Using a special query to the Google database, it is possible to extract certain metadata for an indexed page. When querying Google for the URL http://www.apple.com/macbookpro/, we would get something like this:
URL_1:32:http://www.apple.com/macbookpro/ Title_1:19:Apple - MacBook Pro Rank_1:1:9 CrawlDate_1:10:1 feb 2006 Summary_1:169:You’ve dreamed [...] of the PowerBook G4. ... Link_1:0: CacheSize_1:3:21k Related_1:0:
The line that says Rank_1:1:9 is the important one, and
it is believed to represent the current PageRank for this URL, ie
(9/10).
Most other PageRank prediction tools use this method of extracting the current PageRank value. What this tool does different though is that it makes sure it gets the current PageRank value for the exact URL entered. Where other future PageRank tools get it right most of the time, we make sure we get it right everytime.