How do search engines work and what are the particularities of each one of them?

Although there are a huge number of search engines, they all have the same process for answering queries. This three-step-process begins with the crawling, follows the indexing, and finally comes the ranking.


Crawling

Before the search: web crawlers – which also can be referred to as ‘Spiders’ software – will fetch webpages, then follow the links on those pages and then fetch the pages they point to etc. In the end, this will lead to the establishment of an index made up with billions of pages which will now be connected together.


Indexing

After the web crawlers have gathered webpages, they will order them in their index, which is a database - similar to a digital gigantic library - where the search engine will collect the webpages corresponding to one’s query.


Ranking

The first step made by the algorithm during the search is that it will search for every page that includes the words corresponding to the query. Then the algorithm will sort the pages by solving several questions such as ‘How many times does the requested words appear on the pages?’ or ‘Are there synonyms for those words on the pages?’ and ‘Is the page from a quality website?’ etc.

Let us take the example of the Googlebot - Google's algorithm - which has several ways to search, analyse and classify the results:

  • In order to find the most relevant answers possible, the first step consists in analysing the meaning of the words in the query.
  • Google will also look for the webpages that match the best for the query in the index, by counting how many times the keywords appear on the different parts of a webpage.
  • On the index, every webpage has its own ranking – also calculated by an algorithm which will surface the most relevant pages with the most appropriate information.
  • The context is also taken in count when analysing pages throughout the index. Criteria such as the user’s location, whether he/she uses the ‘Safesearch’ or not etc will also make a difference in the final result the algorithm will present.
  • And finally, in order to present the best results, the algorithm will fix issues such as the different topics that are broached and then count how many of them there are. The idea is to give a wide range of results, which stays relevant to the search of origin, while avoiding a narrow interpretation of these ones.


How to improve a site’s ranking ?

  • Offering fresh contents enables a website to attract more often the Googlebot/Bingbot etc. and therefore, it is more likely to have its pages among the search engines’ index.
  • Backlinks also have a major part in the ranking process: they correspond to the links on an outside website to your own website (in a quantitative but also qualitative way). It is a good way to measure a site’s approval rating.
  • Relevant and useful information
  • A consistent user-friendly experience
  • Keyword usage ( in the page title and headers, ensuring they’re visible throughout the page)
  • Technical website optimization such as: optimizing the meta descriptions, but also the site’s structure, its speed etc.
  • Try to make the time that the user will spend on the site longer
  • Include URLs in your social media posts that sends back to your website to build your ranking


Differences between Google and Bing

  • As for the SEO strategy and the different search engines, even if each search engine has its own algorithm and way to process for ordering results, their aim is to provide the best answer for a user’s query. Therefore, it is not necessary to have completely different SEO strategies, even if there can be some benefit in studying the nuances between both engines. For instance, we know that Google’s search algorithm is the best at matching results with synonyms and relevant words, whereas Bing requires more to enter the right keywords to get the relevant search results.
  • As for page ranking, Bing will show first information related to recent events related to your search, while Google will give you information more related to the topic itself, giving less consideration to what happened lately.
  • If a website includes Flash content, it is more likely to be ranking higher in Bing than in Google.
  • Bing also tends to give more importance to the lifespan of a website than Google. It also crawls only through the first 100 kb of a webpage, whereas Google crawls through the whole page.

When studying the algorithms and page results of both search engines, we only notice a slight difference between Google’s results and Bing’s. These differences can be explained by the algorithms themselves, which – even if they look alike – were made by different companies.


Yandex

This search engine was created even before Google. It is actually Russia’s largest search engine. Unlike Google, Yandex takes into account the distance between words along with the relevance of documents to a user’s query, but except from this, the principle of crawling, indexing and ranking stays the same.

Let us see the main differences between Yandex and a western search engine such as Google.

  • In addition with being a search engine, Yandex is also a portal (like Yahoo) where many Russians begin their day.
  • Yandex is best to use for Russian language search. The Russian language has many particularities, for instance, each noun has a grammatical gender which depends on the construction of the sentence, and this gender will influence the rest of the sentence and so the whole query. As we know, Google only search for ranks pages that are accurate to the user’s query, whereas Yandex is able to parse the synonyms and user intent regardless of the spelling.
  • As for Yandex’s algorithm, it does not take into account links in general. That is to say, the algorithm will only use the user’s experience and keyword ranking metrics. This may give better credibility to Yandex and enables it to generate better quality results, as the risk to give credit to low quality links is negligible.
  • Yandex also has a high priority on geo-targeting. Indeed, all queries are divided between geo-dependent and geo-independent searches. As a consequence, people from different locations will not be confronted to the same results even though they had the same query.
Continue reading?

Share this article

asds