Website archival takes place by scheduling a "bot" (short for "robot") to view all of the content on a website. When a bot looks at the homepage of a website, it takes note of all of the navigational links on the page and creates a map of all of the pages on a website and then visits those pages as well. This process is called "crawling" and a it creates a comprehensive list of pages on a website. This list is then crawled again on a scheduled basis to look for new pages or for changes on those pages.
This process takes place several times per week. When a change is found on a page, a screenshot is taken and added as a version of that page in the archive history for that site. This allows us to create a historical archive for how every page on a website changes over time.
Not every page on a website can be found by clicking through it's links. An example could be a marketing landing page that is only used for sharing to social media. Other links might only exist as a link on someone's "About" section of their Facebook Page. When our system sees these links, they are added to the crawling schedule and archived with the rest of the website.