Background: For patients with cancer, being well informed by their oncologist about treatment options and the implications thereof is highly relevant. Communication skills training (CST) programs have shown to be effective in improving clinicians' communication skills, yet CSTs are time-consuming, inconvenient to schedule, and costly. Online education enables new ways of accessible learning in a safe and personalised environment. Aim and methods: We describe the design of a digital CST-tool for information provision skills that meets oncologists' learning needs. We used the CeHRes Roadmap for user-centred design as a guiding framework. Phase 1 (Contextual Inquiry) involved consultation of the literature and a focus group interview study to uncover the learning needs and training preferences of clinicians' regarding a digital training for the skill of informationprovision. In phase 2 (Value Specification), two multidisciplinary expert panels specified the learning content and format of a digital training. Phase 3 (Design) encompassed an iterative development process, including two user group assessment sessions and 5 individual user sessions in which prototypes were tested. All sessions were recorded and independently analyzed by two researchers. Results: Based on literature and consultation of the users in the inquiry phase of the development process, and on expert opinion in the value specification phase, relevant (sub) skills and user requirements were defined to consider for the digital training format. It was decided to develop a conventional e-learning and a chatbot. Personalization and interactivity were integrated in the prototypes by including features that allow for e.g., choosing text, video or animation; to upload video-recorded consultations to receive peer-feedback; and to consult a communication expert. Results revealed that, overall, participants expressed a willingness to use a digital training tool to acquire information-provision skills. Individual user testing (including junior clinicians), indicated a preference for the chatbot over the e-learning.