Author: Zen and the Art of Computer Programming
1 Introduction
In 1997, the first website advertising system appeared in the Google online search engine. At that time, in order to increase revenue, Google proposed the term "network effect", which is the ability to convert users' clicks and purchasing behavior on page information into advertising costs. This technology first focuses on search rankings and continuously improves the quality of query results by optimizing keywords and advertising density. But with the development of the Internet and the rise of e-commerce, this system has gradually been proven to be a huge "bubble" - as the number of users increases, its advertising effects inevitably become worse and worse. On the other hand, people's focus on search results has also changed - people expect to browse the most valuable content rather than simple page rankings, which also provides advertisers with new marketing strategy directions.
At the end of 2006, Facebook launched its own advertising system, whose main goal is to provide personalized advertising services to individual users. Facebook founder Mark Zuckerberg once described the four major characteristics of this system:
- Provide personalized advertising based on user data: Facebook combines people's habits, preferences, interests, preferences and other data to accurately deliver ads;
- Constantly iterating advertising algorithms: Facebook updates its advertising algorithm every few months so that it can more accurately match user needs;
- Large-scale data accumulation: Facebook generates a large amount of data every day, which helps it quickly analyze users’ interests, habits, hobbies, preferences, etc., so as to accurately target advertisements;
- Highly scalable advertising platform: Facebook can easily scale horizontally to create a massive advertising network around the world.
At the end of 2013, Baidu also launched its own advertising system. Its functions are similar to Facebook's personalized advertising services, but the usage scenarios are different. Baidu's advertising system consists of two modules, namely search and Tieba. The former focuses on providing personalized push for the relevance of search results and topic construction, while the latter focuses on providing relevant information for specific industries or topics.