The CRISP-DM (CRoss Industry Standard Process for Data Mining) project proposes a standard process model for data mining. This model is independent of industry and technology, and is designed to provide structure and flexibility for data mining projects. The paper argues for the need of a standard process model in data mining and reports on practical experiences with the CRISP-DM methodology. The CRISP-DM methodology is based on previous attempts to define knowledge discovery methodologies. It is described as a hierarchical process model with four levels of abstraction: phases, generic tasks, specialized tasks, and process instances. The generic CRISP-DM reference model provides an overview of the life cycle of a data mining project, consisting of six phases: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. The paper discusses the application of the CRISP-DM methodology in a response modeling project. The goal was to develop a standardized process that could be reliably performed by marketing professionals with limited data mining skills. The project team included both experienced data miners and marketing specialists. The initial case studies focused on acquisition campaigns, where the challenge was to select prospects likely to buy a Mercedes for the first time. The paper highlights the importance of a standard process model for data mining, as it helps to ensure consistency, improve communication, and facilitate the reuse of knowledge and experience. It also discusses the challenges of applying the model in practice, including the need for flexibility and the importance of quality assurance. The paper concludes that CRISP-DM works well for data mining projects, and that the generic process model is useful for planning, documentation, and communication. It is also important to adapt the process model to specific applications and to ensure that all documents are living and flexible.The CRISP-DM (CRoss Industry Standard Process for Data Mining) project proposes a standard process model for data mining. This model is independent of industry and technology, and is designed to provide structure and flexibility for data mining projects. The paper argues for the need of a standard process model in data mining and reports on practical experiences with the CRISP-DM methodology. The CRISP-DM methodology is based on previous attempts to define knowledge discovery methodologies. It is described as a hierarchical process model with four levels of abstraction: phases, generic tasks, specialized tasks, and process instances. The generic CRISP-DM reference model provides an overview of the life cycle of a data mining project, consisting of six phases: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. The paper discusses the application of the CRISP-DM methodology in a response modeling project. The goal was to develop a standardized process that could be reliably performed by marketing professionals with limited data mining skills. The project team included both experienced data miners and marketing specialists. The initial case studies focused on acquisition campaigns, where the challenge was to select prospects likely to buy a Mercedes for the first time. The paper highlights the importance of a standard process model for data mining, as it helps to ensure consistency, improve communication, and facilitate the reuse of knowledge and experience. It also discusses the challenges of applying the model in practice, including the need for flexibility and the importance of quality assurance. The paper concludes that CRISP-DM works well for data mining projects, and that the generic process model is useful for planning, documentation, and communication. It is also important to adapt the process model to specific applications and to ensure that all documents are living and flexible.