What does a search engine refer to

Search engines explained: definition, alternatives, market shares, functionality

You are interested in the topicsearch engines. Then you have come to the right place in our glossary for technical terms from (online) marketing. In this post you get all the important information on the topicsearch engines.

Table of Contents

What are search engines?

ASearch engine is an application for researching information stored on the Internet or in a closed database. After entering a search query, as text or via voice, a search engine provides a list of search results that refer to relevant documents. The results in the so-called SERPs are usually presented with a title and a short extract from the respective document.

The essential components or areas of responsibility of a search engine are:

  • Creation and maintenance of an index (data structure with information about documents)
  • Processing search queries
  • Find and sort results
  • Preparation of the results in a user-oriented manner.

Usually theData acquisition automatically, on the Internet by web crawlers, on a single computer by regularly reading all files in directories specified by the user in the local file system.

Types of search engines

Basically, you can differentiate between search engines with their own index and so-called proxy search engines that use a third-party index.

Examples of search engines with a self-crawled own index:

  • Google
  • Baidu
  • DuckDuckgo
  • Qwant
  • Yandex
  • Bing

Examples of proxy search engines:

  • Ecosia (Bing Index)
  • Lycos (Bing Index)
  • T-online search (Google index)
  • AOL Search (Google Index)
  • Yahoo (Bing Index)

There are also special search engines that are limited to special use cases:

  • Youtube (video search engine)
  • Amazon (product search engine)
  • Google image search
  • Google video search

Overview: The largest search engines by market share

Below is a list of the most popular search engines and their market shares in Germany and worldwide. In countries such as Russia, China, but also the USA, there are sometimes significant regional differences.

 

 

Search engines in Germany

Source: StatCounter Global Stats - Search Engine Market Share

Search engines in Russia

Source: StatCounter Global Stats - Search Engine Market Share

Search engines in china

Source: StatCounter Global Stats - Search Engine Market Share

 

How do search engines like Google, Bing & Co. work?

Search engines with their own index crawl freely available content, e.g. via links or manual requests for indexing. The crawler stores links it finds in a plan list (scheduler), which is processed later to find new or updated content. The search engine bots can be prevented from crawling certain content via the robots-txt file. Indexable contents are saved in raw form in html format and cleaned up. In a second step, each document goes through a rendering for the interpretation of the markup instructions such as CSS and html, but also for the representation of Java Script. It should be possible to display and understand the document as a user sees it in his browser.

 

Crawling and indexing, © Olaf Kopp

 

Modern semantic search engines such as Google extract information on entities from the indexed documents in order to relate them to each other and to enrich them with further information / attributes.

Most search engines and their ranking algorithms are relatively simple and one-dimensional. Keywords that appear in the search term are compared with the content available in the index. Simple document search query matching based on the keywords used is common to all search engines. All search engines have in common that they work on the basis of text analysis. A query processor is responsible for interpreting the search query and a scoring engine for evaluating the documents with regard to the search query. Individual layers such as a cleaning engine or a personalization layer are applied by the major search engines as a refinement to the search results created by the scoring engine in order to avoid duplicates, apply possible penalties or output individualized search results based on historical user behavior.

How simple search engines work, © Olaf Kopp

 

Over the years, however, Google and Bing have developed a significantly more complex, multi-dimensional system to determine rankings for content or documents and to output them sorted as search results.

The difference between Google and Bing is the influence of so-called off-page factors such as links on the rankings and the ability to semantically interpret search queries and content.

In this way, these search engines are able to determine the relevance of content based on text analysis methods and external links and put them in order accordingly. But Google not only wants to determine the objective relevance of documents, but also the pertinence and usefulness in relation to the individual user. In addition this video:

For this purpose, Google uses Rankbrain or BERT, an algorithm based on machine learning, which is supposed to simulate a kind of semantic understanding of search queries. In this way, Google is also able to determine the real search intention behind a search term.

Furthermore, the individual user context plays a major role in personalizing the rankings. For search terms with a regional search intention, the location of the user and the type of device used are an important ranking factor. When logged in, Google is able to output personalized rankings based on search history or click behavior. Here is a graphic of how Google works today in terms of retrieval, scoring and ranking:

Overview: Possible information retrieval on Google, © Olaf Kopp

The following is one of the most recent presentations on SEO and how Google search works:

Alternatives to Google

Further sources and information on search engines


Blog article on the subject

Quo Vadis SEO? What role will (the) SEO play in the future

The SEO world has changed a lot in the last 10 years and with it the position and role of SEO in companies and in the (online) marketing industry. In particular, Google introductions such as Panda, Penguin, Hummingbird, Machine Learning, E-A-T ... have the cards for the Bran ... View article

SEO Costs: How Much Does Search Engine Optimization (SEO) Cost?

How Much Does Search Engine Optimization (SEO) Cost? What are the SEO costs? For most of them, this question cannot be answered across the board at first. The simpler question would be how much does SEO not cost? to answer. We try to show articles with the help of selected SEO professionals such as Jens Fauldrat

My SEO analysis of the Google December Core Update

In this post I analyze the impact of Google's December 2020 Core Update on this domain. This content is interesting for SEOs and those responsible for marketing. Basics on the subject of Google Core Updates I usually don't deal very intensively with the now ... Show articles

Chinese Search Food - What does EAT mean to Baidu?

Guest contribution by Marcus Pentzek - This contribution tries to deduce that expertise, authority and trust (E.A.T.) are not really new aspects and come into play not only with Google but also with the Chinese search engine Baidu. Concrete hints are worked out how ... Show article

Digital Voice Assistants - The Marketing of Intentions

Since digital voice assistants have found their way into the perception and purse of the masses, not only has the development of the systems themselves been driven with high pressure, but brands and companies are also looking for their entry point into the market and the use of "Smart Voice". In the following ... Show article