Development of Omnichannel Product for Modern Retail Ecosystem (Online and Offline Retail, Marketplace)

This paper examines the use of customer clustering algorithms to solve the problem of building an omnichannel communication strategy for a modern retailer with online, offline and marketplace sales channels. The topic of this paper is motivated by the real business problem of the Magnit retail company. The paper studies the experience of using an omnichannel approach to communications in retail and proposes a solution to this problem using clustering algorithms. Next, a review of algorithms and the practice of their use for solving such practical problems is made. A practical experiment is carried out using data on customer purchases in different sales channels. At the end of the work, the results are reviewed, proposals are made for using the results obtained to form a communication strategy, and the vector for further research is determined.