PageRank, described in a 1998 Stanford technical report by Lawrence Page, Sergey Brin, Rajeev Motwani, and Terry Winograd, is the link-analysis algorithm that gave the original Google search engine its name and its ranking power. The companion paper, “The Anatomy of a Large-Scale Hypertextual Web Search Engine,” presented the full Google prototype; the PageRank report focused on the ranking idea itself.
The insight was to treat a hyperlink from one page to another as a vote of importance, but to weight votes by the importance of the page casting them, so a link from a widely respected page counts for more than a link from an obscure one. The report frames this as modeling a “random surfer” who keeps clicking links at random and occasionally jumps to a random page; the PageRank of a page is, roughly, the long-run probability that this surfer lands on it. This made it possible to rank pages objectively using only the link structure of the web, rather than the contents of the page, which earlier search engines had relied on and which was easy to spam.
PageRank is not a recommender system in the collaborative-filtering sense, but it is the same kind of object: an algorithm that ranks a vast set of items by importance using the recorded behavior of a network. The eigenvector computation at its heart is closely related to the matrix methods that later recommender and graph-ranking systems would use.
Why business readers should care: PageRank reorganized how the world finds information and became the foundation of one of the most valuable companies ever built. It is the canonical example of how a single ranking algorithm can become an economic moat.