Entering the 21st century, the Internet industry is developing rapidly, and all walks of life are facing huge challenges. How to attract more customers under the huge impact of the Internet is a problem that enterprises urgently need to solve. This article starts with precision marketing and introduces banking products. With the deepening of internet big data technology, the bank has its mobile client and studied how the banking industry uses various means to collect data in the big data environment, establish its user profile, and adjust its marketing strategy. When the bank recommends to users, it must choose an appropriate algorithm to serve its marketing. Collaborative filtering algorithm is the most classic and easyto-operate recommendation algorithm. This article provides an improved collaborative filtering algorithm, which is a method for calculating the similarity of recent interests related to time. Through this improved collaborative filtering algorithm, it can provide a new reference for the precise marketing of banks.