In January of this year, Google updated the Knowledge Graph. The knowledge graph 2.0 has not yet appeared in the Russian-language search. However, this technology has already made marketers working with English-speaking Google think. Some experts claim that Google uses an updated Knowledge Graph to steal someone else's content and traffic. How fair are these accusations?
What is the Knowledge Graph
Knowledge graph or Knowledge Graph is a Google technology that allows you to generate answers to some user requests directly on the search results page. The search engine analyzes the semantics of the query and provides answers using information databases. Here's how it works:
In response to a request, the search engine issued a brief and specific answer, drawn up in the form of an answer box or knowledge graph card. Thanks to the Hummingbird algorithm, the largest search engine correctly analyzed the semantics of the user question. And where did Google find the right answer? To understand this, take a look at the illustration:
Google took the information for the answer box from the knowledge graph panel, which is located on the right side of the page. In turn, the search engine uses a limited list of sources to form Knowledge Graph, which includes Wikipedia, Google+, Freebase.
Why Google had to update Knowledge Graph
As mentioned above, the search engine uses several sources to form Knowledge Graph panels. All these sources are developed through the efforts of staff and volunteers. Therefore, Google ran into a problem that caused all search engines to be created.
Search engine robots appeared at a time when people stopped coping with the addition of information in the catalogs of sites. The knowledge graph 2.0 appeared at a time when there was a need to go beyond the limited range of sources of information created by people. Simply put, Google does not want to take information for the knowledge graph only from Wikipedia. The search engine learned how to automatically answer questions using all indexed sites.
Here is the answer to the English query "social security tax rate":
Say, the answer box of knowledge graph 2.0 is no different from a similar card of the first Knowledge Graph? Pay attention to the link. The search engine generated a response using data taken from a government site. Thus, Google uses the entire Internet as a database for generating answers.
The new answer box is located in the upper part of the left column of the SERP, as well as the cards of the first knowledge graph. Google has learned in sufficient detail to answer user questions. Here is what the answer box formed in response to the query "stone for those born in September" looks like:
Please note, the search engine bold direct answer to the question. The answer cards of knowledge graph 2.0 became similar to the block "Definitions on the Internet", which in Russian-language search is formed mainly from Wikipedia data.
The knowledge card answers of the knowledge graph 2.0 actually became the new first position in the issue. They are the first to attract the attention of users, as they are located above organic results. According to MOZ, very often the answer box is formed from data taken from the site located at the first place of issue. In this case, the user immediately sees two links to the donor resource: the first one in the response card, and the second directly in the search results. Here's what it looks like:
However, in some cases, Google uses as a donor resources that are not in the first place of issue. For example, Forbes occupies the first three lines of issue on request "the richest man on the planet." And the data for the answer box is taken from the resource that occupies the eighth place:
It is noteworthy that the site Time.com, whose data entered the answer box, directly refers to Forbes. According to MOZ, Google has preferred Time to Forbes, since the latter used a bad page design and endless scrolling. In addition, semantic relevance played a significant role. On the Time website, Bill Gates's name is located next to the phrase "the richest person on the planet." On the Forbes website, there is no exact entry at all on the phrase "the richest person on the planet."
The value of semantic relevance is confirmed by the following mini-experiment. If you enter the phrase used on the Forbes website as a request, this resource is placed on the response card.
It's only the beginning
Google has yet to improve the knowledge graph 2.0 and the answer box feature. Sometimes answer cards look funny:
However, it is already clear that the search engine will improve the quality of the answers by using the content of the indexed sites. By the way, the search engine has already significantly increased the number of information sources for the formation of an extended knowledge graph panel. Note the related topics block in the following illustration:
Also note the link to the donor site. It is barely noticeable on the background of the definition. A blue highlighted links to get acquainted with the results of the issuance of Google on the relevant request.
Knowledge graph 2.0: more questions than answers
Thanks to Knowledge Graph 2.0, Google can respond to user requests without being limited to online encyclopedias as sources of information. The largest search engine in the world has learned to use to answer the data that you publish on your website. If before Google simply referred to your resource and sent visitors to it, now it displays basic information already on the issuing page.
This situation is certainly beneficial to Google users. But the knowledge graph 2.0 breaks the balance of interests of the search engine and site owners. The search engine actually deprives the resources of part of the traffic. It uses information from third-party resources to keep users from going to the site or to force them to make an additional search.
In this regard, the question arises: will not the actions of Google cause a drop in the quality of content across the Web? After all, Google takes the information for the answer box from the highest quality resources. At the same time, it deprives them of traffic, as users receive the necessary data without leaving the SERP. Google makes site owners think that they get in return by posting information.
What situation do you think is more preferable: when does the information from your website or from your competitor's website get into the answer card? Why?