Dries Buytaert

Structured data is the new search engine optimization

Two days ago, Google announced "Rich Snippets", a move that is sure to shake up the SEO industry, and cause hundreds of thousands of people to reconsider their skepticism of the semantic web. Yes, that probably includes many of you.

Google's Rich Snippets provide summary information to help users quickly identify the relevance of their search results. For example, if you search for a restaurant, rich snippets may include an average review score, a price range, or more. As users get more sophisticated at search, they'll ask Google increasingly complex questions. Rich Snippets allow Google to stay on top of that trend, and prevents losing users to competitors.

It is very hard for search engines to understand the structure and semantics of data embedded in an HTML page. To create these snippets, Google needs the help of hundreds of thousands of webmasters around the world, and by extension, content management systems like Drupal, Joomla!, and others. Specifically, Google is asking all of us to surface structured data to their crawlers by marking up our HTML with RDFa and Microformats. When Google announced Rich Snippets this week, they really announced support for RDFa and Microformats, and the semantic web in general. This is big.

Initially, Google's adoption of RDFa will disrupt the current approaches to Search Engine Optimization (SEO). With Google entering the RDFa game, the words "semantic markup" will get redefined. Every webmaster wanting to improve click-through rates, reduce bounce rates, and improve conversation rates, can no longer ignore RDFa or Microformats. Structured data is the new SEO.

As I've written before, search engines like Google and Yahoo! will provide the killer apps (e.g. vertical search engines) that the semantic web has been waiting for. Five years from now, we'll look back and say: "All it took was some incentive for the SEO industry." ...

Rich Snippets is a natural step in making search better. It provides a glimpse into the future of search, and tempts us with the possibilities of the semantic web. Right now, Google has a database of pages. If you read beneath the lines of their announcement, what Google is really asking is for us to help them in building giant specialized databases of all products, people, places and events in the world. This provides opportunities well beyond providing rich search snippets. We're turning the web into a giant database for Google (and others) to slice and dice as needed.

For example, it is easy to see that a database of all the job applications in the world, built by crawling hundreds of thousands of independent RDFa-enabled sites, will impact specialized job sites. Or how a database of all the product or movie reviews in the world could affect specialized review sites. It might seem scary at the surface, but it really isn't. On the web, scale and reach are more important than scarcity — you win by setting data free, not by holding it close to your chest.

For many of us in the Drupal community, Google's announcement couldn't be more timely. The Drupal community has been working on adding RDFa support to Drupal 7, and at this very moment, people from the community are gathering in Galway for a week long code sprint to get more RDFa support in Drupal 7 core. Once again, Drupal proves itself to be on the cutting edge, and is taking a leadership role in adopting semantic web technologies. As I said in my DrupalCon Boston keynote 1.5 years ago, I believe that Drupal can become a significant player in the development of the semantic web. It's bullish, and maybe even naive, but I couldn't be more excited about giving the semantic web snowball a small push.

— Dries Buytaert