Disclosed is a method and computing device that receives a web page and locates text elements in the web page, where each text element includes a set of one or more characters or symbols. For each text element found, the computing device assigns a weight value to the each text element. The computing device then stores the text from the each text element in a relevant text storage if the weight value for the each text element is above a threshold weight or similar in layout or content to elements above the threshold weight.