0, which is exclusive to content including Structured Data Markup.
According to Blue Nile Research firm, Rich Media results get significantly more clicks than non-rich media, even if the non-rich media is in position 1.
Users are drawn to Rich Search Results because they are better at indicating to the user, what they can expect by visiting the website. Therefore, Rich Results can increase quality CTR, which increases the time spent on site, decreases bounce rate, and more — all of these factors contribute to organic search ranking.
This article from the Search Engine Journal, does a great job at explaining the wide ranging benefits of Structured Data Markup from an SEO perspective. And if you’re still unsure whether Structured Data would be worth the investment, take a look at the list of Rich Search Result features offered by Google categorized by website, and content type.
Schema + AMP
For publishers, Structured Data + AMP is required for many of the Rich Search Results features, including the News Carousel. More information about the requirements for these features, are included on Google’s Webmaster documents, here.
Implementing Structured Data, is essential to prepare for the future. As search becomes more personalized, more accurate, and other applications and tools start utilizing structured data. If your information isn’t readable by these applications using the common schema’s made available, you may be missing out on massive opportunities for visibility on the internet.
In order for your data to appear in personalized results, you must ensure that the relevant search engines are able to understand the data on your site.
If you manage a website, ask yourself this:
- Is my data accessible to Google?
- Why should Google choose my website as a data source?
