What Is Hummingbird & The Entity Search Revolution – SMX West 2014
By Tad Miller
Mar 13, 2014
More Articles by Tad
This session stars our own Janet Driscoll Miller, Warren Lee from Adobe, Marcus Tober from Search Metrics, Inc. and is moderated by an admittedly punchy Danny Sullivan. The session description says:
At this session, you’ll learn the difference between entity vs. keyword search on both Google and Bing, and how SEOs can – and should – take advantage of this change.
Warren is responsible for global Search for Adobe. His points on what is Hummingbird & Entity Search:
- The world is changing in how it searches and Hummingbird is Google’s response to those changes
- Hummingbird is a brand new search engine that impacted 90% of search results
- It does’t impact the original Google SEO ranking factors
- Hummingbird addresses the huge growth in mobile devices
- The Google Knowledge Graph was the first step in Entity Search
- Search Entities are about relationships between keywords. Bill Slawski has the definitive data on how this works at SEO by the Sea
- Keyword co-occurrence is a method for substituting search terms like “place”
- Query expansion based on previous search is another way of establishing relationships between keywords and eliminating problems like keyword horizontal relevance
- Previous search queries can change what ranks on your keywords
- It doesn’t impact people in SEO 99% of the time, but there are some things to consider with your content strategy going forward
- LSI Keywords and Ubersuggest are tools that will give you different keyword differences that you can focus on to capitalize on search entities
Searchmetrics did a study on the impact of Hummingbird. Their findings are:
- Contextual Search – the context matters. Previous searches really matter.
- Conversational Search – Hummingbird is the foundation for voice search and understanding meaning and semantics.
- Semantic Search
- Google is trying to understand complex entities to understand and answer questions.
- Google is trying to understand desire or intent in search queries with Hummingbird.
- Google is delivering search results for many similar queries now under Hummingbird – like they were all the same query (kind of like broad matching in paid search).
The number of unique domains in search results dropped by 6% after Hummingbird rolled out. Findings by Query Type were:
- With Navigational Queries, total unique domains dropped 4%
- Informational Queries, total unique domains dropped 6.5%
- Question Queries, total unique domains dropped by 4.5%
Hummingbird’s focus is on what the user wants. Hotel Berlin or Berlin Hotel now deliver the same results and it used to produce different results for each. Google now delivers fewer URLs per query than before Hummingbird.
Knowledge Graph: Hummingbird isn’t just about Knowledge Graph. There hasn’t been an increase in appearance of Knowledge Graph after Hummingbird’s launch. Only question queries increased the use of Knowledge Graph results after the launch of Hummingbird.
Janet Driscoll Miller
Don’t Press the Panic Button: Hummingbird is tied to the mobile revolution. Google is just trying to prepare for it with the use of Hummingbird and Knowledge Graph.
Set Your SEO Priorities: Do You Want to Be Part of the Knowledge Graph? It’s a Conundrum. Information Publishers are the most threatened by Knowledge Graph. Non-profits are also threatened by not getting click throughs and getting a chance to give calls to actions to readers. Danny Sullivan added that publishers need to do more than just give facts if they don’t want to be as negatively impacted by Knowledge Graph. Advertising within the knowledge graph is also a threat to your AdWords Advertising because the Knowledge Graph is pushing down the ads.
Should you optimize for the Knowledge Graph? Yes. It is an opportunity. But you need to prepare:
- Is your content well written?
- Is it conversational?
- Use pronouns.
- Be clear and don’t use “jargon”. Write how people actually speak.
If you want to get into Google Now on mobile devices, you need to optimize for Knowledge Graph. Use of structured markup or Schema can do this. You can test your markup with Google’s Structured Data Testing Tool. You can also enter the data into Freebase.com.
Share this article