Good search engines will rank the results according to how relevant they are to the user's query, with the most relevant first. Each search engine uses its own ranking algorithm. Original Google PageRank™ algorithm was effective because it used high reputation sites linking to and from the page as a marker that the page contained high quality content - it was a form of 'wisdom of the crowds' and it was the basis of Google's success.This is the original algorithm as published by Lawrence Page and Sergey Brin PR(A) = (1-d) + d(PR(T1)/C(T1) + .... + PR(Tn)/C(Tn)
- PR(A) is the PageRank of page A
- PR(Tn) is the PageRank of pages Tn that link to page A
- C(Tn) is the PageRank of outbound links from page A
- d is a damping factor to keep it stable values 0 to 1
Current PageRank™ algorithm is more sophisticated and has over 200 factors in its calculations. Most of these are no longer public, but some are known:
- Keywords found within the content. Must be relevant to the main content. A site that is trying to 'game' the engine by inserting incorrect or unrelated keywords will be downgraded in ranking.
- Last modified
Algorithm may consider how old the page is and when was it last modified. A page that is years old is less relevant than a recent one with similar content.
- External links and link reputation
If a large number of high-reputation (high PageRank) sites link to the page or resource then the algorithm takes this into account because it assumes these sites would only link to high quality material. It considers the PageRank of the outgoing links on the page - because why would a page want to link to low quality material.
- SEO companies
Getting a high ranking is important for businesses and so an industry has grown up called 'Search Engine Optimisation' where consulting companies offer their expertise to improve their customer's ranking.