|
|
|
Next: Recycling Domains
QueriesA knowledge of how search engines work and they are actually used is very useful to search engine optimizers. Google heralded the era of lazy searching, or search for the masses, depending on your point of view. Before Google using the top dog, AltaVista, needed an arcane knowledge of boolean logic and obscure syntax to get the most out of the engine. With Google, and increasingly with rivals such as Yahoo! and MSN's so called "algorithmic search" you simply have to type your query and normally you find what you are looking for in the first few results. The relevancy is so good that the majority of searchers don't look beyond the first results page. A survey by OneStat that examines search engine use has shown that users are becoming more sophisticated with an increase in the use of two and three search terms in queries. The implication for page optimization is that specialist pages targeting niche terms are the immediate future. Some searchers are surprised to type in a query:
and find pages where the keywords do not actually appear on the page. This is because of off-page optimizations such as anchor text user by the search engine algorithm. On Google you can use the intext operators to restrict the search to page contents. In terms of boolean logic the search includes all the terms, in any order and is the equivalent of
If you want to find any of the words you can use OR instead of AND. Some search engines will exclude common words. These can be included by putting a '+' operator in front of the keyword:
Conversely the '-' operator can be used to exclude keywords An exact phrase can be matched by putting the words in quotes: "to be or not to be" Stop words will still be excluded from the search. You can mix and match "to be or not to be" Jack Benny the asterix '*' character can be used as a wildcard, it will match any word:
Matching exact terms can be useful for search engine optimizers who want to see if a new page that has recently been spidered by the search engine robot can be found in the index. Initially these pages may rank too low to be found using target keywords but may be found by entering a phrase from the page content. General search queries often return far too many results. With a pool of some billions of pages to choose from it doesn't really matter if the results are comprehensive so long as some relevant links are included on the first page. To cope with information overload when searching for specific information go detailed first then zoom out and favour obscure keywords in the search. There are some useful operators that can be used to analyse site rankings in search engine results pages. The "site:" operator restricts the search to a specific web site:
If you just enter the site restriction with no keywords you get an idea of how many pages from your website are indexed by that search engine:
Clearly there is some problem with seochat.com at Yahoo! You can restrict your search to top level domains:
This search will only find pages from educational establishments. Or you could search just in a particular country:
When I tried this the first page returned was from the French President's web site! The link operator shows the number of inbound-links (backlinks) for a site:
The figure for the new MSN search is maybe not surprising but why is the value for Google so low compared to Yahoo!. Google doesn't show low PageRanked inbound-links. This makes some sense for them as their PageRank algorithm will place much less importance on such links. In addition the Google toolbar often doesn't give the exact ranking for sites making analysis of Google's algorithms more difficult. The special operators intitle and inurl work with both Google and Yahoo search and restrict the keyword matching to the web address and page title. This can be useful for understanding which elements of a page are important to the search engine results and and can be used to analyse competition. Search engines are constantly adding new features. For example, Yahoo and Google let you specify file formats. Sometimes new features have unintended consequences. When Google introduced number range searching the first thing many people thought of was looking for credit card numbers that had inadvertently been indexed by search engines:
Google supports date ranges as well. This shows when the page last changed in the Google index. The Google web interface only gives a limited range of options but you can use the daterange operator directly:
finds all the documents updated between the 23rd and 24th of November, well over a million of them. The date format uses the Julian calendar which is a count of days since January 1, 4713 BC. The Julian calendar doesn't require complicated leap century calculations or those involving missing days involved with the changeover from Julian to the Greogorian calendar. The search engine's own help pages are often wrong or simply out of date. As an example Google claims that the site: operator can be used on news stories but this is not supported.
See AlsoFagan Finder has an alternative interface to the major search engines giving all the available search options |
|
©1994-2006 All text and images copyright: www.abcseo.com; last updated: |
