Workplace-based assessment (WBPA) was introduced in 2007 as a new approach to monitoring competence of GP specialist trainees (GPSTs). It includes a raft of assessments carried out in the workplace to assess what a trainee actually does in clinical practice. The assessment tools used are adapted from other contexts of doctors' training but little is known about how they function in day-to-day practice within GP training or how valid and useful they are found to be by trainees. To establish how the new system of WPBA is working in day-to-day practice for GPSTs in hospital posts. A mixed methods design including quantitative and qualitative phases of data collection. Two training locations within Severn Deanery. A questionnaire was completed by 52 GPSTs (67% response rate) currently in hospital posts. Twenty-two took part in focus groups and semi-structured interviews to explore key findings from the questionnaire in greater depth. There is value in the face-to-face contact between trainees and senior doctors. However, quality and depth of feedback are not consistent and there is evidence of poor use of the tools, reducing the value of the assessments. The system is further undermined by a clear perception of bias and lack of honesty in judgements which limit the scope for assessment to lead to learning. Overall, these weaknesses may impair the validity and usefulness of the system and its potential to improve the performance of doctors. General practice trainees in this study have a low opinion of how WPBA assessments function in the hospital setting. Changes are needed to optimise the potential of WPBA to improve the performance of doctors in training and to increase its credibility.