Google’s Gary Illyes confirmed on stage at the Search Masters Brazil event that while the Google Panda algorithm collects its data at the page level, it will impact the site at the host level. This differs to the Google Penguin algorithm that both collects data at the page level and also impacts the site on a page level, not a host level. Truth is, this is nothing new – I covered this when we were in the thick of Penguin and Panda going real time and such, which feels like ages ago. But Pedro Dias, who interviewed his former colleague, said on Twitter this has been confirmed again. Here is his tweet: So while both these algorithms use data based off of page level signals, only Penguin zooms in when it actually has an impact on the site by impacting a specific page versus the whole site at the host level. How does this help you as an SEO? I am not exactly sure unless you are looking to game Google a bit. But either way, it is an important distinction between Panda and Penguin. Forum discussion at Twitter.