Google’s Panda and Penguin algorithm updates targeted low-quality sites, or sites that had thin content, content farms, spam links, over optimized anchor text links, and unnatural backlink profiles. But these were only updates to the then current search algorithm. Quietly launched in late summer 2013, and publicly announced a month later, Hummingbird is an entirely new algorithm representing a significant shift in search science.
What is a search algorithm? It is essentially a computer program that analyzes search queries and returns to the user the most relevant results possible. For any given query, there are hundreds of thousands, even millions, of webpages that could potentially provide relevant information. In today’s mobile-heavy technological landscape, the information hasn’t necessarily changed but the method by which users search has.
Hummingbird was built to support semantic and conversational search, both of which are heavily focused on “long-tail searches.” With “speaking” technology advances such as Siri & Google Now, Google needed to prepare itself for the future of search. More and more people are asking Google direct questions, and they are expecting the search engine to understand what it is exactly that they are trying to find out. Google’s previous algorithm did support conversational search to a small degree with its Knowledge Graph results (you could search for “Who is Henry Ford? and it would give you a “Card” on the right side in the Knowledge Graph section) but this new algorithm is a much greater advancement. Now, Hummingbird not only applies that logic to Knowledge Graph results, but to organic results as well. Let’s take a look at an example query and how Google viewed it pre-, and post-Hummingbird launch.
Sample query: Where can I find the closest dealership with a black Audi S4 with leather?
In a pre-Hummingbird world, Google would scan its index of webpages for content matching the query or containing a slight variation of the query. It would find the pages with relevant content, order them based on the more than 200 signals it receives such as inbound links or domain authority, and serve them to the user in a search engine results page.
With Hummingbird, Google now deconstructs the query by focusing on the meaning behind each word to gain better understanding of the user’s intent:
“[Where : Adverb meaning at or in what place] can I find the [closest : Superlative adjective meaning “a short distance away”] [dealership : Local business type] with a [black : color] [Audi : Automotive manufacturer] [S4 : Product | Car model] with [leather : Product feature | interior fabric]?”
In the post-Hummingbird example above, we can see that Google is not just looking to match a grouping of words within a document. It is dissecting the query to understand the intent behind the search. It even understands that when you say “the closest,” your physical location needs to be considered (based on history and/or IP address).
To exploit Google’s changes, websites need proper microdata markup (schema.org, hProduct, etc). Dealer.com continuously provides this to Google to deliver your dealership’s website for queries like the one above. Our platform currently uses hProduct markup for our vehicle details pages and product pages. Our new templates, however, will start to utilize schema.org as a replacement, allowing for more in-depth product-based microdata.
We get asked pretty frequently how we’ve been preparing for Hummingbird and if we’ve seen any changes because of it. Our SEO team has been testing schema.org markup since the summer 2011. We are currently building schema.org into our new templates, particularly for products, videos, blogs, and images.
In analyzing the data, we haven’t seen any negative effects on Dealer.com websites that are tied to the Hummingbird algorithm. In fact, we’ve seen positive ranking results as well as more rich snippets displayed on search engine results pages, which is increasing click-through rates and providing dealerships greater exposure.
If you have any questions, please feel free to reach out to me – pete.bruhn@dealer[dot]com.
About the Author: Pete Bruhn is the Associate Product Manager, SEO at Dealer.com. He has eight years of experience in the SEO & search marketing industry, five while working at DDC. He can be found on Twitter at @petebruhn and on G+ at google.com/+PeteBruhn.