The digital transformation of industries and companies is an integral element of competitiveness at the present stage. However, companies are experiencing problems in systematizing and finding effective marketing digital approaches, including in the field of pricing. The purpose of this chapter is to develop proposals for improving approaches for companies in the residential real estate market. Interest in the study is due to the lack of effective pricing methods that would meet the requirements of modern companies in this area. The methodological basis of the work is the study of the theory and practice of applying innovative technologies in the field of deep learning based on the use of neural networks. The analytical and desk research of companies in the residential real estate sector, the results of which are presented in the article, was conducted based on general scientific principles of comprehensive study of economic phenomena, methods of systematic, logical analysis, generalization, as well as methods of mathematical statistics and machine learning algorithms. The collected data were processed using Excel, R software tools. For data analysis, studies were conducted using ANN, XGBoost, Ridge, Lasso, linear and multiple regression methods. The results of the research can be summarized in two main blocks. Firstly, analysis of existing pricing methods in the real estate market. It includes parametric and non-parametric pricing models as well as modern machine learning methods, such as random forest and gradient boosting. Secondly, a multifactorial pricing model in the residential real estate market was developed, based on machine learning methods and deep neural networks. The model takes into account both internal characteristics of real estate objects and external macroeconomic and socio-demographic factors influencing demand and prices. The developed algorithm allows more accurate forecasting of residential real estate prices, considering demand dynamics and elasticity in various market segments. Key parameters influencing pricing were identified, and their preliminary selection was carried out using correlation analysis and feature selection methods. This reduced the model's complexity and increased its accuracy.