Recent advances in the field of intrusion detection brought new requirements to intrusion prevention and response. Traditionally, the response to the detected attack was selected and deployed manually, in the recent years the focus has shifted towards developing automated and semi-automated methodologies for responding to intrusions. In this context, the cost-sensitive intrusion response models have gained the most interest mainly due to their emphasis on the balance between potential damage incurred by the intrusion and cost of the response. However, one of the challenges in applying this approach is defining consistent and adaptable measurement of these cost factors on the basis of requirements and policy of the system being protected against intrusions. In this paper we present a structured methodology for evaluating cost of responses based on three factors: the response operational cost associated with the daily maintenance of the response, the response goodness that measures the applicability of the selected response for a detected intrusion and the response impact on the system that refers to the possible response effect on the system functionality. The proposed approach provides consistent basis for response evaluation across different systems while incorporating security policy and properties of specific system environment. We demonstrate the advantages of the proposed cost model and evaluate it on the example of three systems.