Google’s Rich Snippets: An Introduction to Semantic Markup

At the end of last month Google announced an update to Rich Snippets. Rich Snippets, which debuted in mid-May of this year, are intended to provide more in-depth information – such as ratings for videos or restaurants or people’s social and professional information – in the snippets on the search result pages, giving the searcher a “richer” searching experience.

How do Rich Snippets work?

In order to produce a Rich Snippet, Google looks at semantic markup (programming code that explains meanings) that webmasters have added to their pages, marking data that humans understand naturally so that machines can understand it also. Basic HTML has plenty of semantic markup that explains data structure, for example, the <h1> tag to denote a heading or the <p> tag to denote a paragraph.

To see a simple explanation of how more complex semantic markup works, take a look at the following:

You know that what you are looking at is a recipe. (Looks pretty good, no? If you are so tempted that you end up making the recipe, tell us in the comments whether it’s worth it.) You know that the text contains (among other things) the title of the recipe and a list of ingredients consisting of the types of ingredient and how much of each you will need.

If basic HTML is used, all a computer knows is that it is looking at a title and a list that contains numbers. It could be a recipe; it could be the inventory of a store; it could be the list of final grades for Geology 101.

More advanced semantic markup tags different parts of the page with code that “explains” to the computer what the data is referring to, and the relationship between different parts of the data. In the above example, the semantic markup would effectively tell the computer something like this:

<recipe title> Chocolate Chip Cheesecake Brownies

<list of ingredients>

<amount of ingredient>1 <measure of amount>cup <type of ingredient>shortening

<amount of ingredient>1 <measure of amount>cup <type of ingredient>brown sugar

<amount of ingredient>1/2 <measure of amount>cup <type of ingredient>white sugar

<amount of ingredient>1 <measure of amount>teaspoon <type of ingredient>vanilla extract

There are a number of existing vocabulary systems for semantic markup, enabling you to let machines understand and make connections about information on people, organizations, products, locations, events, media, and jobs, among other things (there’s even a vocabulary for recipes!). If all information was tagged, a search engine would ostensibly be able to return “all brownie recipes with no oil that are written by people who live in Oklahoma and are friends of my friends on Facebook.”

We’re not there yet. Google only visibly uses this information for its Rich Snippets, and only for reviews and for social/people profile sites, although they say that they do use the information from other markups. The two systems that Google supports are microformats and RDFa. As you may have noticed, the above mouth-watering brownie recipe does have ratings, but they do not appear in the search result snippet for that page:

as they are using a system other than microformats and RDFa.

More Rich Snippet support will be added in the future for markup that is helpful to searchers and that gains wide use among webmasters. E-Commerce sites, in particular, may be likely to benefit.

Which system should you use and how should you use it? Stay tuned for the next blog post on semantic markup and Rich Snippets! In the meantime, if you have your own opinions on the use of semantic markup, let us know in the comments below.

Bookmark