To know the headline location in a web page/blog automatically needed an extraction method because each website/blog has own unique characteristic in advertisement, headline, and link list placement. This research developed a system to mapping tag pairs, and single tag of HTML on website/blog code, the result can be used as a pattern or comparison against already saved patterns in the database. Extracting and mapping method use both vertical and horizontal multilevel numbering. Data example has taken from 20 websites of government, school, and also private companies to obtain the result of tag mapping that is going to be used to know the amount of pattern that produced by a website. Template comparisons are up to 5th level, the result of extracted website/blog contains hundreds of pages starting from the lowest that is 110 and the highest 280 and it is only less than ten patterns for each website. Conclusion of this research is that all extracted website has pattern less than 10% of the total pages.