Rich snippets for smarter search engines
If you’ve got a website full of useful written content but the search engines seem to be ignoring it, you may be feeling quite frustrated, asking yourself why you bother and cursing the stupid search engines for not appreciating your excellent content.
If so, you’re quite right to curse them. But if you think there’s nothing you can do about it, think again. Microdata, also known as “rich snippets”, may be the answer to your problem.
Stupid search engines?
Fact: search engines are stupid. They’re automatons, not sentient beings. Although they can pick up the words on a web page, they can’t process them in the same way that human readers can. So if you write a sentence such as, “To make ale you need water, malt, hops and yeast”, a search engine won’t recognise that water, malt, hops and yeast are ingredients and that they can be used in a recipe to produce a finished product called ale.
Rich snippets are a way of telling the search engines what part or all of the content of your web page is about, and explaining the connections between a main concept (like a person, a business, a recipe or a film) and its various attributes (name, address, age, ingredients, actors, ratings, etc). They rely on exactly the same content that the human reader sees, but they’re included in the page source code in such a way that they don’t affect what the human sees.
So yes, it’s a form of SEO. But just for a change, it’s one that’s actively promoted by the search engine providers themselves so that the SEs can provide better results. In fact, Google, Bing and Yahoo! were responsible for the whole initiative in the first place.
Scheming for success
The whole system’s based on a “schema”: a hierarchical way of organising things into different types, each with its own set of attributes.
As it happens, the top level of the hierarchy is called, very simply: “Thing”. It has a set of just five very basic attributes that you might expect to find for anything mentioned on the Web: “additionalType”, “description”, “image”, “name”, “url”.
Below that, you have eight second-level types, each with its own sub-levels: CreativeWork (eg Article, Book, Movie, WebPage); Event (BusinessEvent, MusicEvent, etc), Intangible (Audience, Enumeration, Rating etc), and so on.
You can find the whole schema set out, appropriately enough, at http://schema.org/. It’s still a work in progress, of course, so not everything you could ever think of may necessarily fall neatly into just one of the categories. (The people at schema.org explicitly recognise this, but they’ve also taken some account of it in devising the schema: for instance, AccountingService falls under both the FinancialService and the ProfessionalService categories.)
Micro management
Not in this post! I’m not going into the nuts and bolts of how to add microdata here. Suffice to say that it’s not too hard to implement – it’s just a few extra attributes to existing mark-up tags. Anything you want to add a snippet to gets wrapped in a <div> tag with a reference to the appropriate schema.
You can nest schemas, as well, so if for instance you have a page about an organisation and want to give its address, you start the page with a <div> linked to the “Organization” schema and insert another <div> linked to the “PostalAddress” schema.
Riches beyond compare?
Well, maybe not. But some commentators do reckon that you’ll get an instant boost to your pages’ SERPs if you implement rich snippets. I’ve only just started dabbling (a pal over on the Experienced People forum mentioned them), so it’s too early for me to have an opinion on whether it works or not.
One thing you may notice is that the results for your pages get some extra bits for people to look at. For instance, if you’ve included the address of an organisation then you may get a little link underneath your entry, giving people the chance to look at a map of the location.
As with any kind of SEO, though, the law of diminishing returns is likely to set in. In theory you could implement rich snippets for everything you mention on every page on your website. In practice, that would take a huge amount of extra time, and I suspect you wouldn’t get a great deal of reward for it. So it’s probably best to start with just the most important information that you think is likely to attract visitors, and then just experimenting here and there to see what works.
2 Responses to “Rich snippets for smarter search engines”
Great topic for discussion and one that is still causing a lot of head aches for some websites. The one important thing we noted, was as you mentioned diminishing returns it is also important to begin by tackling terms your not ranking for. It is all to easy to lose or diminish SERPS temporarily or even permanently by applying schema.
In much the same way as one can rank no1 for M.D in Detroit but not appear for Medical Doctor Detroit. Implementing new mark up can change what the serach engines perceive as your target or they will “help” you by changing your relevance. This may not be to the most searched or most profitable terms though.
So slowly, test, implement, test, wait, check and compare as with all SEO efforts.
Good post and definitely an important tool webmasters need to start paying attention too.
Excellent point, Brian – thanks. It’s always a good idea to keep an eye on how implementation of a newly added feature affects things. All the more so for a new system like this that is clearly still bedding down and hasn’t been widely taken up yet.
(Sorry not to reply earlier – we’ve been travelling!)