Taxonomy and Search
A taxonomy is a system devised to classify things. The classic example comes from the world of biology. You might think you’re just some guy eating handheld food with your mouth open while reading this blog post. To a biologist, you are of the species Homo Sapiens, of the genus Homo (no sniggering) of the family Hominidae of the order Primates of the class Mammalia of the phylum Chordata and of the kingdom Metazoa.
The hell you say?
All of that stuff lets biologists determine your relationship to other parts of the animal kingdom. It’s a way of ordering the biological world through shared characteristics. I’ve got a backbone and breathe oxygen and so does a whale. But we can’t mate because there are lots of differences between us.(like I don’t have time to hang around the ocean deploying chat-up lines like “oooowweeeiiioouuu – wooooobbbbbiiiiiooot.”*)
Search is increasingly about classification
Remember meta tags? They were an early attempt to establish a taxonomy for web pages. These are the keywords… this is the author… this is the date… etc etc. Abused beyond parody, the search has been on ever since for a way to decipher those signals from natural elements within a page. The principle behind meta tags was sound – it was just that they were such an obvious target that they were rendered useless within days.
You can see it in the flesh in Google Base or whatever the hell they’re calling it these days. They’ve identified a taxonomy for products which they can use to structure their results. To you and me, that means a load of database columns that you have to fill in but to Google it’s an important way for them to determine rankings. Showing items in price order is only possible because all products share a common taxonomy.
Now on the regular schmo version of the web, populated by blogs, semi-literate forums, flash microsites and a billion other formats, establishing a taxonomy is very difficult. But among the more significant and lasting changes in the main Google SERPs over the last 18 months has been the prevalence of news results, maps, sitelinks, localised results and product feeds into the main results as part of blended search.
All of these are evidence of how Google is attempting to create a taxonomy for web content.
Taxonomy of the search itself
Google are clearly looking for a way to create a taxonomy of searches themselves. This is most evident when you hop on and search for Tiger Woods today. Google makes an assumption that there is a likelihood that you fit into an informal taxa of ’prurient sleazeball’ and so you will see today’s searches mixed up with news results about the continuing fallout of his well-publicised affairs.
This is probably organised by responses to search volume. Overnight, millions of people are suddenly searching for Tiger Woods. While that trend happens, those searches become classified under Query Deserves Freshness. And so the rankings are tipped towards news stories from authority domains and fresh, relevant content.
That’s a really obvious example, but deeper within the taxonomy of searches come groups like refinements and modifiers, all of which Google use to try and decipher a searcher’s intent, and therefore the kind of results to serve. Search for JLS and Google thinks: you’re a teenage girl and weirdly obsessive middle aged mum – have some fan pages and images. Search for “JLS album” and Google guesses that you’re a long-suffering dad near to your Christmas deadline and will help you out by pointing you to Amazon.
Products
An obvious one in light of the example I gave earlier, but when you search for something that is a product name, catalogue number or [brand product] then Google will show product listings. Not just because they want to own the shopping space, but because there is a verifiable taxonomy to go on. If you know enough to use a product code then Google shopping is a perfect fit.
Another example are results for hotels in Google maps – now updated to include reviews for specific aspects of a hotel (HT to Patrick Altoft for the example). Google have identified that review sites have developed a kind of industry specific taxonomy of their own. That means they can aggregate information together in interesting ways. So if you want to break into the hotel review/booking market, you’d better be sure that you pay attention to the taxonomy of the sites that Google is using if you want to piggyback your way into these results.

On your pages, using lists and tables to deploy data like product codes can help Google identify your page in this kind of context even if you aren’t in their supplied platform.
Architecture
When you look at the structure of your site, it creates its own internal taxonomy – and Google is keen to leverage it. There’s a hint about it in the appearance of Google Breadcrumbs. It’s long been understood that the priority pages of your site have the best chance of ranking. If there’s a page on your site that is 4 clicks down your navigation, it isn’t going to get much equity from your domain and will struggle to rank in the absence of deep linking or massive domain equity in the first place.
So when you’re busily drawing up all the categories of stuff you sell, think long and hard about where the money is. Putting some gimcrack piece of plastic toss with an 8p margin at the same level of navigational priority as your high value items is a deep failure to understand how Google uses your internal taxonomy.
Local
Stick a place name into a search and Google will give priority to maps listings, because geography is itself a taxonomy. So get your Google local listing and make sure your address is displayed on your site in a recognisable format (I suspect that post/zipcode would be the critical information to include) to give Google as much help as possible in putting your company in some kind of geographic context. Even if you don’t use maps, Google are increasingly using IP addresses to try and tailor local results into search wherever they feel it can be relevant and practical.
Content
When it comes to things like news – triggered by QDF and a great chance to grab a spell on the front page while a story is hot – Google is looking for obvious signals like the inclusion of a date, an author name, a blockquote, an image etc. All of that helps to establish that your content sits in the category of ‘news’ and is therefore perhaps worth a kick up in the rankings.
Similarly, blog posts can be identified through the use of tags, categories and dates. This is all old news to most of us – organisation of information can be as critical as the information itself – but there are always ways to deploy your content in such a way as to help Google identify priorities and relevance.
How can you utilise all this guff?
Firstly, you need to look at how the architecture of your site reflects Google’s increasing attempts to impose its own taxonomy on the web. If If your site uses a blog format, is the timestamp immediately obvious and somewhere high up in the content where Google can look for it and use it? Is there a list of tags? If you’re hosting product reviews, take time to look at the sites Google is listing and look for commonalities in the things they include – as in the hotel example back up the page there. If you’re putting out news content, make sure that you are writing is timely in the wider context of what the world might be searching for. Include an easily identifiable date stamp and author name to help Google file your content in the taxa of ‘news’.
More obviously, take advantage of all the opportunities that Google are giving you for free – news, local search, blog search etc.
What thinking about your taxonomy doesn’t do is establish authority. That still comes from brand exposure and backlink profile. Without those things, the best structured site in the world will still fail to rank. In the end, we’re still coming back to the same things people: good content, properly architected and demonstrating the quality signals that should be self-evident from any 5 minute trawl through the SERPs.
———————————————————————-
* “Your eyes are like spanners – every time you look at me my nuts tighten”

Chris Peterson 889 days ago
http://bloggingwithchris.com./Thanks for great posting, it contains number of applications that helps people create taxonomies and place information objects within their categories, although the amount of automation can vary.
But I will research on this topic and analyze how far it will be work in case of SEO Point of view
Sam Daams 889 days ago
http://twitter.com/samdaamsA great post as usual. SEO is becoming more and more about a holistic, structured approach than anything else.
I guess that’s what the bot in comment 1 was also trying to convey
Amelia Vargo 888 days ago
It always goes back to the same thing, a great site will always list well, because it is offering something of value to people. That’s all Google want to do, serve up great websites, sure they get it wrong from time to time but mostly they give great search – that’s why people use them. It’s al about providing something of quality. And quality always stands out, whichever taxonomy it’s in.
NGT Web Site Design 887 days ago
Nice article. You have a typo here: If If your site uses a blog format
Rifki Jones 881 days ago
http://www.seotra.comA very informative post, like this article, in fact I like this site. Great quality content. We can use our SEO and marketing more effectively if fully aware of Google’s focus on providing the most relevant content with taxonomy.
Regards
Rifki
Michael Jones 879 days ago
http://freemicrosoftpoints.netVery informative post and I agree. Google is constantly trying to adjust their architecture and does seem to have some coincidences for when you search certain keywords. It seems like they are just trying to adjust for the search habit of each user, then classifying sites by rank.
Kieron Hughes 874 days ago
http://webtoastie.co.ukGreat post. I think taxonomy is a topic that people may be a little bit skeptical about when they first hear the word, because it can be confusing, but it makes perfect sense.
Conforming to some form of taxonomy is a good way of allowing the search engines to quickly establish what your content is about, and (hopefully) boost your rankings because of that.
PS – interesting whale chat up lines!
Ralph Windsor 863 days ago
http://www.daydream.co.ukInteresting how the debate about the value of taxonomies and metadata seems forever destined to repeat itself. This article from 2001 makes precisely the opposite point – that Google is destroying the need for human taxonomies by the creation of ‘observational metadata’ (i.e. links)
http://www.well.com/~doctorow/metacrap.htm
It’s ironic that Google are now leveraging the power of taxonomies and metadata, but not so funny that they have total power over their structure or how the content we all provide them for free via the organic SERPS will be end up being classified.