“Hummingbird is not just a standard newly released algorithm, it is a learning algorithm that will get better and better over time.” This sentiment by Marcus Tober, the founder and CTO at Searchmetrics Inc., accurately summarized an extensive and intriguing review of how Google’s Hummingbird Algorithm, along with the Google Knowledge Graph, are revolutionizing how SEOs need to approach search optimization. Here are summaries of the SMX East sessions from Warren Lee, Bill Slawski, and Marcus, who each looked at the issues and opportunities from very diverse perspectives.
Warren provided a fascinating look into the immediate effects of the Hummingbird Algorithm. According to Warren, Hummingbird is a new search engine algorithm impacting 90% of search queries in Google. The update was done to cater better to mobile users, and to attempt to learn the contextual meaning behind searches.
Hummingbird uses the Knowledge Graph (see my previous blog post on the topic for more information) in an attempt to understand both search entities and entity search.
Search Entity Includes:
Google is trying to use search entities to learn about relationships between keywords.
Entity Search: Contextual targeting based on search history
Warren provided several methods to optimize for these changes. His self-started main takeaway though is that the long tail is as important as ever to provide search entity relational information. In addition, he provided the following resources to use to get ideas for keywords related to your primary phrase:
Blog: SEO by the Sea
Bill is brilliant at researching Google’s patents to provide insights into what Google is doing. One word he used repeatedly to describe what Google is doing with Hummingbird was “patterns.” Google is monitoring search patterns in an attempt to learn about contextual relationships between words (Search Entity). Furthermore, Bill said that patents show that Google is using aggregate and individualized user behavioral data to impact search results (Entity Search). By doing this, Google is beginning to understand the meaning behind more ambiguous searches to provide users the most relevant results.
One before & after example that was provided by Warren was for a search of the word “headache.” In the past, that search would provide results about what a headache is. However, after Hummingbird was implemented, that same search would instead result in treatment results such as Tylenol’s page. Bill expanded upon this with an example on searches for the word “cats.” Normally Google would send you to pages about the pet, but after searching for Broadway plays in New York, the results for Bill were for the musical Cats. As Google learns more about individual search patterns and preferences, they are better catering results to each user.
Bill also touched on a new Google Knowledge Vault that is extracting chart data across the web to try and fill in any gaps in the Knowledge Graph’s knowledge base. It is yet to be seen how large of an effect this will have on search, but we know Google is collecting a tremendous amount of data.
Marcus delved into the current effects of Hummingbird and what the long-term implications are. Perhaps the most interesting and important aspect he mentioned was that Google is looking for more than just your primary keywords – they’re also looking for related keyword phrases in your content that help give that keyword phrase more context. Marcus says that Google is looking for your page to have a “holistic” view of a topic that will provide searchers a complete answer. Specifically, he stated that SEO has gone from go-down-the-list tactical, to strategy including understanding a topic and building a complete view to answer all of a searcher’s questions.
As a result of these changes, an interesting effect has taken place. According to Marcus, link diversity has decreased as much as 7% since the launch of Hummingbird. That means that fewer websites that Google “trusts” are now ranking, where lesser-known smaller websites used to. It will be interesting to see if this trend continues and what effect it has on the smaller websites, but as usual it means that SEOs will need to adapt to survive.
While these changes are very significant, I see them as encouraging the things Google has always recommended. Significant depth of good content and long tails are as important as ever to show Google that your site is an authority on a topic. However, title tags, meta descriptions, and header text may not be enough anymore. SEOs will need to focus on research into keywords Google thinks are related to the main keyword phrase to provide Google the contextual clues it is looking for to say your website is a legitimate resource.
Join over 4,000 marketers who receive actionable digital marketing insights.