Minggu, 24 November 2013
Q: Does the domain extension actually affect the ranking by Google and other search engines?
The likely answer is NO. +Matt Cutts (homepage) , an outspoken senior engineer at Google, with few search engine adaptations passing by him unnoticed, had this to say
Technically, it is not sensible to a priori discredit certain domains based on their domain extension. So far no recent evidence exists, from any search-engine company to penalize either of .info, .biz, .com, .org, .net, .name domain extensions or weigh them differently.
Why not?
The ranking process for a site to end up at a certain position at the search results is complex, and highly depending on network science. A short introduction into this fascinating field is provided in the Further Reading section. Network science basically attempting to mathematically describe and classify networks. Most important for a site are its internet-wide ingoing links, or indegree in networking lingo.
One thing to keep in mind is that a domain is not the primary score factor. What is, is the content behind it. A domain, once resolved by a DNS, and ideally an accompanying reverse DNS (rDNS) lookup will point to the Internet Protocol (IP) address of a server. Reverse IP lookup is a means to discover all the other domains hosted by a given IP. It is easy to see how the immediate neighbors of a site can affect the standing of a credible site negatively, if all other sites hosted on the same IP are classified as SPAM.
A domain may contain keywords, matching a user's search-terms. By reasoning that firstly, a domain has to be bought on the free market at relatively high prices, and secondly cannot be changed afterwards, inevitably leads to buyer intent. Naturally, matching keywords in domains should be factored in by certain scaling factors of a search algorithm when computing search result scores. Yet there is a but, which Matt Cutts indirectly addresses in the video: a decreased need of relying on old ways.
In times of the social Web or Web 2.0, linkage building has stepped up from being static to being a dynamic process, that is self-organizing and orchestrated by web-users. Users link and vote on the significance of content and indirectly the content's immediate links, which current search-engine algorithms use for their score computation. In the process it makes these algorithms less dependent on the weight of domain keywords than they used to be.
If you mistrust what you just read, simply try to directly find any number of unknown TLDs/Top-Level Domains within the first 100 SERPs or Search Engine Result Pages.
Verdict:
It is not worth considering domain extension in regard to Search Engine Optimization/SEO. Buy cheap!
Pointers:
Lastly, a psychological effect should be acknowledged when considering to buy a new domain: Users will be most familiar with .com,.net, .org domains and will associate .com and .net domains with commerce. To which degree this may affect an editor's, poster's, commenter's decision of linking a given domain is a matter of debate at best. But human perception is undoubtedly influenced by a constant weighting of normal vs. abnormal by thresholding a stimulus against a set-point. This way of human perception affects the overall context of any decision making process. If it is worth considering depends on the context, and if your gut feeling is telling you to lean towards a certain domain-extension then by all means, go for it.
Further Reading:
B.J. Jansen, D. Booth, Classifying Web Queries by Topic and User Intent, CHI 2010 Posters, pg 4285 [link, backup]
L.Sauer, Scalable Web-Application Development in Science, Network and Graph theory, unpublished. Download (pdf), View Online
The likely answer is NO. +Matt Cutts (homepage) , an outspoken senior engineer at Google, with few search engine adaptations passing by him unnoticed, had this to say
1:38 For me, I tend to lean a little more towards things that are brand-able because, for example, if you have 15 sites about Android and they all have Android, Android, Android, Android, it's going to be a little hard to remember to rise above the noise, to rise above the din.
2:31 Some people have complained that we're giving a little too much weight for keywords in domains.
2:36 And so we have been thinking about adjusting that mix a little bit and sort of turning the knob down within the algorithm so that given two different domains, it wouldn't necessarily help you as much to have a domain with a bunch of keywords in it.
Technically, it is not sensible to a priori discredit certain domains based on their domain extension. So far no recent evidence exists, from any search-engine company to penalize either of .info, .biz, .com, .org, .net, .name domain extensions or weigh them differently.
Why not?
The ranking process for a site to end up at a certain position at the search results is complex, and highly depending on network science. A short introduction into this fascinating field is provided in the Further Reading section. Network science basically attempting to mathematically describe and classify networks. Most important for a site are its internet-wide ingoing links, or indegree in networking lingo.
One thing to keep in mind is that a domain is not the primary score factor. What is, is the content behind it. A domain, once resolved by a DNS, and ideally an accompanying reverse DNS (rDNS) lookup will point to the Internet Protocol (IP) address of a server. Reverse IP lookup is a means to discover all the other domains hosted by a given IP. It is easy to see how the immediate neighbors of a site can affect the standing of a credible site negatively, if all other sites hosted on the same IP are classified as SPAM.
A domain may contain keywords, matching a user's search-terms. By reasoning that firstly, a domain has to be bought on the free market at relatively high prices, and secondly cannot be changed afterwards, inevitably leads to buyer intent. Naturally, matching keywords in domains should be factored in by certain scaling factors of a search algorithm when computing search result scores. Yet there is a but, which Matt Cutts indirectly addresses in the video: a decreased need of relying on old ways.
In times of the social Web or Web 2.0, linkage building has stepped up from being static to being a dynamic process, that is self-organizing and orchestrated by web-users. Users link and vote on the significance of content and indirectly the content's immediate links, which current search-engine algorithms use for their score computation. In the process it makes these algorithms less dependent on the weight of domain keywords than they used to be.
If you mistrust what you just read, simply try to directly find any number of unknown TLDs/Top-Level Domains within the first 100 SERPs or Search Engine Result Pages.
Verdict:
It is not worth considering domain extension in regard to Search Engine Optimization/SEO. Buy cheap!
Pointers:
- .info, .biz, .com, .org, .net, .name domains differ significantly in price, whose appreciation or depreciation is largely controlled by the free market. See the Internet Domain Name Index.
- General Top Level Domains /gTLD carry with them buyer/user intent, which will slightly affect the overall standing within the SERPs, but the choice of gLTD is insignificant compared to the weight of even one strong/credible ingoing link.
- Contrarily, Sponsored Top Level Domains, which are generally not accessible to everyone do significantly impact a a site's search engine score.
- Country Code Top Level Domains such as .us, .ca, .uk, .in, .ir will will significantly affect the ranking of a site among the national search results, unless the user searches internationally. Of much greater significance is however the detected language and character-set of the site's content, to which the domain points to.
Lastly, a psychological effect should be acknowledged when considering to buy a new domain: Users will be most familiar with .com,.net, .org domains and will associate .com and .net domains with commerce. To which degree this may affect an editor's, poster's, commenter's decision of linking a given domain is a matter of debate at best. But human perception is undoubtedly influenced by a constant weighting of normal vs. abnormal by thresholding a stimulus against a set-point. This way of human perception affects the overall context of any decision making process. If it is worth considering depends on the context, and if your gut feeling is telling you to lean towards a certain domain-extension then by all means, go for it.
Further Reading:
B.J. Jansen, D. Booth, Classifying Web Queries by Topic and User Intent, CHI 2010 Posters, pg 4285 [link, backup]
L.Sauer, Scalable Web-Application Development in Science, Network and Graph theory, unpublished. Download (pdf), View Online