Phrase Based Indexing

Phrase Based Indexing is a system developed by Google that takes into account the other components of search queries and not just a single keyword.

Definition

Google developed and patented Phrase Based Indexing in 2004 and continues to innovate and develop it. In the meantime, it is considered a standard system with which search engines crawl, capture, sort and assign web documents and content.

Phrase Based Indexing is a further development of Single Word Indexing, where content is defined and searched for by search engines based on a specific keyword.

In addition to the actual keyword, the system also takes into account the semantic environment, i.e. the entire sentence structure. From a query typed into the search box, Phrase Based Indexing matches several important words (omitting any stop words) to existing web content. If the keywords also appear in the document in a close, semantic context, the website is visible in the SERPs. The more congruent the search phrase and content (including any stop words), the higher the ranking of the page in the search results of the query.

Phrase Based Indexing also checks how different phrases are used on the internet and decides if they are acceptable.

Phrase Based Indexing and SEO

Phrase Based Indexing has the greatest impact in ranking web pages by relevance to the search query, as well as link analysis.

Phrase based content relevance

For the relevance of a web page, even the use of a keyword could lose importance in the long run with Phrase Based Indexing. For example, if a content piece uses phrases that, according to the database of allowed phrases, appear mostly in other web content with the same context.

Phrase based link analysis

Natural anchor texts might become more important than SEO-optimized link texts in the long run. Background: If the anchor text for an outbound link consists of a phrase that is not used on the target page, but according to the lists for permissible phrases is closely related to phrases as they are present on the target page, this target page is evaluated as more relevant for the phrase from the anchor text used on the referring website.

Problems and dangers with Phrase Based Indexing

With the indexing of phrases that are related in content, the semantic understanding of the search engine algorithms and thus the precision of output SERPs can be improved. Google’s auto-complete feature while typing search queries can also see steady improvement through Phrase Based Indexing.

However, Phrase Based Indexing requires servers with much higher storage capacity than Single Word Indexing (also: Term Based Indexing). And tokenizing search queries can be challenging for AI. These are common arguments for the assumption that Google, Yahoo as well as other search engines do not use Phrase Based Indexing.

Although Google has never officially confirmed it, there is still evidence that this system is used in indexing and ranking taxonomy. Numerous patents on Phrase Based Indexing strongly suggest this, at least. The fact that the corporation has apparently been testing it as part of its TeraGoogle project also points to this. Especially since the initiator Anna Patterson, on whose programming Phrase Based Indexing is supposed to be based, temporarily left Google to establish the competitor search engine Cuil with her patent. Yahoo has also registered similar patents via so-called superunits.

Just because Google files a patent doesn’t mean that’s exactly what the corporation will implement:

Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren

Video laden

YouTube immer entsperren

Conclusion

Phrase Based Indexing allows search engines to start building keyword tables for documents based on whole sentences/phrases. Contrary to a broad match keyword level. If these phrases and their relationship to each other are understood, you can use this information to improve relevancy scores, link analysis, and spam detection.