If a collection item is found that has the same identifying field as an object previously scraped (e.g. the same ID), then the new object will replace the old object. This makes it easy to continually scrape a source and not have to worry about deduplicating the results when there is overlap.